Download (6Mb) - Warwick WRAP - University of Warwick

Loading...
University of Warwick institutional repository: http://go.warwick.ac.uk/wrap A Thesis Submitted for the Degree of PhD at the University of Warwick http://go.warwick.ac.uk/wrap/57704

This thesis is made available online and is protected by original copyright. Please scroll down to view the document itself. Please refer to the repository record for this item for information to help you to cite it. Our policy information is available from the repository home page.

Assessment of demersal fishery resources in Brunei Darussalam Syazana Ebil BSc., MSc.

Thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy in Biological Sciences School of Life Sciences (Environmental Resource Management) University of Warwick March 2013

Contents Contents ....................................................................................................................... ii List of Figures ............................................................................................................ vii List of Tables................................................................................................................ x Acknowledgements .................................................................................................... xii Declaration ................................................................................................................ xiii Summary ................................................................................................................... xiv List of Abbreviations.................................................................................................. xv CHAPTER 1

General Introduction ....................................................................................... 1 1.1 Background ............................................................................................................ 2 1.2 Brunei Darussalam, coastal features, and climate and environmental status ......... 4 1.3 State of marine capture fisheries in Brunei Darussalam ........................................ 8 1.3.1 Fisheries management and research ................................................................ 9 1.3.2 The fishing fleets ........................................................................................... 11 1.3.2.1 The large-scale sector............................................................................. 12 1.3.2.2 The small-scale sector ............................................................................ 13 1.3.3 Marine fishery resources of Brunei ............................................................... 15 1.4 Research aims and objectives............................................................................... 17 1.5 Thesis organisation: Summary outline of chapters .............................................. 18 CHAPTER 2

Review on Assessment of Tropical Fishery Resources ............................... 19 2.1 Overview of tropical fish stock assessment ......................................................... 20 2.2 Data requirements for development of indicators ................................................ 26 2.2.1 Assessing stock status using catch-effort data .............................................. 28 2.2.2 The ‘Catch per Unit Effort’ model ................................................................ 30 2.2.3 The ecosystem-based approach ..................................................................... 33 2.3 Contribution of local ecological knowledge (LEK) in stock assessment and fishery management ................................................................................................... 37 2.3.1 The “Shifting Baseline Syndrome” ............................................................... 39 2.3.2 Formalisation of LEK with geographical information system (GIS)............ 41 2.3.3 Application of fuzzy logic concept in quantification of LEK ....................... 42 2.4 Marine environmental variability and fishery resources ...................................... 43 ii

2.4.1 Impact of climate change on tropical marine ecosystem and fishery ........... 45 2.5 Earlier studies on fishery and stock assessments in Brunei Darussalam ............. 47 CHAPTER 3

Overview of Brunei marine ecosystem in the past decade via trophodynamic indicators and community structure analyses ................. 50 3.1 Introduction .......................................................................................................... 51 3.2 Materials and methods ......................................................................................... 52 3.2.1 Study area ...................................................................................................... 52 3.2.2 Catch data and trophic level estimates .......................................................... 52 3.2.3 Data analyses ................................................................................................. 56 3.2.3.1 Mean trophic level (MTL) and Fishing-in-Balance (FiB) index ........... 56 3.2.3.2 Trophic spectra (TS) .............................................................................. 57 3.2.3.3 Statistical analyses ................................................................................. 58 3.3 Results .................................................................................................................. 59 3.3.1 Exploratory analysis: Trends in annual catch and effort ............................... 59 3.3.2 Trends of trophic-based indicators ................................................................ 62 3.3.3 Trends in species composition of catch ........................................................ 63 3.3.4 Catch trophic spectra analysis ....................................................................... 66 3.4 Discussion and Conclusion .................................................................................. 68 CHAPTER 4

Exploring human and environmental influence on recent abundance trends of selected demersal stocks in Brunei ............................................... 74 4.1 Introduction .......................................................................................................... 75 4.2 Materials and Methods ......................................................................................... 77 4.2.1 Study area ...................................................................................................... 77 4.2.2 Official fishery-dependent data and data selection ....................................... 77 4.2.3 Environmental data collation ........................................................................ 79 4.2.3.1 Sea-surface temperature (SST) data ....................................................... 79 4.2.3.2 Chlorophyll-a concentration (chl-a) data ............................................... 79 4.2.3.3 Precipitation data.................................................................................... 80 4.2.3.4 Indices of the Southern Oscillation (SOI) .............................................. 80 4.2.4 Effort unit selection ....................................................................................... 80 4.2.5 Standardisation of CPUE .............................................................................. 81 4.2.6 Definition of stock status based on catch ...................................................... 83 4.2.7 Statistical analyses ........................................................................................ 83 iii

4.3 Results .................................................................................................................. 84 4.3.1 Exploratory analysis: distribution by gear .................................................... 84 4.3.2 Exploratory analysis: short- and long-term changes in environmental variables ................................................................................................................. 85 4.3.3 Best measure of effort ................................................................................... 87 4.3.4 Standardisation of CPUE .............................................................................. 88 4.3.5 Lagged-effects of environmental variables on stocks’ CPUE index ............ 90 4.3.6 Trends in catch and CPUE index of demersal stocks in Brunei between 2000 and 2009 ................................................................................................................. 93 4.4 Discussion and Conclusion ................................................................................ 101 4.4.1 Best measure of effort ................................................................................. 101 4.4.2 CPUE standardisation ................................................................................. 103 4.4.3 Potential influence of environmental variability on selected demersal stocks in Brunei ............................................................................................................... 106 4.4.4 Probable cause of non-proportionality ........................................................ 111 4.4.5 Spatio-temporal changes in CPUE of demersal stocks in Brunei ............... 113 4.4.6 General limitations of analyses ................................................................... 115 CHAPTER 5

Utilisation of local ecological knowledge to assess status of selected demersal stocks in Brunei ........................................................................... 117 5.1 Introduction ........................................................................................................ 118 5.2 Materials and Methods ....................................................................................... 119 5.2.1 Data collection ............................................................................................ 119 5.2.1.1 Interview preparation ........................................................................... 119 5.2.1.2 Interview design and procedures .......................................................... 120 5.2.1.3 Survey sampling strategy ..................................................................... 123 5.2.2 Data assembly and analyses ........................................................................ 126 5.2.2.1 Fishers’ perceptions of fish depletion and changes in indicator species (Lutjanus erythropterus) – descriptive approach ............................................. 126 5.2.2.2 Estimating past abundances from LEK - Fuzzy logic approach .......... 127 5.2.2.3 LEK assessment using spatial information technology – GIS approach .......................................................................................................................... 130 5.3 Results ................................................................................................................ 133 5.3.1 Exploratory analysis: distribution of small-scale fishers by gears .............. 133 5.3.2 Perceptions of decline in indicator species (Lutjanus erythropterus) and changes in selected stocks from fishers’ interviews ............................................ 134 5.3.2 Change in abundance of selected stocks from fuzzy logic output .............. 139 iv

5.3.3 GIS outcomes as indicators of spatial and temporal patterns of red snapper distributions and fishers’ fishing activities .......................................................... 140 5.3.4 Qualitative assessment of fishers’ opinions on current fishery management in Brunei ................................................................................................................... 146 5.4 Discussion and Conclusion ................................................................................ 147 5.4.1 Extracting Brunei’s local ecological knowledge on fishery ....................... 147 5.4.2 Evaluating past abundance of Brunei fishery resources ............................. 154 5.4.1 State of ‘shifting baseline syndrome’ in Brunei .......................................... 156 5.4.2 Concluding remarks .................................................................................... 159 CHAPTER 6

General Discussion – Towards sustainable management of the demersal fisheries of Brunei Darussalam................................................................... 160 6.1 Introduction ........................................................................................................ 161 6.2 Managing the demersal fishery resources of Brunei Darussalam ...................... 162 6.2.1 Legal and institutional framework .............................................................. 162 6.2.2 Objectives of fisheries management in Brunei ........................................... 163 6.2.3 Current aspects of demersal fishery management ....................................... 165 6.3 Key issues .......................................................................................................... 166 6.3.1 Overexploitation of the fisheries resources ................................................. 166 6.3.2 Inadequate information on the fisheries resources ...................................... 167 6.3.3 Open access to the fisheries resources ........................................................ 169 6.3.4 Inadequate monitoring, control and surveillance capacity .......................... 170 6.3.5 Regulatory ambiguity on recreational (sport) fishery ................................. 171 6.3.6 Consideration of climate change and environmental variability................. 173 6.3 Monitoring, assessment and/or management recommendations ........................ 174 6.3.1 Licensing requirement ................................................................................. 175 6.3.2 State-of-the-art monitoring system ............................................................. 175 6.3.3 Access rights ............................................................................................... 176 6.3.4 Co-management and the participatory approach......................................... 177 6.3.5 National fishery statistics ............................................................................ 179 6.4 Concluding remarks ........................................................................................... 179 CHAPTER 7

General conclusions and suggestions for future work ............................. 181 7.1 Conclusions ........................................................................................................ 182 7.2 Suggestions for future work ............................................................................... 184 v

REFERENCES......................................................................................................... 186 APPENDICES ......................................................................................................... 202 Appendix A-3.1 (Trophic spectra trends) ............................................................ 202 Appendix B-4.1 (GLM models of stocks’ CPUE) ............................................... 203 Appendix C-4.2 (Missing value patterns of data) ................................................ 211 Appendix D-4.3 (Plots of CCF analyses) ............................................................. 212 Appendix E-5.1 (Interview questions) ................................................................. 215 Appendix F-5.2 (Example of ‘logical consistency’) ............................................ 217 Appendix G-5.3 (Interview survey input data) .................................................... 218 Appendix H-5.4 (List of depleted species from interview survey) ...................... 219 Appendix I-5.5 (Selected pelagic stocks’ trend from official data) ..................... 221 Appendix J (Manuscript published) ..................................................................... 222

vi

List of Figures 1.1

Map of Brunei Darussalam and the surrounding area ......................................... 6

1.2

The fishing zones in Brunei waters, used for management purposes. .............. 10

1.3

Three types of vessels used in Brunei’s large-scale fishery.............................. 12

1.4

The small-scale sector of Brunei fishery........................................................... 13

1.5

Annual change in number of small-scale fishers in Brunei from 2000 to 2008 .................................................................................................................. 14

2.1

Trends in status of fisheries stocks on the basis of stock assessment time series of (a) catches and (b) biomass for the same set of stocks....................... 29

2.2

Relationship between CPUE and abundance based on different values of the shape parameter (β) .......................................................................................... 31

3.1

Study area corresponding to Zone 2 and 3 of Brunei fishing zonation ............. 52

3.2

Annual fishery catches of Brunei waters between 2000 – 2009 based on functional groupings ......................................................................................... 60

3.3

Annual trends in effort for the three LS fishery fleet ........................................ 61

3.4

Seasonally-adjusted mean trophic level (MTL) of fishery catches in Brunei during the period 2000 – 2009 .......................................................................... 62

3.5

Absolute Fishing-in-balance (absFiB) index of fishery catches in Brunei (2000-2009) ...................................................................................................... 63

3.6

Changes in composition of species caught in Brunei waters between 2000 and 2009............................................................................................................ 64

3.7

Analysis of Brunei fishery catches discriminated by trophic level ................... 65

3.8

Catch trophic spectra of Brunei’s LS fishery from 2000 to 2009 ..................... 67

4.1

Proportion of number of different types of species caught by each gear in Brunei ............................................................................................................... 84

4.2

Contribution of the different gears on the overall demersal finfish catch by zone over the studied period (2000-2009) ........................................................ 85 vii

4.3

Seasonal variation of (a) sea surface temperature (SST), (b) chlorophyll-a concentration (chl-a), (c) rainfall, and (d) the Southern Oscillation Index (SOI) in Brunei between 2000 and 2009 .......................................................... 86

4.4

PCA outcome of four different types of efforts ................................................ 87

4.5

Overall demersal (a) catch (in MT) and (b) standardised CPUE (MT/day) of Brunei waters between 2000 and 2009 ............................................................. 93

4.6

Annual catch (MT) of selected family stocks between 2000 and 2009 ............ 95

4.7

Annual standardised CPUE (MT/day) of selected family stocks between 2000 and 2009 in Brunei................................................................................... 99

5.1

Obtaining local ecological knowledge from fishers in Brunei, whereby interviews were conducted in private ............................................................. 121

5.2

Map showing four access-points for the interview survey of fisheries in Brunei ............................................................................................................. 124

5.3

Fuzzy set membership for fisher abundance scores ........................................ 128

5.4

Process flow for the construction of the two classifications using GIS .......... 131

5.5

Distribution of gears among small-scale fishers interviewed in Brunei between March and May 2011 ....................................................................... 133

5.6

Linear regressions of interview data showing (a) best day’s catch, (b) weight of largest red snapper, (c) distance offshore associated with best day’s catch, and (d) water depth associated with best day’s catch, against fisher experience ....................................................................................................... 135

5.7

Linear regressions of interview data showing (a) best day’s catch, (b) weight of largest red snapper, (c) distance offshore associated with best day’s catch, and (d) water depth associated with best day’s catch, against reported year . 136

5.8

Proportion of fishers reporting changes in depletion indicators since when they first started fishing .................................................................................. 137

5.9

Evidence of shifting cognitive baselines in Brunei fishers ............................. 138

5.10 Relative abundance of selected stocks in Brunei over five time periods from fuzzy logic algorithm ...................................................................................... 139 5.11 Nearby major natural / artificial reefs or shipwrecks associated with the reported red snapper harvesting location in Brunei waters ............................ 140 viii

5.12 Distribution of red snapper presence scores based on fishers’ LEK in Brunei141 5.13 Spatiotemporal pattern of fishers’ best day’s catch in Brunei ....................... 142 5.14 Distribution of high pressure harvest area (HPA) based on 236 fishers’ most frequently visited sites .................................................................................... 144 5.15 Range of HPA scores generated for Brunei waters using 236 fishers’ most frequently visited sites .................................................................................... 145

ix

List of Tables 1.1

Differences between the large-scale and small-scale fisheries sectors in Brunei. .............................................................................................................. 11

3.1

Trophic level (TL) of the main species landed from Zone 2 and 3 of Brunei fishing zonation, from 2000 to 2009 ................................................................ 54

3.2

Results of linear regression analyses of catches of different groups as a function of years in Brunei waters .................................................................... 60

3.3

Results of linear regression analyses for efforts of different fishing fleet as a function of years in Brunei waters .................................................................... 61

3.4

Results of linear regression analyses of different type of MTL as a function of time in Brunei waters ................................................................................... 62

4.1

Break down of SS and LS fisheries by gears and principal family/group of demersal fish exploited in Brunei ..................................................................... 76

4.2

Original criteria used by Froese and Kesner-Reyes (2002) for assigning exploitation stages to fisheries .......................................................................... 83

4.3

Summary statistics of seasonally-adjusted environmental variables in Brunei over the study period, from January 2000 to December 2009 .......................... 86

4.4

Principal component loadings from the correlation matrix between the four different types of effort ..................................................................................... 87

4.5

Most parsimonious models used in the standardization procedure of nominal CPUE of overall demersal resources and of specific categories at family level for Brunei waters between 2000 and 2009 .............................................. 89

4.6

Hypotheses linking climate variability and demersal stocks CPUE generated from the CCF analyses...................................................................................... 92

4.7

Trends in annual catch of specific categories of fish and in total demersal finfish catch (MT) from 2000 to 2009 .............................................................. 94

4.8

Summary statistics of test for linear trend in one-way ANOVA to assess trends in annual standardized CPUE of specific categories of fish and in total demersal finfish catch (MT/day) from 2000 to 2009 in Brunei ............... 97

4.9

Comparison in stock abundance between Zone 2 and Zone 3 of Brunei fishing zonation, using standardized CPUE ................................................... 100 x

5.1

Distribution of small-scale fishers interviewed between March and May 2011 in Brunei ................................................................................................ 125

5.2

Results of linear regression analyses of four different red snappers catch variables as a function of fishing experience, ................................................. 135

5.3

Results of linear regression analyses of four different red snappers catch variables as a function of actual years ............................................................ 136

5.4

Effect of fishers’ experience (β) on their perceived change in abundance between the time when they first started fishing and present ......................... 138

5.5

Basic statistics of HPA scores ......................................................................... 145

5.6

Fishers’ negative comments on the local fishery authority and current management practices in Brunei were assessed qualitatively and divided into eight themes ............................................................................................. 146

xi

Acknowledgements Alhamdulillah. My research would not have been possible without generous funding. For essential and much appreciated personal financial support, I am grateful to the Brunei Government for awarding me a Ministry of Education Scholarship. Thank you as well to the Scholarship Unit, Ministry of Education (Brunei) and University of Warwick (UK) for funding my research. No thesis is accomplished alone. My most sincere thanks go to my ‘official’ supervisors, Dr. James Bull and Prof. Charles Sheppard for taking me under their wings after a year into my PhD, following the retirement of my original supervisor, Prof. Andrew Price. I am especially thankful to Jim for challenging me and helping me to broaden my horizons. His guidance, questioning, advice and contributions to this research have been plentiful that I could count on. My original supervisor deserves special thanks, as he gave me the opportunity to study for PhD and has encouraged me from the conception of my research plan through to the submission of my thesis, even when it is beyond his call of duty. Components of my research benefitted from intellectual input from my PhD supervisory committee, Prof. Laura Green and Dr. Paul Neve, whom I gratefully acknowledge. Also thanks to Mr Andrew Mead for his advice on the modelling in Chapter 5. I am tremendously indebted to the Department of Fisheries (Brunei) for their kind permission in letting me carry out this research project. They allowed me to access their fishery databases and reports in which they have invested a lot of resources. My thanks extend to the DOF officers and staffs: Awang Hj. Abdul Halidi Mohd. Salleh (Director, DOF), Awang Matzaini Hj Juna (MFDM), Awang Elviro A Cinco (MFDM), Awang Abdul Hamid Hj Zainin (MFDM) and Awang Ishak Jarudin (MAS) for their help and useful ideas and suggestions in our discussion sessions; Awang Bidin Suru, Awang Mahmud Abdullah, Awang Hj Amzah Hj Edin, and Awang Hj Aji Hj Sapar for their experienced assistance in the field. I am especially grateful to Dayang Ranimah Hj Abdul Wahab, who was then Head of MFDM when I first started my PhD, but now the Deputy Director of DOF. She not only helped me navigate the bureaucracy at DOF, but has also been considerably supportive throughout my PhD. She has been a true inspiration both as an exceptional leader and as someone who is always approachable and ever patient with my queries. I owe much thanks to all the fishers in Brunei who participated in the interview survey. I must also express my gratitude to all contacts I made in my pursuit of maps and data; Dayang Hjh Roseidah Hj Kamis (Survey Department, Brunei), Mr. Warwick Newall (Integrated Environmental Consultants Sdn. Bhd.), and personnel from Brunei Darussalam Meteorological Service and Department of Economic Planning and Development (JPKE). Appreciation is also due to my research assistants, Sarif Mahmud and Aizuddin Roslan, who were with me through every step of data collection in Chapter 5. Also thanks to Hilni Harunsani and Ewana Yusop for proof reading the thesis. Above all, I am grateful to my family and friends for their constant prayers and unwavering support.

xii

Declaration I hereby declare that the work presented in this thesis is the result of original research carried out by myself, under the supervision of Dr. James Bull and Prof. Charles Sheppard, unless stated otherwise. All sources of information presented in this thesis have been acknowledged by a reference. No part of this thesis has been submitted for a degree at another university.

Other than my supervisors, Prof. Andrew Price was part of the development of the proposals of the chapters of the thesis, and specifically provided edits and suggestions for improvement of the drafts of Chapter 5. A manuscript related to Chapter 3, titled “Changes in community structure of finfish catches in Brunei Darussalam between 2000 and 2009”, co-authored with CRC Sheppard, R Wahab, ARG Price and JC Bull, has been published in the peerreviewed journal, Ocean & Coastal Management, 76, 45-51. I conducted data analysis and wrote the manuscript, under the supervision of JC Bull. R Wahab provided the data, while CRC Sheppard and ARG Price reviewed and edited the manuscript.

xiii

Summary A problem commonly encountered in stock assessments of tropical marine resources in developing countries is data paucity, which invariably results from the lack of both human and economic capacity within the government to implement and maintain programmes for data collection and analysis. With special reference to the demersal fishery of Brunei Darussalam, this thesis examines approaches for extracting useful information from data-poor fisheries to assess the state of resources and inform fishery management actions. By using official fishery statistics, augmented by local ecological knowledge (LEK) obtained from fishers engaged in either the large-scale (LS) or small-scale (SS) fisheries in Brunei, changes in demersal fishery resources over the years were assessed. The sustainability of Brunei’s demersal capture fishery was evaluated in the face of its ongoing development and climate change.

Using trophodynamic indicators such as mean trophic level (MTL), Fishing-inBalance (FiB), trophic spectra (TS) and community structure analyses, LS fishery catches of Brunei between 2000 and 2009 revealed a deteriorating state of the coastal demersal ecosystem. Closer examination of the abundance of overall demersal finfish stocks, using the Catch-Per-Unit-Effort (CPUE) index – standardised for other factors not related to abundance – indicated a declining trend, even when total catches remained stationary, although trends in abundance of the different demersal fish families varied. This rapid significant change in recent years is further supported by fishers’ LEK on relative abundance of Brunei marine resources. The study on LEK has also revealed the ‘shifting baseline syndrome’ (SBS) among currently active fishers and their exploited populations, a phenomenon not previously reported for Brunei fisheries.

Findings from the study are synthesised with other information, where a number of key issues and policy options are discussed, and recommendations for the management of the fishery are made. This thesis demonstrates that researchers in data-poor fisheries can utilise different assessment tools, given the resources at their disposal, to assist in the management of marine resources.

xiv

List of Abbreviations C

Catch

CCF

Cross-Correlation Function

chl-a

chlorophyll-a concentration level

CPUE DOF EAFM EEZ ENSO FAO FDFW FiB FiSAT

Catch Per Unit Effort Department of Fisheries, Brunei Ecosystem Approach to Fisheries Management Exclusive Economic Zone El Niño Southern Oscillation Food and Agriculture Organisation of the United Nations Fishing Down the marine Food Web Fishing-in-Balance FAO-ICLARM Stock Assessment Tools

FLC

Fish Landing Complex

GIS

Geographic Information System

GLM

General Linear Model

IUU

Illegal, Unreported and Unregulated

LEK

Local Ecological Knowledge

LS

Large-Scale

MEY

Maximum Economic Yield

MSY

Maximum Sustainable Yield

MT MTL

Metric Tons Mean Trophic Level

nm

nautical mile

SBS

Shifting Baselines Syndrome

SOI

Southern Oscillation Index

SS

Small-Scale

SST

Sea Surface Temperature

TL

Trophic Level

TS

Trophic Spectra

VMS

Vessel Monitoring System

xv

To Mama and Babah, and Brunei Darussalam…

xvi

Chapter 1

Chapter 1 General Introduction “Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for a lifetime.” – Chinese Proverb

Chapter 1

1.1 Background For centuries, because of the vastness of oceans and seas, it was widely believed that their resources were inexhaustible and that fishing could go on indefinitely (Costanza, 1999). Time and research have proven that this idea of infinite marine resources is far from right, as the world witnessed the collapse or decline of several fishery resources, the classical example being the Peruvian anchoveta and the North Atlantic cod (Hilborn and Walters, 1992; Myers et al., 1997). Human population growth throughout the world has resulted in immense and growing pressure on most natural resources, and it is generally agreed that conventional approaches to control fishing have failed to sustain fish populations or fisheries (Pauly et al., 2002).

The demersal fishery resources of tropical coastal areas such as Brunei Darussalam consist of highly diverse, multi-species complexes, which are predominantly Kstrategy species (Pauly, 1979), with a general trend for long-term site attachment (Jones et al., 2002) and a relatively small home range (Jones et al., 2008). Unlike their pelagic counterpart, influence of climate variability on this ecological group has not been clearly identified (e.g. Klyashtorin, 2001), and this is further exacerbated by the relatively short time-series of many of the fishery datasets from the tropical regions (e.g. Venkatachalam et al., 2010).

In this study, particular emphasis has been placed on Brunei waters. Despite its relatively small landings and fishing area covered, Brunei marine ecosystem is considered a valuable reference point – a lightly harvested system within a region chronically overexploited (Silvestre and Garces, 2004). Brunei’s stable economy which is largely based on oil and gas, small population, and low reliance on coastal

2

Chapter 1

resources place Brunei in a unique position in terms of fishery development and management challenges. Yet, fisheries research progress in Brunei is relatively slow compared to the neighbouring countries, and the impact of fisheries and climate variability in Brunei waters remains poorly understood. Although fisheries data became available from 1980s onwards, Brunei’s fishery was rendered “data-poor”1, not because of lack of data, but mostly due to lack of reliable data with adequate and appropriate analyses. Previous studies of the demersal stocks in Brunei have not examined the impacts from fishing and climate in great detail, and existing management strategies do not consider natural climate variability of climate change on the dynamics of demersal stocks abundance.

Globally, the estimated fish consumption per person per year reached an all-time high of 18.8 kg in 2011 (FAO, 2012b). In Brunei and other coastal Southeast Asian countries, fish is the traditional and preferred source of animal protein, and will remain so in the foreseeable future. Despite a decrease in fish contribution to supplies of total protein from 16% in 1961 to 10% in 2007, the per capita consumption of fish for Brunei remained one of the highest in the world at 33 kg per year for 2010 (FAO, 2012a). However, FAO’s The State of World Fisheries and Aquaculture 2012 reported that about 87% of monitored marine stocks worldwide are now fully exploited, overexploited or depleted with no further potential for increasing marine catches. A sustainable fishing industry, therefore, is of considerable importance in Brunei, in line with the country’s efforts to diversify from the dominant petroleum industry, and reduce the heavy dependence on imported

1.

“Data-poor” – a condition to describe a fishery that lacks sufficient information to conduct a conventional stock assessment; this includes fisheries with few available data, as well as fisheries with copious amounts of data but limited understanding of stock status due to poor data quality or lack of data analysis (Honey et al., 2010)

3

Chapter 1

fresh fish supply and attempts at self-sufficiency (Silvestre and Matdanan, 1992; DOF-MIPR, 1992). To understand the fishery and enable sustainable exploitation of fishery resources, it is essential, not only to recognise the dynamics of the resources, but also to understand the marine environment and their effects on fishery production, hence this study.

1.2 Brunei Darussalam, coastal features, and climate and environmental status Brunei Darussalam is a coastal state with a land area of 5 765 km2, located in the north-western part of Borneo Island and bordered by the east Malaysian state of Sarawak (Figure 1.1). The country has a 161 km long coastline fronting the South China Sea, dominated with sandy beaches, mud flats and estuaries with mangroves and peat swamps. The total marine territory covers an area of 38 600 km2, within the 200 nautical mile (nm) limit of Exclusive Economic Zone (EEZ). This area is bounded by the Malaysian state of Sabah on the eastern side and Sarawak on the western side. Brunei waters is also included in the Palawan/North Borneo Marine Ecoregion, which borders the north western part of the Coral Triangle – the most diverse and productive marine system in the world (The Nature Conservancy, 2004; Allen, 2008).

Coastal features The territorial waters of Brunei are characterised by a narrow continental shelf (within the 200 m isobaths) of about 8 600 km2, and an offshore area of about 30 000 km2 within which depths reach ≥ 2 000 m depth. The bottom substrate is sandy between the shoreline to around 30 m depth and around Ampa Patches. Depth gradients between 10 m and 40 m are irregular due to sand deposition by longshore 4

Chapter 1

drift and raised areas covered by coral/hard grounds (e.g. Ampa and Champion areas; Silvestre and Matdanan, 1992).

Brunei has three large estuaries (Brunei, Tutong and Belait) covering 400 km2, which serve as important nursery/feeding grounds for fishery resources and are sites of significant fisheries and traditional human settlements (i.e. water villages) (DOFMIPR, 1992). The mangrove forests of Brunei make up 3.2% (184 km2) of the country’s total land area and provide coastal protection and harbour unusual wildlife (DOF-MIPR, 1992). The highly diverse coral reefs of Brunei cover 50 km2 as claimed by the government of Brunei (DOF-MIPR, 1992; Turak and DeVantier, 2011), although authors from the Reef at Risk in Southeast Asia (RRSEA) project estimated the coral reef and coralline areas to cover roughly 200 km2, including fringing reefs, patch reefs, and one atoll (Burke et al., 2002)1. An estimated 185 species of stony corals belonging to 71 genera and over 150 fish species from more than 30 families have been reported (DOF-MIPR, 1992). These numbers were found to increase in the recent Rapid Marine Survey of Brunei’s Coral, Fish and Sea Shell Biodiversity in 2009 – over 650 species of fish, 404 species of reef building coral and 330 species of bivalves and gastropods were identified (Scubazoo, 2009). The major reefs include Ampa patches, Brunei patches and Champion shoals (Figure 1.1b). There are also several artificial reefs made up of tyres, galvanised pipes, concrete piles and redundant oil platforms which have been placed at several sites parallel to the coastline, mainly for fisheries purposes (DOF, 1999; DOFpers.comm.).

1

Meanwhile, the GIS analysis carried out in this study revealed the coral grounds of Brunei to cover an area of c.154 km2.

5

Chapter 1

FIGURE 1.1: Map of Brunei Darussalam and the surrounding area (a) Geographic location of Brunei in relation to other Southeast Asian countries; (b) Depth distribution, as well as the distribution of oil industry structures and coral grounds in the coastal waters of Brunei. The four districts of Brunei are also shown.

6

Chapter 1

At present, there are nine offshore oil/gas fields in operation, with connecting pipelines between the platforms (Figure 1.1b). Over 60% of Brunei’s gross domestic product (GDP) is contributed by the oil and gas industry, so any fisheries and environmental development program must ensure safety and unhampered operation of the sector’s offshore installations. Since 2000, a 500 m no fishing “buffer zone” surrounding the oil platforms has been prescribed by the petroleum companies, while the Government implements a “no take” zone of 1 nm from both sides of the oil/gas pipelines and structures.

Environmental status Much of the natural environment in Brunei is claimed to be in good, pristine or near pristine conditions (Khartini, 2004). However, the coastal region, where more than 85% of Brunei’s population is living (Ranjith and De Silva, 1998 - figure might actually be higher now), is under increasing anthropogenic pressures as the population continues to increase. Brunei’s population growth rate is reported to be one of the highest in Asia, at around 2% per annum (WHO, 2012), and in the light of industrialization and urbanisation, almost all of the major economic activities are confined to the coastal zone. Since almost the whole of the northern coastal belt between Kuala Belait and Muara consists of soils not suitable for any large-scale agriculture use, the coastal land has been used for urban, residential and industrial development (dela Cruz et al., 1987; DOF-MIPR, 1992). Consequently, the considerable land developments that took place resulted in some undesirable coastal environmental impacts, including siltation from land due to erosion, as a result of unplanned land clearance, untreated waste discharges as well as indiscriminate

7

Chapter 1

cutting of mangroves for residential and industrial development, aquaculture, fuel wood and timber.

Climate The climate of Brunei is governed by its location within the equatorial tropics, and the wind systems of Southeast Asia – in a form of two monsoon seasons – which result from the atmospheric pressure distribution over the region as a whole (Sirabaha, 2008). During the northern hemisphere winter months from December to March, the northeast monsoon winds affect the South China Sea and Borneo, while the broad flow of southwest monsoon winds move across Brunei from June to September. The first transitional period occurs in April and May while the second one in October and November. The mean monthly rainfall indicates certain seasonality which reflects the two monsoon seasons in conjunction with the related movements of the “Inter Tropical Convergence Zone” (ITCZ) and the influence of the localized land-sea circulations. On an inter-annual timescale of three to seven years, the climate of Brunei is also influenced by El Niño Southern Oscillation (ENSO). The warm (or cold) episode of El Niño (or La Niña) is normally associated with prolonged dry conditions (or wetter than normal) in Brunei (Sirabaha, 2008).

1.3 State of marine capture fisheries in Brunei Darussalam Brunei has a long fishing tradition and early European accounts as far back as the sixteenth century attest to the significance of fisheries as a way of life. The fishing industry, comprising capture, aquaculture, and seafood processing sectors, is one of the most important non-petroleum industries in Brunei, and under the current National Development Plan (2007 – 2012), BND $115M (around GBP £65M) was 8

Chapter 1

allocated for the continued development of the country’s fishing industry (BEDB, 2012).

In Brunei, the marine capture fisheries contributed about 70% of the total supply of fish products and can be characterised as multi-species and multi-gear (Matzaini et al., 2007; JPKE, 2010). Total annual production from the capture fisheries increased markedly at the turn of the 21st century, from 6 705 metric tons (MT) in 1997 to 15 329 MT in 2010 (JPKE, 2010). The bulk of fisheries activities are concentrated in Brunei’s estuarine system and in the coastal waters off Brunei-Muara district which has the highest concentration of human population.

1.3.1 Fisheries management and research The Department of Fisheries (DOF), under the Ministry of Industry and Primary Resources, is the authoritative body in charge a wide array of functions, including fisheries research, extension, enforcement, marketing, conservation, development and management of Brunei fisheries (Silvestre and Matdanan, 1992). In particular, the Marine Fisheries Development and Management division, under this department, is responsible for the research, assessment, monitoring and management of marine fisheries resources in Brunei waters (DOF, 2009). Like many other Asian countries, the fishing area in Brunei is divided into “fishing zones” established in 1990, which act as a spatial management tool to restrict fishing in particular areas, based on the distance from the coastline (Garces et al., 2006 ; Figure 1.2). Other existing management mechanisms are reviewed in Chapter 6 (Section 6.2).

9

Chapter 1

Fundamentally, the Government of Brunei started to manage the fisheries in Brunei waters since the late 1960s after the establishment of the DOF in 1966. However, technical capacity and scientific knowledge in the country are quite limited, so Brunei is seeking help from regional and international organisations in executing comprehensive assessments and monitoring programs (Burke et al., 2002; DOFpers.comm.). Earlier studies on fishery and stock assessments of Brunei are reviewed in Chapter 2 (Section 2.5).

FIGURE 1.2: The fishing zones in Brunei waters, used for management purposes. Red circle shows the location of Muara Fish Landing Complex (FLC), which is the main FLC in Brunei. Map adapted from Matzaini et al, 2007.

10

Chapter 1

1.3.2 The fishing fleets The marine capture fisheries of Brunei, well described – albeit outdated – in a review by Silvestre and Matdanan (1992), are divided into large-scale (LS) and small-scale (SS) sectors based on the level of capital inputs and associated characteristics (Table 1.1).

TABLE 1.1: Differences between the large-scale and small-scale fisheries sectors in Brunei.

Large-scale sector Permitted fishing area

Zone 2, 3 and 4

Common types of gears(a)

Purse-seines Bottom trawls Horizontal bottom long lines and large fish traps

Vessels characteristics

Inboard engines Equipped with navigational instruments No refrigeration system

Fish marketing

1) Fishmongers Retailers Consumers 2) Retailers Consumers

Working time(b)

Full-time

Landing sites

1) Muara Fish Landing Complex (FLC)

Small-scale sector Zone 1, 2, 3 and 4 – Individuals Zone 2, 3 and 4 – Companies Surrounding nets Seine nets Lift net Falling gear Gill nets Scoop nets Traps Hook and Line Others (i.e. gleaning) 0 or 1 or 2 outboard engines Some equipped with navigational instruments No refrigeration system 1) Fishmongers Retailers Consumers 2) Retailers Consumers 3) Consumers 4) Own consumption Full-time Part-time Any place suitable for landing, but common sites include 1) Jerudong Beach 2) Pengkalan Sibabau 3) Danau 5) Seria Beach 6) Lumut Beach 7) Kuala Belait FLC

(a) Only main categories of fishing gears used by small-scale fishers are shown, following the classification and definition of FAO. (b) Full-time fishers refer to those solely dependent on fishing for their livelihood (time spent fishing ≥ 80%), while part-time fishers are those whose chief income sources are from other occupations but who fish during weekends, holidays or times of peak fish abundance (time spent fishing ≤ 10%) (Silvestre and Matdanan, 1992, Ranimah 2008).

11

Chapter 1

1.3.2.1 The large-scale sector Silvestre and Matdanan (1992) classified the large-scale sector as “modern”, “commercial”, “industrial” and “offshore”. LS vessels operate only from Muara where a fish landing complex (FLC) is available. There is also another FLC facility offered in Kuala Belait, although there are no LS companies currently based there.

The LS vessels can be subdivided into three types based on the gears used, namely trawlers (kapal pukat tunda), purse seiners (kapal pukat lingkung) and long liners (kapal rawai) (Figure 1.3). Although identified as long liners, the latter group of vessels are also known to utilise large fish traps (bubu) in most of their operations.

(a)

(b)

(c)

FIGURE 1.3: Three types of vessels used in Brunei’s large-scale fishery; (a) bottom trawlers (kapal pukat tunda), (b) purse-seiners (kapal pukat lingkung), and (c) long liners (kapal rawai).

The LS sector was launched in the 1980s, and all vessels were locally owned (Khoo et al., 1987). In 2011, there are a total of 43 LS vessels (21 trawlers, 12 purse seiners and 10 long liners) registered with DOF. LS vessels’ skippers are required to be certified in Fishing Technology and Navigation course conducted by DOF, and are

12

Chapter 1

also mandated to complete a monthly fishing logbook. All of the vessels are fitted with radars, echo-sounders and VHF radios. The LS vessels are restricted from operating in Zone 1 but this separation does not always work in practice.

1.3.2.2 The small-scale sector The SS sector, which is also indicated as “traditional”, “artisanal” and “near shore” by Silvestre and Matdanan (1992), is quite old but its nature has changed with time. The sector’s fishing techniques had improved over the years as modern electronic devices such as global positioning system (GPS) equipment, echo-sounders and fishfinders become cheaper and easily available. The boats are also well-constructed, easy to manoeuvre and commonly fitted with the latest models of outboard motors. (a)

(b)

(c)

(d)

FIGURE 1.4: The small-scale sector of Brunei fishery can be divided into two groups; (a-b) single boat operated small-scale fishery, and (c-d) fleet operated small-scale fishery.

13

Chapter 1

At present, this sector can be further classified into two groups based on their mode of operation (Figure 1.4): single boat operated (“small-scale individual” or SSi) and fleet operated (“small-scale company” or SSc) (Ranimah, 2008). The SSi group use only a single boat operated by the owner himself, and can either be full-time or parttime fishers. The SSc group is differentiated by having more than one fishing boat managed by one owner. A company usually owns the fleet, with foreign labour commonly employed to operate the boats and fishing gears. All SSc fishers are fulltime fishers.

On average, 70% of the total marine capture production is contributed by SS fisheries. However, this trend has only been observed in the recent years, which seems to coincide with the drastic increase in the number of fishers. Indeed, the total number of full-time fishers in Brunei nearly tripled, while the number of part-time fishers almost quadrupled between 2000 and 2005 (Figure 1.5; Matzaini et al., 2007; Ranimah, 2008).

Total number of fishers

Part-time fishers

Full-time fishers

6000

Number of fishers

5000 4000 3000 2000 1000 0 2000

2001

2002

2003

2004

2005

2006

2007

2008

Year

FIGURE 1.5: Annual change in number of small-scale fishers in Brunei from 2000 to 2008, taken from Ranimah (2008).

14

Chapter 1

1.3.3 Marine fishery resources of Brunei The marine resources of Brunei waters, typical of fish communities in dominant softbottom areas of the central Indo-West Pacific, are characterised by high species diversity. About 500 species of fish and invertebrates have been reported from the catch of various fishing gears used in Brunei waters – 80% are demersal species while the rest are pelagics. Overall, the fisheries resources of Brunei are estimated at about 21 300 MT in terms of potential yields, which comprised of 12 500 MT for demersal resources (500 MT contributed by the shrimp fishery) and 8 800 MT for pelagic resources (Matzaini et al., 2007).

Assessments of the pelagic resources using hydro-acoustic technique revealed that the pelagic resources were dominated by sardines (Dussumieriidae, Dussumieria), horse mackerels (Carangidae, Carangoides), scads (Carangidae, Decapterus) and driftfishes (Ariommatidae, Ariomma), which account for over 80% of standing stock (Silvestre and Matdanan, 1992).

Catches from trawl surveys conducted in the 1970s and 1980s showed that the demersal resources of Brunei have been consistently dominated by low-value families/groups such as ponyfish (Leiognathidae), goatfish (Mullidae), threadfin breams (Nemipteridae) and sharks/rays, as these groups collectively account for c. 47-68% of demersal biomass between 1970 and 1990 (Silvestre and Matdanan, 1992). In their most recent demersal trawl survey in 2008, the DOF noted that the top three most dominant families for each hauls were Leiognathidae, Mullidae and Priacanthidae, representing up to 27%, 9% and 6% respectively (DOF-pers.comm.).

15

Chapter 1

Overall, fish abundance exhibits spatial and temporal fluctuations, which are strongly related to depth, seasonal life cycles and productivity phasing, and time of the day (Silvestre and Matdanan, 1992).

The demersal resources of Brunei contribute up to 75% of the annual capture fishery production value for the LS sector, which was worth BND $7.4M (GBP £4.5M) in 2008. While assessments of fisheries have been undertaken in Brunei, they are often published in “grey literature” and so not widely available. This ecological group of fish has been classified as over-exploited in the region over past few decades (Pauly, 1988), but its status in Brunei waters remains unclear from the existing literature.

When DOF carried out trawl survey stock assessments in 2000 and again in 2008, an alarming declining trend was detected. In 2008, only a fifth of the overall demersal stock abundance in 2000 was left, based on the mean catch rate during the trawl survey (DOF-pers.comm.). Their finding was rather surprising considering the ‘lightly exploited’ status of demersal stocks in Brunei up till the late 1990s (Silvestre and Matdanan, 1992; Silvestre and Garces, 2004) and the prudent development of the marine capture industry practised by the authority.

16

Chapter 1

1.4 Research aims and objectives In the light of the above gaps in knowledge, the overall aim of this study is to assess changes in demersal fishery resources and harvesting pressure imposed by the fisheries in Brunei Darussalam, while simultaneously evaluating the sustainability of its demersal capture fishery in the face of its on-going development and climate change.

This thesis has four specific objectives: 1) Examine the current state of Brunei marine ecosystem (Chapter 3). 2) Distinguish the effect of fishing, environmental and other drivers of abundance and/or productivity of selected demersal stocks, pertaining to the large-scale fishery sector (Chapter 4). 3) Analyse population changes of selected demersal stocks, based on standardised Government catch statistics and information from an interview survey administered to fishers. (Chapter 4 & 5). 4) Critically review the policy and management framework governing Brunei’s fisheries, and make recommendations to improve management of Brunei demersal fishery and its future research needs (Chapter 6).

The four objectives are addressed over four chapters. The chapters are conceptually connected, with the collective aim to examine the possibilities of extracting useful information from, and to contribute to the improvement in management of, data-poor fisheries in tropical countries such as Brunei Darussalam.

17

Chapter 1

1.5 Thesis organisation: Summary outline of chapters This chapter (Chapter 1) provides a general introduction to the thesis. In addition to defining overall aims and specific objectives of the study, the chapter also describes the current available information and key literature review pertaining to the study area, its fish resources and the fisheries. Chapter 2 is an overview of fish stock assessments as commonly carried out in the tropics, with particular emphasis on ‘data-poor’ fisheries. It includes a comprehensive literature review for subsequent chapters and underpins the conceptual hypotheses and methodology developed in undertaking this study. In Chapter 3, a ‘holistic’ picture of Brunei marine ecosystem during the study period is described via trophodynamic indicators and community structure analyses. Chapter 4 is a detailed analysis on recent abundance trends of overall demersal resources, as well as several commercially-important demersal stocks in the study area, by investigating the joint responses of fish productivity to fishing and climate variability. Through an interview survey of fishers in Brunei, Chapter 5 then explores the utilisation of local ecological knowledge in assessing long-term abundance trends of selected finfish stocks. Chapter 6 provides a synthesis of the study, through which the implication on current management of the fishery resources in Brunei is examined and recommendations on possible improvements to management of the demersal fishery are made. Finally, Chapter 7 presents conclusions and suggestions for future research on demersal fish resources of Brunei.

18

Chapter 2 Review on Assessment of Tropical Fishery Resources “All models are wrong, but some are useful.” – George Box, 1978

Chapter Summary Using published research and opinion pieces, this review illustrates the issues commonly associated with tropical fish stock assessments in developing countries. Taking the limitations of fishery-dependent data into account, development of simple indicators based on such data can provide useful insights into the state of fish stocks and fisheries. In addition, fisheries scientists and managers are gradually engaging with fishers’ local ecological knowledge (LEK) to provide novel and useful information with regards to fish stock status. It has also been increasingly recognised that a likely reason for the causes and mechanisms responsible for variability in fish stock abundance to remain poorly identified may be due to interactions between fishing and the environment. In Brunei Darussalam, earlier studies on fishery and stock assessment revealed that further research on the demersal fishery resources, to complement the fishery authorities’ monitoring programme, is in order.

Chapter 2

2.1 Overview of tropical fish stock assessment The basic purpose of fish stock assessment is to provide advice on the optimum exploitation of aquatic living resources (Sparre and Venema, 1998), and this generally involves performing robust stock assessment techniques on appropriate data for comparison to reference points, and eventually the identification of actions required to achieve the aims of fisheries management. Hilborn and Walters (1992) defined stock assessment as being “the use of statistical and mathematical techniques to make quantitative predictions about the reactions of fish populations to alternative management choices”. Therefore, assessment approaches cannot be developed in isolation from the aims of management, nor the management actions that are to be applied to achieve them (Pilling et al., 2009). Fishery managers increasingly realise that the benefits of fisheries management are not always measured in terms of increases in yields and that the fundamental purpose of fisheries management is to ensure sustainable production over time from fish stocks, in the face of environmental variability (Hilborn and Walters, 1992; Gallucci et al., 1996; Hart and Reynolds, 2002a). Hilborn and Walters (1992) organise the objectives of fisheries management into four categories, namely (1) biological, (2) economic, (3) recreational and (4) social. In most cases the objectives for a particular management regime will be a mixture of the four. In general, however, management often strive to achieve the delicate balance between potentially conflicting aims of ensuring a profitable but sustainable fishery, without specifically assigning priorities to individual aims.

Through stock assessments, fisheries scientists attempt to estimate the amount of fish in a stock and its rate of growth and mortality, and compare those estimates to a 20

Chapter 2

stock’s biological reference points. Numerous definitions of stock concept have been canvassed throughout the fisheries literature (Sparre and Venema, 1998), and a diversity of techniques has been used to identify and classify fish stocks (review by Begg and Waldman, 1999). In its current simplified form, a stock refers to a managed unit of a semi-discrete group of fish, typically based on their geographical location (Begg et al., 1999), while a cohort is a group of fish born in the same year within a population of stock (Hart and Reynolds, 2002b).

The early works in stock assessments took place between the 1890s and 1950s (review by Munro, 2011), where means to estimate growth rates (i.e. via rings in otoliths and scales) and mortality rates (i.e. based on relative abundances of successive year classes) were developed, and foundations of the modern theory of fishing were established. The development of stock assessment concepts was greatly facilitated by the compilation of a set of 16 “key papers on fish populations” published between 1931 and 1981 by Cushing (1983). Eventually, the availability of personal computers started to facilitate enormously the application of stock assessments into feasible undertakings by small government and university fisheries laboratories with limited funding (Munro, 2011).

Stock assessment of a tropical fishery is usually adapted from traditional stock assessment techniques of temperate regions, as mostly described by Beverton and Holt (1957), Ricker (1975) and Gulland (1969, 1983), whose works form the backbone of fisheries science. The development of tropical fisheries science began in the 1980s, where Daniel Pauly became the dominant figure in this field, by introducing and developing suitable methodologies for use by scientists in

21

Chapter 2

developing countries (Pauly et al., 1987). This was motivated by the fact that ageing tropical fishes was, until recently, assumed to be virtually impossible (Morales-Nin, 1992; Choat and Robertson, 2002). For instance, the lack of strong seasonality makes the distinction of seasonal rings, and hence, of year-rings, problematic for many tropical species. In addition, most tropical species would have less distinct spawning periods, with some spawning at least twice per year and often over long periods (Sparre and Venema, 1998). Hence, to apply conventional stock assessment methods of temperate waters on tropical fish, the required parameters must be estimated from length-frequency data, as opposed to data collected for the age-based assessment (Pauly et al., 1987), since growth of fish is revealed to be well approximated by a von Bertalanffy curve (Jobling, 2008; Pitcher, 2008). In a length-frequency analysis, peaks of numbers in the length-classes are used to estimate the mean length of successive cohorts at integer intervals of age, and the relative numbers in these cohorts to estimate total mortality rates. Therefore, when properly sampled and analysed, length-frequency data can provide estimates of key parameters used in conventional stock assessments (Pauly et al., 1987).

With the establishment of the Network of Tropical Fishery Scientists (NTFS) in 1982, a lot of effort was devoted to the development of tropical stock assessment techniques, as well as the development of specialized courses for fisheries scientists from developing countries. This led to close collaboration between groups working at the Food and Agriculture Organisation (FAO) and the then International Centre for Living Aquatic Resources Management (ICLARM) (Venema et al., 1988), which then, among others, led the development of FiSAT (FAO-ICLARM Stock Assessment Tools) software (Gayanilo, 1997). FiSAT made use of applications

22

Chapter 2

based on length-frequency distributions, including length-based Virtual Population Analysis (VPA) and yield assessments for tropical stocks assessments. FiSAT II software is now available in a Microsoft Windows operating environment, with several additions of improved models and packages (Gayanilo et al., 2005). Consequently, a host of new stock assessment programs have emerged in recent years, with some being well established and others in various stages of development (Munro, 2011).

The breakthrough in multispecies management of multi-gears fisheries in the tropics, however, arose with the development of Ecopath, a steady-state ecosystem model, which was further developed to the degree that it has become the de facto world standard for ecosystem modelling of marine resources (Christensen and Pauly, 1992; Christensen and Walters, 2011). The Ecopath with Ecosim (EwE) software suite, which is composed of three main components, namely Ecopath (a static, massbalanced snapshot of the system; Christensen and Pauly, 1992), Ecosim (a time dynamic simulation module for policy exploration; Walters et al., 1997) and Ecospace (a spatial and temporal dynamic module primarily designed for exploring impact and placement of protected areas; Walters et al., 1999), was recently recognised as one of NOAA’s (National Oceanic and Atmospheric Administration, USA) top ten scientific breakthroughs in the last 200 years (Ecopath, 2011).

Nonetheless, the greatest problem commonly encountered in stock assessments of tropical species in developing countries is the lack of funds as governments are reluctant to spend large sums of money on fisheries research and management, especially on the small-scale sector that is not perceived to be of great value (Munro,

23

Chapter 2

2011). Of course, this view is inaccurate, as shown by Sary et al. (2003) who demonstrated that the cumulative cost of non-management of Jamaica’s trap fisheries amounted to US$1.3 billion over the previous 25 years. In fact, employment in the fisheries and aquaculture primary sector has continued to grow faster than employment in agriculture, whereby 54.8 million people worldwide were estimated to be engaged in the primary sector of fish production in 2010, of which Asia accounts for more than 87% of the world total (FAO, 2012b). The reality, however, is that against the priority central governments place on spending on health, education and military, fisheries will remain the poor relations (Prince, 2010).

Subsequently, obtaining a satisfactory sample size for assessment can be expensive, time consuming and labour intensive, particularly for the larger, rarer and more valuable species (Pauly, 1988; Munro, 2011). Where catch data are collected, they are mostly aggregated by families or into even broader groups, and hence, are of little value for stock assessment. With the few exceptions that gather fairly detailed catch data on regular basis, majority of the countries in the tropics would not have detailed catch statistics apart from periodic sample surveys (Munro, 2011). Furthermore, especially for the SS sector in the tropics, the numerous and scattered landings sites, as well as complex and variable market chains often limit the collection of robust data from these fisheries (Stobutzki et al., 2006). In addition to the multispecies and multi-gear nature of the fisheries, the species caught often overlap between gears and sectors, and in some cases the same species may even be targeted at multiple life history stages by different sectors.

24

Chapter 2

Data paucity invariably results from the lack of both human and economic capacity within government to implement and maintain programs for data collection and analysis. Where the “best scientific information available is simply inadequate for determining meaningful reference points and/or current stock status with respect to such reference points” (Richards and Maguire, 1998), fisheries may be considered as “data-poor” (Pilling et al., 2009; Honey et al., 2010). While the term “data-poor” is sometimes confounded with SS fisheries, mostly due to reasons identified earlier, these terms are not synonymous. Large-scale, but recently developed fisheries where fisheries research and management have lagged behind exploitation, or LS fisheries where the quality of data is poor or variable and difficult to assure, may also be deemed data-poor.

However, the Precautionary Approach, proposed by FAO in the International Code of Conduct for Responsible Fisheries (FAO, 1995), declares that the limitations, uncertainties or lack of data for the assessment of stocks or estimation of parameters, cannot be justification for not applying regulation measures. Therefore, managers of data-poor fisheries often face additional pressure to perform their responsibilities to achieve success within data and resource constraints. Realisation is growing across the field of fisheries science that the use of classic stock assessment models to derive relevant science-based reference points such as biomass (B) or effort (F) at maximum sustainable yield (MSY) and/or maximum economic yield (MEY) (i.e. BMSY or FMSY) may have limited application and success in much of the developing world fisheries, particularly with regard to facilitating timely local level fisheries management decisions, and that different approaches are needed (Mullon et al., 2005; SEAFDEC, 2006; Honey et al., 2010; Froese et al., 2012).

25

Chapter 2

One way to overcome this is through the use of simple indicators, or measures of performance, aimed at providing an insight into the development and management of the fishery. Simple indicators are consistently found to outperform more complex (model-dependent) indicators, which are sensitive to data quality (Fulton et al., 2005; Link, 2005). While potentially less precise than complete stock assessments, simple indicators can still provide the necessary information for the formulation of fisheries management policy and enable timely day to day management of fisheries resources by local authorities using identified targets or goals. They may not simply be biological in origin – for instance, indicators can also be socioeconomic in nature, usually by measuring changes in the well-being of the people who are dependent on the fish populations – but indicators need to be locally specific, practical, easy to understand and comprehensible to all local stakeholders (Nielsen et al., 2001).

2.2 Data requirements for development of indicators Conventional stock assessments models require three primary categories of information, namely catch, abundance and biological data (Cooper et al., 2006). Where the data are derived from the fishing process itself, they are classified as fishery-dependent data, and are collected through avenues such as self-reporting (e.g. log books filled by fishers themselves), landings surveys or vessel-monitoring systems (VMS). Fishery-independent data, on the other hand, are collected independently of fishing activities and ideally come from statistically well-designed research surveys and tagging experiments.

The fishery-independent approaches involve a range of standardised sampling gears such as trawls, seines, hydro-acoustics, video and side-scan sonar. Maintaining 26

Chapter 2

standard survey practices over time is crucial, so when new gear or sampling methods are adopted, they should be calibrated so that the results can be directly compared to results from the old gear or method (Cooper et al., 2006). Expected outputs from the fishery-independent methods are then used for estimating fish abundance (can be relative or absolute), demographic structure at sea, as well as for the collection of other biological and biophysical information. For instance, by sampling stomach contents, one can determine a species’ diet (e.g. Ibrahim et al., 2004), while results from tagging experiments may be used to estimate movement or migration rates between stocks, and natural mortality rate of the fish (e.g. de Pontual et al., 2003), all of which enhances stock assessment models (Cooper et al., 2006).

While stock assessments ideally require a combination of both fishery-dependent and fishery-independent data to answer the questions imposed for fishery management, fishery-dependent data are relatively easier and much less expensive to acquire. In most cases, much of the fishery data already exist, either from within or outside the fisheries sector. Undeniably, misreported landings, misinformation and illegal activity hamper the accuracy of such data, but their cautious use may still provide useful insights into the state of fish stocks and fisheries (Sadovy, 2005). Furthermore, considering that most, if not all, countries already have data collection systems of some forms in place, a sensible first step in improving the available information on the fishery sector is to use and assess the current available data (SEAFDEC, 2006). In the event where the statistics are found to be insufficient, corrective action can then be taken to improve their quality.

27

Chapter 2

2.2.1 Assessing stock status using catch-effort data Under limited resources and capacity, efforts have been made to ensure that the minimum requirements of national fishery statistics system are met, often through regional and international initiatives led by FAO and other fishery research and/or management agencies such as, among others, the Southeast Asian Fisheries Development Centre (SEAFDEC) and the Latin American Organisation for Fishery Development (OLDPESCA) (FAO, 2005a). Consequently, the most basic and informative data in fisheries science is the catch and effort data (Caddy and Gulland, 1983). Interestingly, there is an on-going debate on the usefulness of fishery catch data to evaluate stocks status, which specifically causes leading fisheries scientists to disagree over the size of the overfishing problem around the world (Pauly et al., 2013).

In the last 60 years, unquestionably remarkable changes in the nature of fisheries have been observed, especially with the rapid geographical and bathymetric expansion (Morato et al., 2006), along with the ever-increasing efficiency of more powerful fishing fleets (Pauly and Palomares, 2010). Using FAO catch data, the world catch has been observed as stagnating, then slowly declining since the late 1980s (Watson and Pauly, 2001; FAO, 2012b), and this has been widely interpreted as being the result of widespread overfishing leading to sequential depletion of exploited stocks (Jackson et al., 2001; Worm et al., 2006; Froese et al., 2009). Worm et al. (2006) further claimed that most large fish populations are now 10% of their original size and that the oceans could collapse by 2048. However, their claims were challenged by a few authors asserting that the estimates of declining global fish stocks were overstated because they relied on a flawed methodology (Branch et al., 2011). When assessments were based on biomass data (from stock assessments), 28

Chapter 2

instead of catch data in previous studies, the authors found that the global fish stocks are not as badly off as has been widely believed (Figure 2.1). Subsequently, Branch et al. (2011) and a few others (Hilborn, 2010; Daan et al., 2011) suggest that fisheries management has led to stabilisation of fish stocks in most regions, and that the partial stock rebuilding scenarios that have recently occurred in the United States and a few other developed countries may be representative of a global trend.

FIGURE 2.1: Trends in status of fisheries stocks on the basis of stock assessment time series of (a) catches and (b) biomass for the same set of stocks, taken from Branch et al. (2011).

While the former group acknowledged that the stock assessment techniques, and the biomass estimates upon which they depend, are preferable when assessing stock status, they maintained that with cautions and rigorous statistics, data from catches are equally useful, if not better, especially as they have greater coverage in space and time than fishery-independent surveys, and hence more applicable on a global scale than other datasets (Maunder and Punt, 2004; Ye and Dennis, 2009; Froese et al., 2012; Kleisner et al., 2012). Indeed, many reviews have found the efficacy and reliability of stock assessment techniques debatable (Parsons, 1996; Conn et al., 2010). Their main critique is that it is difficult to obtain reliable estimates of stock biomass, even in the best of circumstances, as evidenced by the collapse of Canada’s well-studied Northern cod (Gadus morhua, Gadidae; Myers et al., 1997). In view of 29

Chapter 2

the enormous contributions of developing countries to world fisheries catches, it is unwise to draw credible inferences for the world as a whole from stock assessments which are mostly carried out properly in developed countries. In the context of this thesis, therefore, there is obviously a value in keeping on track with the trends and patterns extracted from fishery catch data.

2.2.2 The ‘Catch per Unit Effort’ model Based on the observation that the size of catch from a fish population typically increases when either population density or effort increases (Hilborn and Walters, 1992), an approach was developed using catch-effort data whereby the catch rate or catch-per-unit-effort (CPUE) is used as an index of abundance, which then in principle can be used to detect declines in the same way as abundance itself. CPUE as an abundance index is based on the relationship that relates catch to abundance and effort:

where Ct is catch at time t, Et is the effort expended at time t, Nt is the abundance at time t, and q is the portion of the stock captured by one unit of effort which are often called the catchability coefficient (Hilborn and Walters, 1992). By rearranging the equation,

CPUE is now proportional to abundance, provided that q is constant over time.

As a result, the assumption that CPUE is directly proportional to abundance is one that is most widely made in quantitative fisheries analysis, with CPUE forming the 30

Chapter 2

basis of stock assessments using catch data for many commercially important species worldwide (e.g. Stergiou et al., 1997; Battaile and Quinn, 2004; Matsunaga and Nakano, 2006; Maunder and Hoyle, 2006; Ortiz de Zarate and Ortiz de Urbina, 2007; Md Nurul Islam et al., 2011). Despite its extensive use, however, there are a number of conditions which may violate the proportionality assumption, namely hyperstability and hyperdepletion.

FIGURE 2.2: Relationship between CPUE and abundance based on different values of the shape parameter (β), taken from Harley et al. (2001).

Hyperstability refers to the situation when the CPUE index remains high while fish abundance is actually low (or abundance declines faster than CPUE). The first two curves where β < 1 in Figure 2.2 show hyperstability, and for resource managers, this means that a population is declining without any change in CPUE to arouse concerns. Hyperstability is considered to be one of the biggest problems for fisheries managers (Hilborn and Walters, 1992), whereby some of the most well known fisheries collapses in the world have been ascribed to hyperstability (Shelton, 2005). The last two curves where β > 1 show hyperdepletion, which happens when CPUE is 31

Chapter 2

actually underestimating abundance (or CPUE drops faster than abundance). The fish population appears to be depleted, despite the fact that abundance overall may be relatively stable. These non-linear relationships between CPUE and abundance may be a result of fishers’ behaviour and/or fish distribution, biology and behaviour. For instance, the inherent schooling behaviour as present in many fish species or fishers’ efficient search behaviour when fishing can lead to hyperstability (Hilborn and Walters, 1992).

Therefore, in the use of catch-effort data, thoroughly understanding the fishery system in question is necessary, especially since other than issues of nonproportionality, there may also be factors other than abundance that can affect CPUE trends over time. For instance, changes in species preference, changes in catchability of a population, and changes in fishing technology, can potentially lead to spurious conclusions regarding the state of fish stocks (Myers and Worm, 2003; Maunder et al., 2006). For this reason, CPUE can only be used as valid indicators of abundance if they are adjusted to account for changes in factors other than abundance that can potentially drive changes in the fisheries (Maunder and Punt, 2004; McCluskey and Lewison, 2008). This is commonly done by standardisation of catch-effort data (see Chapter 4).

Regardless of its well-documented shortcomings, the CPUE index remains an important tool in fishery stock assessment. Since its first use in fisheries by Baranov (1918), numerous studies have then been made to resolve the problems of scale and distribution, non-random variations, and data incompatibilities, along with various procedures suggested for standardizing and model-fitting of the CPUE data

32

Chapter 2

(Maunder and Punt, 2004). Ultimately, development of CPUE as an index of abundance lies in the improvement in the proportionality between the derived index and the true abundance of the stock in the ecosystem (Ye and Dennis, 2009).

2.2.3 The ecosystem-based approach The failure of the traditional stock assessment and management is generally recognised (Garcia and de Leiva Moreno, 2003), and often they are attributed to the inadequacy of the approach to capture multispecies and ecosystem effects (Cury et al., 2005a). This led to a current global call for more use of an ecosystem approach to fisheries management (EAFM) to provide holistic view of ecosystem-fisheries interactions. The EAFM is defined by FAO as “the extension and integration of conventional methods of management of marine resources, stressing the close interdependence between the welfare of humanity, the preservation of the environment and the need to maintain productivity of ecosystems for present and future generations” (Garcia et al., 2003). To cater to this new type of ecosystem management, there is a need for descriptive indicators that reflect and describe the complex interactions between fisheries and marine ecosystems (Pauly and Watson, 2005). This is recognised and broadly accepted by the international community, thus leading to the recent development of several ecosystem-based indicators formulated for fisheries management purposes. In view of the relevance to the relatively datapoor multi-species tropical fisheries, several fisheries scientists agreed that trophic indicators are most useful in supporting the implementation of EAFM, especially as they are easily understood, measurable and can be used by non-scientists to make management decisions (Cury et al., 2005b; Munyandorero and Guenther, 2010; Pennino et al., 2011).

33

Chapter 2

The Marine Trophic Index (MTI) is one of the eight indicators which the Conference of the Parties of the Convention on Biological Diversity (CBD) identified in February 2004, to monitor progress toward reaching the target of significantly reducing the current rate of biodiversity loss (CBD, 2004). Essentially, the term MTI is the CBD’s name for the mean trophic level (MTL) of fisheries landings, which was originally introduced by Pauly et al. (1998), who had demonstrated that fisheries, since the 1950s, are increasingly relying on smaller, short-lived fishes and invertebrates from the lower parts of both marine and freshwater food webs. The origin of the trophic level concept and their usefulness in summarising fisheries impact on marine ecosystems has been reviewed by Pauly and Watson (2005). In brief, the MTL of fisheries landings can be used as an index of sustainability for the exploited ecosystems because fisheries tend to remove large, slow-growing fishes at first, thus reducing the MTL of the fish remaining in the ecosystem. This eventually leads to declining trends of MTL in the catches extracted from that ecosystem, a process now known as ‘fishing down marine food webs’ (Pauly et al., 1998). Using FAO catch data and trophic levels (TLs) of all species or groups of species that contribute to global catches, Pauly et al. (1998) showed that the global MTL significantly declined from 1973 to 1994, at a rate of about 0.1 TL per decade. Their original work gave rise to criticisms (Caddy et al., 1998; Caddy and Garibaldi, 2000), which led to further elaboration of the fishing down concept (see Pauly (2010) for discussion). Subsequently, Essington et al. (2006) pointed out that instead of sequential collapse/replacement of the high-trophic level fisheries, a declining MTL may also indicate a serial addition of the low-trophic level fisheries, a process they termed as ‘fishing through marine food webs’. Essington et al. (2006) further emphasized that the sequential addition mechanism was by far the most common one

34

Chapter 2

underlying declines in catch MTL of large marine ecosystems worldwide. Specifically, they found that out of the 30 ecosystems studied where a decline in MTL was observed, only 9 ecosystems showed declining catches of higher-TL species, compared with 21 ecosystems that exhibited either no significant (n=6) or significant increases (n=15) in high-TL catches (Essington et al., 2006). Their work, however, was contradicted by the fact that globally, catches are not increasing, and in fact are going down (FAO, 2012b).

A more elaborate review on the use of catch MTL as an indicator appeared recently, refuting Pauly et al. (1998) by indicating that the MTL of the worlds’ oceans are in fact stable or increasing (Branch et al., 2010). Instead of using the catch data, as done in earlier studies, Branch et al. (2010) had used scientific estimates of abundance within ecosystems (i.e. from stock assessments) to come to their conclusion. In response, Pauly argued that their analysis might be flawed by their non-consideration of fisheries expansion, which are particularly evident in the ‘Sea Around Us’ data they used (see Swartz et al., 2010; Stergiou and Tsikliras, 2011).

While such divergence in findings had attracted large media attention (e.g. Reuters1, The Guardian2, The Economist3), more important for fisheries scientists and managers alike is the fact that these studies have shown that MTL, especially with better taxonomic resolution of catch data, could be a well-suited indicator that can measure the overall health and stability of the marine ecosystem. Thus mandated by the CBD as a primary indicator of marine biodiversity and health, this index can indicate how abundant and rich the large, high trophic level fish are, and also 1

http://www.reuters.com/article/2009/07/30/idUSN30463 http://www.guardian.co.uk/world/feedarticle/8635355 3 http://www.economist.com/node/14159943?story_id=14159943 2

35

Chapter 2

indicate the extent that the fishing effort within a country’s fishing grounds is modifying its fish stocks (CBD, 2004; Pauly and Watson, 2005; Branch et al., 2010).

In response to a critical review by a group of FAO staff (Caddy et al., 1998), who argued that the FAO statistics used by Pauly et al. (1998) were not detailed and reliable enough to support the ‘fishing down’ interference drawn from them, a Fishing-in-Balance (FiB) index was also developed by Pauly et al. (2000) to assess whether a fishery is balanced in ecological terms or not. The theory behind the development of the FiB index is well described in a review by Pauly and Watson (2005) and Pauly (2010). Briefly, FiB is calculated on the basis that any decline in MTL of fisheries catches should be matched by an ecologically appropriate increase in these catches, and the appropriateness of that increase is being determined by the transfer efficiency between TL (Pauly, 2010). Subsequently, values of FiB index should (1) remain constant if the TL-change match ‘ecologically correct’ changes in catches; (2) increase (> 0) if either bottom-up effect occurs (i.e. increase in primary production), or if a geographical expansion of the fishery occurs; or (3) decrease (< 0) if discarding occurs that is not considered in the ‘catches’, or if the fisheries withdraw so much biomass from the ecosystem that its functioning is impaired. While FiB requires the assumption that transfer efficiency is constant across trophic levels (Pauly et al., 2000), it is believed to provide a better indicator of ecosystem change than catch or catch composition, because of its integrative nature (Garcia and Staples, 2000).

36

Chapter 2

2.3 Contribution of local ecological knowledge (LEK) in stock assessment and fishery management Since the pioneering work of Johannes (1981b), the intrinsic value of local operators’ knowledge of the ecosystem within which they work has been increasingly recognised and used (Haggan et al., 2007). Within fisheries science and management, local ecological knowledge (LEK) research further progressed when Pauly (1995) published an article on “Anecdotes and the Shifting Baseline Syndrome of Fisheries”, in response to the collapse of cod and other fish stocks, together with the fisheries and coastal economies they support, in the northwest Atlantic between 1992 to 1995 (Harris, 1998). Accordingly, key information missing from scientific datasets may be found in stories and anecdotes that described past conditions of marine species and ecosystems, although not in a form easily quantified (Pauly, 1995).

LEK can be defined as knowledge held by specific group of people about their local ecosystems (Olsson and Folke, 2001). As fisheries system worldwide move towards EAFM, large amounts of ecological and social information are required and the use of LEK has been described as an asset for an efficient implementation of EAFM (Garcia and Cochrane, 2005; Paterson and Petersen, 2010). Consequently, fisheries scientists and managers are increasingly engaging with fishers’ LEK to provide novel and useful information to improve the legitimacy of fisheries governance, as fishers’ LEK may encompass a finer scale resolution and provides a longer historical perspective than other data sources (Dulvy and Polunin, 2004; Ames, 2007; Lavides et al., 2009).

37

Chapter 2

By conducting personal interviews with fishers, researchers can elicit large amounts of information pertaining to the past and the present for both commercial and noncommercial species – information which can be very useful in scientific stock assessments (Johannes and Neis, 2007). For instance, local knowledge of the time and place fish are caught can indicate seasonal and directional fish movements, while information on spatial and other changes in effort and fishing practices can be beneficial in interpreting catch-rate data (Hutchings, 1996; Neis et al., 1999). Fishers’ LEK is now used for many purposes, including the deciphering of ecological interactions (Folke, 2004) and dealing questions concerning specific issues, such as fish aggregation (Moreno et al., 2007), habitats (Bergmann et al., 2003), trophic relationships (Pikitch et al., 2004) or the identification of fishing areas (Davis et al., 2004). Moreover, researchers are made aware not only of ecological processes but also of customary tenure and traditional management systems that have been eroded through the interactive effects of external management interventions and resource degradation (Johannes and Neis, 2007).

The accuracy and precision of qualitative information such as LEK, nevertheless, might not be comparable to scientific data. Several researchers (e.g. Silvano and Valbo-Jorgensen, 2008; Ruddle and Davis, 2011; Daw et al., 2011) have warned that although fishers’ local experiences and observations are important, they may not be able to characterise accurately such key attributes of the ecosystem processes such as predator-prey dynamics and seasonality. For instance, Ruddle and Davis (2011) demonstrated that the Canadian researchers had disproved fishers’ contentions that white hake (Urophycis tenuis) was the main predator on juvenile lobster (Homarus americanus) in St. Georges Bay of Nova Scotia, while Daw et al. (2011) had

38

Chapter 2

identified factors which may have resulted in fishers’ more pessimistic perceptions of trends in fish abundance in Seychelles artisanal trap fisheries compared to the landings data and underwater visual census (UVC).

As a result, dispute can arise between scientific perspectives on resources and those of resource users (Daw et al., 2011), often focusing on the validity of either perception and the question of which is correct. Interestingly, dispute can arise simply because they are based on observations of different parts of the fisheries system (Daw et al., 2011). Fishers and scientists may perceive the system at different scales (Berkes, 2006), or through monitoring different variables (Verweij et al., 2010). Inaccuracies or biases can affect both scientific and fishers’ perceptions, due to the context in which perceptions are formed. Undeniably, different assumptions underlying the analysis of interview data compared to conventional scientific data could also have led to qualitatively different trend perceptions (Daw et al., 2011).

2.3.1 The “Shifting Baseline Syndrome” While diverging perceptions of resources are common between fishers and scientists (Gray et al., 2008), many studies have increasingly recognised and acknowledged the complementary nature of LEK to scientific knowledge (Silvano et al., 2005; Berkes et al., 2007; Hall et al., 2009). Even though the reliable use of LEK as a basis of management decisions remains a matter of debate, various approaches for increasing its validity have been proposed (Davis and Wagner, 2003; Maurstad et al., 2007). Moreover, in executing fishers’ LEK research, scientists may uncover the shifting baseline syndrome (SBS) mentioned earlier.

39

Chapter 2

SBS is a psychosocial phenomenon whereby each generation of stakeholders – whether fishers or scientists – accepts a lower standard of resource abundance as being normal (Pauly, 1995). SBS can operate at a societal level, resulting from younger generations being unaware of past abundance (termed generational amnesia), or as a result of individuals forgetting previous abundances (termed personal amnesia) (Papworth et al., 2009). Subsequently, while it has been shown that reliance on scientific knowledge only, especially for mixed species and over a limited period, can be risky and easily mask true stock status (Saenz-Arroyo et al., 2005a; Saenz-Arroyo et al., 2005b; Venkatachalam et al., 2010), presence of SBS in a LEK system invoke a potential problem in a variety of conservation and fisheries governance context (Sheppard, 1995; Pauly, 1995; Papworth, 2007).

Indeed, fishers may have experienced and understood behavioural changes and increased fishing power, which may have obscured declines from fishery statistics. However, various psychological factors affect memory and recall, thus may affect fishers’ memory-based estimates of trends (see Daw et al. 2011). Nevertheless, LEK is relatively inexpensive to obtain and particularly beneficial when degradation may be rapid but time and resources for scientific understanding are limited (Johannes et al., 2000). LEK is also perhaps most useful in understanding the non-commercially targeted species for which there are no alternative catch or relative abundance data available, since fishers are known to offer an unbiased view regardless of whether or not they depend on the stock for their livelihood (Ainsworth, 2011).

40

Chapter 2

2.3.2 Formalisation of LEK with geographical information system (GIS) Acknowledging the limitations imposed by LEK (see Rahman, 2000), some researchers argued that LEK needed to be formalised (Huntington, 2000; Hall and Close, 2007; Ainsworth et al., 2008; Ainsworth, 2011), since it is essentially of a fragmentary and provisional nature. Yet, relatively little published research has discussed the use of LEK data collection and analysis methods for inclusion in fisheries management. Therefore, although fishers’ verbal accounts can be used to reconstruct previous ecosystem trends, some effort is required first to convert anecdotal information in a systematic manner to produce data that is acceptable to fisheries managers and scientists. One approach is the use of spatial information technology, specifically geographical information systems (GIS), as a medium to integrate LEK with scientific knowledge systems.

Fishers, because they are on the water most days of the week, experience the marine ecosystem (i.e. climate patterns, fish migration patterns, species’ behaviour, etc.) first hand that may not be fully represented during the times when a scientific study takes place (Johannes, 1989). Unless captured over substantial time periods, fishers, therefore, tend to have better local and temporal knowledge than scientific data gathering can capture. Fishers tend to perceive the environment as a non-linear representation of space, often orientating themselves based on places, such as how far a fishing spot is from a particular island (Brodnig and Mayer-Schönberger, 2000). These types of spatial interactions represent features at a finer, more localised scale than other types of information. In this context, LEK has the potential to be very effective in stock assessments, and if collected over a multi-year period, can even illustrate a temporal representation of the fish stocks. 41

Chapter 2

GIS, on the other hand, has the capability of storing, analysing and manipulating information in various ways, thus serves as a tool to store LEK and to visualise LEK and scientific knowledge in the form of maps. Although GIS in fisheries science have been slow to evolve relative to terrestrial applications, largely due to the fluid nature of the aquatic systems (Nishida et al., 2001), by having LEK and scientific knowledge unified in a GIS environment, the integration of these data can provide a rich and holistic knowledge base for planning and management of fisheries and marine resources.

2.3.3 Application of fuzzy logic concept in quantification of LEK Beside systematic integration with scientific-base knowledge, formalisation of LEK also has to deal with the inherent issue of vagueness in data collection. Depending on age, experience and expectations, one fisher’s ‘high’ could be another’s ‘low’ (Ainsworth et al., 2008). For instance, an experienced fisher who knows the best places to fish for certain species may report high abundance for that species, while a younger fisher who lacks experience may be less successful, hence reporting its abundance as low. Consequently, LEK does not lend itself well to mathematical representation, and hence for most part it has remained absent from stock assessment or during development of management plans.

To address this problem, several researchers turned to ‘fuzzy logic’ in quantifying LEK responses for use in fisheries research (Mackinson, 2001; Ainsworth et al., 2008; Ainsworth, 2011). Fuzzy logic offers a standardisation approach which has been demonstrated as appropriate to address the uncertainty and subjectivity in complex marine environmental problems (Cheung et al., 2005; Ainsworth et al.,

42

Chapter 2

2008; Sylaios et al., 2010). In comparison to classical logic, where input “linguistic variables” need to be arranged into crisp sets (e.g. a Boolean control variable belonging exclusively to either ‘true’ or ‘false’ categories), fuzzy logic enable a gradation of truth (or false) for each input variables. Originally developed by Zadeh (1965) for artificial intelligence and control systems, the fuzzy logic approach made possible for one fisher’s ‘high’ to be treated the same as another fisher’s ‘low’, by emulating an expert’s judgement and combining inputs through a heuristic IF-THEN rule matrix to reach conclusion regarding the data. Depending on the set of linguistic input variables, relevant heuristic values will fire in the fuzzy expert system with certain strengths that reflect the certainty regarding the system condition. Each control rule leads to a conclusion about the system status, and after all relevant control rules fire, the resulting range of possible conclusions is then reduced to a single point output through a “defuzzification” process (see Ainsworth et al. 2008 and Ainsworth, 2011). Since all knowledge, whether scientific or LEK, can be incorporated into the system, the potential of all data sources is also maximized (Mackinson and Nottestad, 1998).

2.4 Marine environmental variability and fishery resources The basis of stock assessment and fisheries management have long evolved around the response of fish populations to harvesting, while the influence of climate variability on the production of marine ecosystems remains poorly understood (Fogarty and Powell, 2002). Indeed, a reliable correlation between long-term stock fluctuations and global climate characteristics has been ignored until recently, especially with the advancement in remote sensing technology capabilities (e.g. Solanki et al., 2003; Klyashtorin, 2001; Qiu et al., 2010). Some studies found that 43

Chapter 2

climate variability affects primary through to tertiary productivity by controlling nutrient supply (Lehodey et al., 2003; Qiu et al., 2008). Others suggest that the physical environmental conditions directly influence larval survival and recruitment (Friedland et al., 2003; Ottersen et al., 2006). Regardless of the source, most agreed that identification of the physical forcing factors and understanding their driving mechanisms greatly benefit stock assessments.

Beaugrand and Kirby (2010) suggest that one likely reason for the causes and mechanisms responsible for variability in fish stock biomass to remain poorly identified may be due to interactions between fishing and the environment (Brander, 2007). For instance, fishing has been observed to reduce the spawning stock biomass and skews the age distribution towards younger ages and earlier maturity at a smaller size (Heath and Brander, 2001), and this can amplify the response of the populations to variability in environmental conditions (Ottersen et al., 1994; Montevecchi and Myers, 1996; Hsieh et al., 2006).

Quantifiable statistical studies linking climate and its variation to commercial fishery catches were first made by Hjort (1926) who demonstrated that environmental conditions, and not migrations, were responsible for the variability in Norwegian herring and cod catches. Much of the subsequent research then focused on the link between climate variability and the responses from large pelagic fisheries such as Peruvian anchovy, Icelandic herring and Japanese sardine (Klyashtorin, 2001). Not only are these pelagic fishes are dominant in marine ecosystems and hence more commercially important globally, but an even more interesting phenomenon is the regular synchronous outbursts of sardine and anchovy catches in different regions of

44

Chapter 2

the world, further suggesting that the fish populations are governed by the same global climatic events (Kawasaki, 1992). At such, the pattern of variation in the demersal fish community structure and its relations with environmental factors have received little attention in comparison with pelagic species (Tian et al., 2011), presumably since changes in demersal fisheries tend to take place over a relatively longer period of time, as well as the lack of large interannual variation in recruitment for the demersal stocks (Myers et al., 1995). Nevertheless, demersal stocks, especially those that form large spawning aggregations such as groupers and snappers, are extremely vulnerable to exploitation (Dulvy et al., 2004). Given the increased knowledge of the linkages between resources and marine environmental parameters, disentangling the influence of fishing and environment therefore becomes essential for effective fisheries management, particularly at a time of rapid global climate change (IPCC, 2007).

2.4.1 Impact of climate change on tropical marine ecosystem and fishery Empirical evidence for climate change effects on marine ecosystems and their component is growing (Blanchard et al. (2012) and references therein). As the many impacts of climate change on marine ecosystems may be additive, synergistic or antagonistic, Intergovernmental Panel on Climate Change (IPCC) in their recent Fourth Assessment Report (2007) concluded that climate change will affect the production of fish and the fisheries they support, and will likely confound the impacts of natural variation on fishing activities, thus further complicating management efforts.

45

Chapter 2

Climate change threats in the tropics are increasing, with coral reefs predicted to be the first major ecosystem to suffer extensive damage, especially from increasing sea surface temperature and elevated concentrations of CO2 (IPCC, 2007). Although there are no parallel climate change threats to mangrove forests and seagrass beds, coastal ecosystems can still be severely damaged by the predicted increase in the incidence of severe tropical storms (IPCC, 2007). Munday et al. (2012) gave an excellent review on impacts and adaptation responses of tropical coastal fish to climate change. Changes to sea surface temperature (SST), ocean pH, and circulation patterns are expected to influence a suite of biological and ecological characteristics of marine fishes, which include physiological condition, life history traits, timing of breeding, reproductive output, larval development, population connectivity and geographical distributions (see Table 1 from Munday et al., 2012).

Since capture fisheries depend on the productivity of the natural ecosystems on which they are based, consequently, climate change impacts will likely exacerbate existing stresses on fish stocks such as overfishing, diminishing wetlands and nursery areas, and pollution (IPCC, 2007). The long-term consequences of climate change to capture fisheries remain highly uncertain (Perry 2011). Climate change can likely cause collapses of some fisheries and expansion of others, as the level of impact vary widely and will depend on the complexity of each ecosystem, the attributes and adaptability of each species, and the nature of human communities that depend on them (IPCC-SAR, 1996). As a result, observational changes in marine environments associated with climate change need to be considered against the background of natural variation on a variety of spatial and temporal scales (IPCC, 2007; Perry 2011). Given that distinguishing the effects of climate change embedded in natural modes of

46

Chapter 2

variability such as ENSO, although challenging (IPCC, 2007), is essential in addressing the uncertainties of climate change on fisheries, these situations further reiterate the importance of being able to discern anthropogenic and natural causes on fishery stock abundance, especially if fishery-dependent data is involved.

2.5 Earlier studies on fishery and stock assessments in Brunei Darussalam Fish stock surveys on the continental shelf of Brunei waters began in the 1950s, but without any significant effect on fisheries development and management in Brunei as these surveys were not conducted by the Brunei government nor any agencies associated with it. Instead, these surveys were carried out under the British Colonial Development and Welfare Scheme based in Singapore, to investigate the fishery potential in Malaya and Borneo waters (Beales, 1982). Brunei’s DOF, as it exists today, was established in 1966, and its first official survey, although limited in scope, was carried out in 1968. Descriptions of these surveys are summarised by Beales (1982), in his seminal report on the “Investigations into Fisheries Resources in Brunei”, which proved to be a watershed in the sense that the results from his report were used as a basis for large-scale fishery development in Brunei. These results were based on the exploratory 1979-1980 trawl survey conducted using DOF’s first research and training vessel K/P Lumba-lumba (=R/V Dolphin).

The next comprehensive study was then conducted in 1989-1990 as part of the ASEAN/US Coastal Resources Management Project implemented in the country (see DOF-MIPR 1992; Silvestre et al. 1993). Demersal trawl and pelagic acoustic surveys were the two techniques employed. However, the analysis of these surveys has been limited to estimating biomass and potential yield (Silvestre and Garces, 2004), 47

Chapter 2

presumably because one of the objectives of the programme was to build up local human capacity in fisheries science and management (DOF-pers.comm.). Following Brunei’s membership in SEAFDEC in 1995, periodic surveys and stock assessments were carried out in Brunei waters, as part of the collaborative research with regional and international organisations, often investigating on South China Sea on a whole (e.g. Ranimah and Cinco, 2006; Syah, 2007; Staples, 2009). As a result, these researches tend to generalise and apply the result of low density sampling to large areas.

The DOF also made an effort to run routine survey programmes for stock assessments, which is mainly made through FAO’s operation of FiSAT. Unfortunately, the results of these surveys were never published other than as an internal government report, although a number of policy recommendations were made based on these results to assist in the management of fisheries at national level (DOF-pers.comm.). In fact, like most other countries in the region, the information content of the data collected from the previous surveys in Brunei waters remains to be fully extracted and utilised (Silvestre and Pauly, 1997b; Silvestre and Garces, 2004). This is because many of the survey data remain in the files of various fisheries agencies and are not being systematically collated and analysed using appropriate techniques – rendering these fisheries to be ‘data-poor’.

Results from previous studies on the demersal stocks of Brunei waters revealed low exploitation rates and a relatively healthy stock biomass, at least up till the year 2000, as verified by the trawl survey carried out by DOF in 2000 (DOF-pers.comm.). This was further confirmed in an independent study by Chirstensen et al. (2003) who

48

Chapter 2

had shown that the predicted fish biomass of trophic level ≥ 3.0 in Brunei in 2000 was roughly 80% of its predicted biomass in 1960. When DOF carried out another demersal trawl survey in 2009, it was obvious that the demersal resources had declined (i.e. mean trawl catch rate significantly lower in 2009; DOF-pers.comm.), but the causes remain unclear. The data which DOF attempted to collect annually by trawl surveys since 2000 were problematic and deemed unreliable to infer trends and other information (DOF-pers.comm.). For instance, other than issues associated with sample sizes and replicates, there were data “gaps” which were often caused by factors beyond DOF’s control (e.g. mechanical problems on research vessel, bad weather condition, withdrawal of permission to carry out survey due to political situations, etc.).

In the face of increased fishing effort, changing climate and environmental conditions, and the boom and bust of some species, it then became apparent that further research is in order to complement DOF’s monitoring programme, and thus provide the stakeholders of Brunei fisheries with more recent and sound scientific outcome to base their decisions in future fishery development and management. As a type of natural resource exploitation, many acknowledged fishery as not only a biological system, but also socio-economic and political (Pauly and Zeller, 2003). Hence, to effectively develop and manage such complicated system, the assessment of the status of the resources will be a critical starting point.

49

Chapter 3 Overview of Brunei marine ecosystem in the past decade via trophodynamic indicators and community structure analyses 1 “… the ocean is a restless and changing environment; its changes may either be sudden and dramatic, or covert and sustained for very long periods…” – Alan Longhurst and others, 1972

Chapter Summary Large-scale fishery catches in Brunei Darussalam between 2000 and 2009 were used to examine changes in community structure of Brunei’s marine ecosystem. This study found that Mean Trophic Level (MTL) has declined at a rate of 0.08 trophic levels (TL) per decade, suggesting a ‘fishing down the marine food web’ process in Brunei waters. In order to focus on changes in relative abundance of the more threatened, higher TL finfish only, MTL was recalculated as 3.30MTL (to exclude catches of finfish and invertebrates species with TL less than 3.30). Here, no overall trend in the 3.30MTL was found over the study period. A more in-depth analysis suggested that MTL fails to capture substantial changes in underlying species assemblage. Over the course of the study period, marked changes in fish community structure were observed. Analysis of official statistics showed an increase in total catch was driven primarily by an increase in pelagic catches, with a similar trend observed when the catches were aggregated into either high- or low-TL. Additionally, the increasing trend of Fishing-in-Balance (FiB) index for Brunei waters, as well as shifts in trophic spectra (TS) patterns suggest expansion of the fishery effort offshore. Whilst this study was not aimed at demonstrating overfishing as the cause of observed trends, there is a danger that current best practice fails to detect what may be an imminent collapse of the demersal stock. In many fisheries around the world, there is a need for detection methods able to identify potential problems within a time scale short enough to be able to react effectively.

1

Syazana Ebil, Charles R.C. Sheppard, Ranimah Wahab, Andrew R.G. Price, James C. Bull (2013) Changes in community structure of finfish catches in Brunei Darussalam between 2000 and 2009, Ocean & Coastal Management, 76, 45-51.

Chapter 3

3.1 Introduction In response to the increasing recognition of the declining state of fisheries and marine ecosystems, there is a compelling need for conservation of ecosystem services, beyond the sustainability of fishery-targeted stocks. This is recognised and widely accepted by the international community, leading to the recent development of several ecosystem-based indicators. This chapter is dedicated to the application of the trophodynamic indicators and community structure analyses to assess the overall state of Brunei marine ecosystem in the past decade.

At present, there are no published analyses of fisheries shifts and their ecosystem effects in Brunei. By synthesising several types of analyses, changes over time in species composition, mean trophic level (MTL), Fishing-in-Balance (FiB) and trophic spectra (TS) of catches were explored. Consequently, this chapter shows that even a well-respected metric such as MTL may fail to capture critical aspects of community architecture. This highlights the importance of documenting and disentangling the variety of overlapping and confounding factors, that potentially affects fishery dynamics, in understanding and interpreting the message conveyed by TL-based metrics. Specifically, this chapter aims to address the following research questions:

1)

Do Brunei catch MTL and FiB indices decline with time?

2)

Do catches of invertebrates and low-TL species affect changes in MTL?

3)

Is ‘fishing down the food web’ (FDFW) prevalent in Brunei waters (i.e. decrease in MTL due to decline in abundance of high-TL species)?

4)

Which ecological domains (or functional group) are mostly affected? 51

Chapter 3

3.2 Materials and methods 3.2.1 Study area The study area (Figure 3.1) corresponds to the continental shelf and continental slope of Brunei waters (within 1000-m isobath), excluding the small-scale (SS)-only fishery zone (Brunei Bay and waters 0 – 3 nautical mile from shore). This study area is known active fishing grounds of pelagic and demersal resources for the large-scale (LS) fishery.

FIGURE 3.1: Study area corresponds to Zone 2 and 3 of Brunei fishing zonation (hatched lines). Shaded area (in yellow) indicates oil industry-related or coral reefs areas.

3.2.2 Catch data and trophic level estimates This study is based on the monthly catch data recorded for the LS sector instituted by the Department of Fisheries, Brunei (DOF). In these fisheries, collection of catch data has been underway since 1984 when LS fishing operation officially started. However, compilation of fisheries data digitally was introduced only in 2000 and is on-going. Thus, this study is based on the data set from 2000 to 2009.

52

Chapter 3

Catches and TL of 88 known species (or species groups) were used to calculate the trophodynamic indicators (Table 3.1). Despite comprising only 55 - 60% of total catches, these selected species were of high commercial importance and benefit from a fine level of taxonomic resolution associated with its catch data. The rest of the total catches comprise of ‘mix’ categories, all representing unidentified fishes, and hence were not included in the analysis due to the impossibility of assigning them a precise TL.

Estimates of TL of each species (TLi) was taken from FishBase (v. 08/2011), which calculates TL from stomach contents data (Froese and Pauly, 2000). When estimates were not available for a species, a closely related species (same genus) was used. When the catch data represented a group of species (e.g. Lutjanidae spp.), an average of the TL of all the species (ever recorded for Brunei waters) within the group would be taken.

53

Chapter 3

TABLE 3.1: Trophic level (TL) of the main species landed from Zone 2 and 3 of Brunei fishing zonation, from 2000 to 2009. Groups(a) DEM

Family name

Scientific name/group

Common name

Local name

Est. TL

Ariidae

Arius thalassinus

giant catfish

gagok

3.10

Balistidae

Abalistes stellaris

starry triggerfish

ayam laut

3.54

Caesionidae

Caesio spp.

fusiliers

sulit

3.40

Carangidae

Alectis indica

Indian threadfish

taweh

4.09

Carangidae

Caranx ignobilis

giant trevally

bamasa

4.22

Carangidae

Caranx spp.

jacks

ikan putih

4.24

Carangidae

Caranx tille

tille trevally

langguran

4.06

Carangidae

Gnathanodon speciosus

golden trevally

bebatik

3.84

Carangidae

Parastromateus niger

black pomfret

duai hitam

2.93

Carangidae

Selaroides leptolepis

yellowstripe trevally

3.53

Drepaneidae

Drepane punctata

spotted sicklefish

temanong besurat saphee

Ephippidae

Ephippus orbis

batfish

awat-awat

3.99

Gerreidae

Gerres filamentosus

whipfin mojarra

kapas-kapas

3.26

Haemulidae

Diagramma picta

painted sweetlips

lapih

3.46

Haemulidae

Pomadasys argenteus

head grunt

garut-garut

3.42

Haemulidae

Pomadasys hasta

silver grunt

umpak

3.55

Lactariidae

Lactarius lactarius

false trevally

kelapa-kelapa

3.60

Leiognathidae

Leiognathus equulus

common ponyfish

pulut-pulut

3.01

Leiognathidae

Leiognathus spp.

ponyfish

bilis

3.02

Lethrinidae

Lethrinus lentjan

pink ear emperor

anduping

3.65

Lutjanidae

Lutjanus argentimaculatus

ungah

3.80

Lutjanidae

Lutjanus johnii

mangrove red snapper John's snapper

beberahan

4.18

Lutjanidae

Lutjanus lutjanus

bigeye snapper

pisang-pisang

4.05

Lutjanidae

Lutjanus malabaricus

Malabar red snapper

membangan

4.48

Lutjanidae

Lutjanus rivulatus

blubberlip snapper

ketumbang

4.13

Lutjanidae

Lutjanus sebae

emperor red snapper

santak

4.27

Lutjanidae

Lutjanus spp.

red snapper

ikan merah

4.46

Lutjanidae

Pinjalo pinjalo

pinjalo snapper

sulit merah

3.80

Lutjanidae

Pristipomoides multidens

goldband snapper

kerosi bali

3.84

Mullidae

Upeneus sulphureus

sulphur goatfish

bantang

3.16

Muraenesocidae

Muraenesox bagio

common pike eel

tingkor-tingkor

3.99

Nemipteridae

Nemipterus spp.

threadfin breams

kerisi

3.60

Polynemidae

Leptomelanosoma indicum

Indian threadfin

kurau

3.60

Priacanthidae

Priacanthus tayenus

spotted fin bigeye

3.58

Psettodidae

Psettodes erumei

Indian halibut

semperiding takat pila-pila

Sciaenidae

Johnius coitor

coiter croaker

gelama

3.25

Sciaenidae

Otolithes ruber

tiger-toothed croaker

jarang gigi

3.60

Sciaenidae

Pterotolithus maculatus

blotched tigertoothed croaker

keropok

3.73

3.32

4.39

54

Chapter 3

TABLE 3.1 (Continued) Groups(a)

PEL

Family name

Scientific name/group

Common name

Local name

Est. TL

Serranidae

Epinephelus bleekeri

duskytail grouper

kerapu hitam

3.90

Serranidae

Epinephelus spp.

grouper

kerapu

4.15

Serranidae

Plectropomus leopardus

leopard coral grouper

penghantaran

4.49

Serranidae

Plectropomus maculatus

spotted coral grouper

kerapu merah

4.11

Siganidae

Siganus spp.

rabbitfish

belais

2.87

Sillaginidae

Sillago sihama

silver sillago

usus

3.37

Stromateidae

Pampus argenteus

silver pomfret

duai putih

3.12

Synodontidae

Saurida tumbil

greater lizardfish

pangual badok

4.40

Trichiuridae

Trichiurus lepturus

largehead hairtail

ikan timah

4.45

Carangidae

Alepes kleinii

razorbelly scad

pelata

3.54

Carangidae

Atule mate

yellowtail scad

temanong

4.45

Carangidae

Decapterus maruadsi

Japanese scad

basong-basong

3.40

Carangidae

Elagatis bipinnulata

rainbow runner

salman

3.59

Carangidae

Megalaspis cordyla

torpedo scad

geronggong

4.39

Carangidae

talang queenfish

bekalang

4.48

Carangidae

Scomberoides commersonnianus Selar crumenophthalmus

bigeye scad

tulai

4.10

Clupeidae

Amblygaster sirim

spotted sardinella

tamban bagol

3.30

Clupeidae

Anodontostoma chacunda

gizzard shad

kuasi

2.83

Clupeidae

Sardinella fimbriata

fringescale sardinella

aur-aur

2.70

Clupeidae

Sardinella gibbosa

goldstripe sardinella

tamban

2.85

Clupeidae

Tenualosa macrura

longtail shad

terubok

2.27

Coryphaenidae

Coryphaena hippurus

common dolphinfish

suhong

4.37

Engraulidae

Thryssa setirostris

longjaw thryssa

kirang-kirang

3.32

Ephippidae

Platax teira

longfin batfish

buna

3.95

Hemiramphidae

Hemiramphus far

black-barred halfbeak

suroi

2.91

Istiophoridae

Istiophorus platypterus

Indo-Pacific sailfish

layaran

4.50

Lobotidae

Lobotes surinamensis

tripletail

pelayak

4.04

Megalopidae

Megalops cyprinoides

Indo-Pacific tarpon

bulan-bulan

3.30

Pristigasteridae

Ilisha megaloptera

bigeye ilisha

bilak-bilakan

3.04

Rachycentridae

Rachycentron canadum

cobia

banglus

3.96

Scombridae

Euthynnus affinis

bonito

bakulan

4.50

Scombridae

Katsuwonus pelamis

skipjack tuna

tongkol

3.75

Scombridae

Rastrelliger brachysoma

short mackarel

rumahan bini

2.72

Scombridae

Rastrelliger kanagurta

Indian mackarel

rumahan laki

3.19

Scombridae

4.50

lamading

4.28

Scombridae

Thunnus albacares

narrow-barred Spanish mackarel Indo-Pacific king mackarel yellowfin tuna

tenggiri

Scombridae

Scomberomorus commerson Scomberomorus guttatus

4.34

Scombridae

Thunnus tonggol

longtail tuna

tuna sirip kuning tuna

Sphyraenidae

Sphyraena spp.

barracuda

titir

4.34

4.50

55

Chapter 3

TABLE 3.1: (Continued) Groups(a)

Family name

Scientific name/group

Common name

Local name

Est. TL

CON

Carcharhinidae

Carcharhinus dussumieri

whitecheek shark

yu

3.90

Dasyatidae

Dasyatis spp.

stingray

pari

3.76

Rhinobatidae

Rhynchobatus djiddensis

giant guitarfish

anunan

3.60

Rhinobatidae

Rhinobatus spp.

guitarfish

paita

3.55

Loliginidae

Loligo spp.

squid

sotong

3.80

Penaeidae

Penaeus spp.

prawn

udang

2.50

Portunidae

Portunus pelagicus

blue crab

ketam

2.50

Holothurian

Holothurian spp.

sea cucumber

timun laut/balat

2.00

Palinuridae

Panulirus ornatus

spiny lobster

bakara

2.00

Scyllaridae

Thenus orientalis

flathead lobster

satak

2.00

Sepiidae

Sepia spp.

cuttlefish

kelabutan

2.00

INV

(a) Groups based on their ecological guilds; DEM – demersals, PEL – pelagic, CON – chondrichthians, INV – invertebrates.

3.2.3 Data analyses The original data set provided by the DOF was stored in Microsoft Access. Data manipulation and analyses were done in other programs, mainly Microsoft Excel and SPSS v.18.

3.2.3.1 Mean trophic level (MTL) and Fishing-in-Balance (FiB) index The MTL was estimated as follows (Pauly et al., 1998):

∑ ∑

where MTL is the mean trophic level of catch in month j, Yij the catch of species i in the month j and TLi is the trophic level of species i.

56

Chapter 3

To account for the effect of non-fish species, as well as species of low-TL, MTL was recalculated using a cut-off trophic level of 3.30 (see discussion; Section 3.4). Termed 3.30MTL, this analysis enabled emphasis on changes in the relative abundance of the more threatened, higher TL species to be explored (Pauly and Watson, 2005).

Subsequently, fishing down the marine food web (FDFW) may be the result of a deliberate choice which can affect the MTL trend but might not reflect the true trophic structure in the ecosystem (Caddy et al., 1998; Pauly and Watson, 2005). This can be explored via the FiB index as developed by Pauly et al. (2000). However, one drawback of the FiB index is its heavy reliance on the catches and their TL in the reference year (or month). Considering that trends in a FiB series are conserved irrespective of the reference year selected (Cury et al., 2005), absolute FiB was calculated for Brunei instead:

( (

)

)

where Yi is the catch at month I, TLi is the mean trophic level of the catch at month i, and TE is the trophic efficiency (here set at 0.10; see Pauly et al., 2000). Consequently, FiB trends will remain constant (slope equals zero) if TL changes are matched by ‘ecologically correct’ changes in catch (Pauly et al., 2000).

3.2.3.2 Trophic spectra (TS) The catch TS were plotted from annual catch data, following the technique described by Gascuel et al. (2005). First, species catches were distributed by trophic class of 57

Chapter 3

0.1 increments according to their trophic level. A smoothed spectrum was then obtained with a three-increment weighted moving average. This second step was required to account for the intraspecific variability (Gascuel et al., 2005) and the uncertainty inherent in TL estimation (Pinnegar et al., 2002).

3.2.3.3 Statistical analyses Monthly MTL and FiB values were subjected to standard time series decomposition to control for the impacts of seasonal factors. Simple linear regression analysis was then used to determine the long term trend between catches, logit-transformed catch proportions and TS, and MTL and FiB indices against time. In monthly MTL and FiB analyses, if the long term trend were found to be non-significant, smoothing splines were used to illustrate shorter-term changes through time. The slope from the regression analysis was also used to calculate the rate of change in MTL per decade (∆MTL) (Pauly et al., 1998; Milessi et al., 2005; Bhathal and Pauly, 2008; Freire and Pauly, 2010).

58

Chapter 3

3.3 Results 3.3.1 Exploratory analysis: Trends in annual catch and effort Annually, the total catch from Brunei waters has shown a significant increase over the years investigated (Figure 3.2a), driven primarily by the significant increase in the catches of pelagics, with a peak of ca. 1360 MT recorded in 2005 (Figure 3.2c). The rest of the catch, which comprised of the functional groups of demersals, chondrichthians and invertebrates (i.e. crustaceans and molluscs) did not show any significant annual trends (Figure 3.2, Table 3.2). Trends in annual effort (taken as maximum number of fishing vessels and total number of days out in the sea – see Chapter 4 for further discussion) varied amongst different gear categories (Figure 3.3, Table 3.3), whereby a pronounced increase in the purse seine fleet was observed.

59

Chapter 3

(a) overall 2200 1800 1400 1000 2000

CATCH (MT)

900

2003

(b) demersals

2006 1500

600

1000

300

500

0

(c) pelagics

0 2000

600

2009

2003

2006

2009

(d) invertebrates

2000

300

400

200

200

100

0

2003

2006

2009

2006

2009

(e) chondrichthians

0 2000

2003

2006

2009

2000

2003

YEAR

FIGURE 3.2: Annual fishery catches (in metric tons) of Brunei waters between 2000 – 2009 based on functional groupings: (a) overall; (b) demersal fishes; (c) pelagics fishes; (d) invertebrates (crustaceans and molluscs); (e) chondrichthians (sharks and rays). Nonsignificant trend is shown with a dashed line. Note the different scales on the y-axis.

TABLE 3.2: Results of linear regression analyses of catches of different groups as a function of years in Brunei waters. Significant p-values are in bold. Annual change Factor R2 Slope b t-stats p-value (catches in MT) (SE) Overall 0.46 78.7 (30.2) 2.61 0.031 Demersal fishes 0.15 -19.0 (16.3) -1.17 0.276 Pelagic fishes 0.67 106.6 (26.5) 4.03 <0.001 Invertebrates 0.06 -6.7 (9.1) -0.74 0.480 Chondrichthians 0.02 -2.0 (5.7) -0.36 0.731 *Degrees of freedom = 9 for all analyses

60

Chapter 3

Total number of days at sea 6500

Maximum number of 'active' vessels

(a) Trawlers

25

5500 20 4500

3500

15 10

1200

8 6

800

4 400

2

0 1200

EFFORT (no. of vessels)

EFFORT (no. days at sea)

(b) Purse seiners

0 (c) Long liners

12

800

8

400

4

0

0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 YEAR

FIGURE 3.3: Annual trends in effort for the three LS fishery fleet: (a) Trawlers, (b) Purse seiners, and (c) Long liners. Non-significant trend is shown with a dashed line. Note the different scales on the y-axis.

TABLE 3.3: Results of linear regression analyses of efforts of different fishing fleet as a function of years in Brunei waters. Significant p-values are in bold. Annual change Factor R2 Slope b t-stats p-value (SE) Effort: Maximum number of ‘active’ vessels Trawls 0.56 -0.63 (0.2) Purse seines 0.78 0.82 (0.2) Long lines 0.00 0.01 (0.2)

-3.20 5.37 0.04

0.013 0.001 0.969

Effort: Total number of days at sea Trawls 0.06 Purse seines 0.74 Long lines 0.46

-0.71 4.79 -2.62

0.496 0.001 0.031

-29.98 (42.0) 109.10 (22.8) -43.71 (16.7)

61

Chapter 3

3.3.2 Trends of trophic-based indicators After correcting for seasonal factors, the MTL of the total catch showed significant temporal decline (Figure 3.4a, Table 3.4), with an apparent drop in MTL values between 2006 and 2007. This resulted in ∆MTL = 0.08 TL per decade. In contrast, 3.30

MTL values showed no significant long-term trend throughout the study period

(Figure 3.4b, Table 3.4), although two distinct perturbations could be identified from the smoothed 3.30MTL trajectory. From 2002 to 2004, 3.30MTL values rose gradually, coinciding with the change in gear legislation in 2002. However, only two years after the restriction was introduced, the 3.30MTL values started to decline again, although a second smaller peak was detected in 2007. Finally, the absolute values of FiB index showed a gradual increase throughout the study period (Figure 3.5, slope (SE) = 0.001 (0.0004), t (df = 119) = 2.11, p = 0.036).

TABLE 3.4: Results of linear regression analyses of different type of MTL as a function of time in Brunei waters. Δ is the magnitude of decadal change. Significant p-values are in bold. MTL

Δ

MTL MTL

0.08 0.11

3.30

R2 0.05 0.11

Monthly change Slope b (SE) -0.0007 (0.0003) -0.0010 (0.0003)

t-stats

p-value

-2.445 -3.727

0.016 0.083

*Degrees of freedom = 119 for all analyses.

3.8

(a) MTL

4.2

3.30

MTL

Change in trawl gear legislation

3.6

MTL

(b)

4

3.4 3.8

3.2

3.6

3 2000

2003

2006

2000

2009

2003

2006

2009

YEAR

FIGURE 3.4: Seasonally-adjusted mean trophic level (MTL) of fishery catches in Brunei during the period 2000 – 2009. MTL of (a) total catches, and (b) catches excluding species with TL < 3.30 (3.30MTL). Non-significant trend is shown in dashed line.

62

Chapter 3

absolute FiB

6.4 6

5.6 5.2 4.8 2000

2003

YEAR 2006

2009

abs

FIGURE 3.5: Absolute Fishing-in-balance ( FiB) index of fishery catches in Brunei (20002009).

3.3.3 Trends in species composition of catch The varying trends in catches of different families within the different ecological groups resulted in a shift in species composition of catches over the years (Figure 3.6a). At the beginning of the time series, total catches were dominated by demersal fishes (mainly snappers and false trevallies) and penaeid shrimp, as these groups comprise about 60% of the total catches. However, in 2009, the proportion of catch contributed by these groups was less than 30%. The proportion of demersal fishes in catches had declined over the years (Figure 3.6c; logit-transformed values of demersal fishes proportions against time – slope (SE) = -0.12 (0.05), t (df = 9) = -2.72, p = 0.03), while on the other hand, the proportion of pelagic fishes such as tunas, sardines and small mackerels had increased with time (Figure 3.6b; logittransformed values of pelagic fishes proportions against time – slope (SE) = 0.21 (0.05), t (df=9) = 4.30, p < 0.01), from just over 10% in 2000 to more than 50% at the end of the time series.

63

Chapter 3

(a)

PROPORTION OF CATCH (%)

80%

Proportion of pelagic fishes (logit-transformed)

90%

70%

1

other pelagic fishes small mackerels sharks and rays

60%

sardines other invertebrates

50%

tunas

40%

false trevallies

20% 10% 0% 2000

snappers and groupers

other demersal fishes 2001

2002

2003

0.5 0 -0.5 -1 -1.5

0

penaeid shrimps

30%

(b)

R² = 0.70

-2

Proportion of demersal fishes (logit-transformed)

100%

2004

2005 YEAR

2006

2007

2008

2009

(c) R² = 0.48

-0.5 -1 -1.5 -2 -2.5 2000

2003

YEAR

2006

2009

FIGURE 3.6: Changes in composition of species caught in Brunei waters between 2000 and 2009. (a) Composition of overall catches, aggregated based on functional groupings (from top) – pelagic fishes, chondrichthians, invertebrates and demersal fishes – which are further subdivided considering the species caught at higher proportion; and proportion of (b) pelagic fishes, and (c) demersal fishes, caught throughout the study period.

64

Chapter 3

(a)

(b)

1400

100%

1200

Invertebrates

Chondrichtians

Pelagics

TL > 3.5

PROPORTION OF CATCH (%)

TL > 3.5

1000 CATCH (MT)

Demersals

800 600 TL < 3.5 400

50%

0% 100%

TL < 3.5

50%

200 0

0% 2000

2003

2006 YEAR

2009

2000

2009 YEAR

FIGURE 3.7: Analysis of Brunei fishery catches discriminated by trophic level: TL > 3.5 and TL < 3.5; (a) Total annual catch of high-TL and low-TL species for Brunei; (b) Species composition of catches for the different trophic level groups, with increasing proportion of pelagics observed toward end of time series.

65

Chapter 3

Similar trends were also observed when the catches were aggregated into either highor low-TL (i.e. species with TL > 3.5 as high-TL and TL < 3.5 as low-TL; Figure 3.7). No significant trend was detected for catches of high-TL species against time (slope (SE) = 1.96 (15.30), t (df = 9) = 0.13, p = 0.90), yet the proportion of pelagic catches had increased with time too (logit-transformed regression: slope (SE) = 0.14 (0.03), t (df = 9) = 3.96, p < 0.01), although not as pronounced as catches of low-TL species (logit-transformed regression: slope (SE) = 0.26 (0.07), t (df = 9) = 3.75, p = 0.01).

3.3.4 Catch trophic spectra analysis The catch TS indicate that Brunei’s LS fisheries catch included a wide range of TLs with multiple peaks (Figure 3.8, Appendix 3.1). Two major opposite patterns emerged from the TS over time. First, catch proportions for low-TL species of between 2.8 and 3.2 had significantly increased from 12.4% to 29.1%, which essentially composed of small pelagics such as sardines and small mackerels. Interestingly, the proportion of high-TL species with TL between 4.4 and 4.6 had significantly increased over time as well, from 4.7% to 12.4%. These are mostly composed of high-valued large pelagics such as tuna and barracuda. Second, catch proportions for species of TLs between 4.0 and 4.3 had significantly dropped from 14.9% to 4.8%. The bulk of these groups composed of sharks and the groupersnapper complex. The rest of the TLs showed non-significant trends over the years.

66

CATCH PROPORTION (%)

Chapter 3

30

YEAR 2000

25

2001 2002

20

2003 2004

15

2005

2006

10

2007 5

2008 2009

0 2.0

2.5

3.0

TROPHIC LEVEL (TL)

3.5

4.0

4.5

FIGURE 3.8: Catch trophic spectra of Brunei’s LS fishery from 2000 to 2009. Downward and upward arrows on catch TS indicate the ranges of TL classes that showed significant declines and increases, respectively, in catch proportions over the studied period. Summary statistics of trends in catch proportions is given in Appendix 3.1.

67

Chapter 3

3.4 Discussion and Conclusion Preliminary exploration of the catch data revealed that the decline in MTL of Brunei marine ecosystem was due to the addition of lower TL catches rather than a decrease of high TL ones. This implies a ‘fishing through’ process, rather than FDFW (Essington et al., 2006), and this is further suggested by a remarkably constant catch of high-TL species over the years (Figure 3.7). However, Stergiou and Tsikliras (2011) pointed out that the ‘fishing through’ mechanism is not exactly an alternative hypothesis to Pauly’s FDFW phenomenon (Pauly et al., 1998), but only appears so because of confusion. Effectively, ‘fishing through’ (or ‘fishing up’ in the case of increasing MTL trends) reflects biases in the data used and/or fishing behaviour see (Stergiou and Tsikliras, 2011).

Indeed, the decline in general MTL values in Brunei waters was accompanied by an increase in total catch, mainly due to the redirection of the LS sector to purse seine fishery in 2001 (Pg.Khairul-Rijal, 2007; JPKE, 2010) thus resulting in a significant increase in catch of low-TL pelagic fishes such as sardines and mackerels – main species targeted by the purse seine fishery. This is in agreement with majority of previous studies worldwide regarding MTL, which showed that an increase in catch is linked to higher catch of species with low TL (Christensen, 1998; Milessi et al., 2005; Arancibia and Neira, 2005; Baeta et al., 2009). Worryingly, the rate of decline in MTL for Brunei waters (ΔMTL = 0.08) is comparable to those estimated for the heavily exploited coastal waters of India (ΔMTL = 0.01 – 0.08; Bhathal and Pauly, 2005) and Gulf of Thailand (ΔMTL = 0.05 – 0.09; Pauly and Chuenpagdee, 2003), albeit lower than the coastal upwelling system such as central Chile (ΔMTL = 0.15; Arancibia and Neira, 2005), the Mediterrenean (ΔMTL = 0.15; Pinnegar et al., 2003) 68

Chapter 3

and coastal waters of Brazil (ΔMTL = 0.10 – 0.17; Freire and Pauly, 2010). Considering that Pauly et al. (2002) deemed changes in MTL of 0.05 – 0.10 as ‘extremely worrisome’, the 0.08 decrease in MTL found in this study is particularly notable, especially since Brunei is lightly fished relative to the rest of the region.

However, other than fishing, long-term environmental variations can also affect the structure and functioning of marine communities (ICES, 2000). Unfortunately, the influence of environmental factors on fishery productivity in Brunei waters is not clear from existing literature (but see Chapter 4). Generally, efforts to distinguish climate effects from fishery effects, to some extent, could be done by excluding from analysis the lower TL species whose abundance is known to vary widely in response to environmental factors (Pauly and Watson, 2005). In this study, a cut-off TL value of 3.30 was chosen to eliminate (besides invertebrates and other planktivores) the sardines, Clupeidae family (i.e. TL of spotted sardinella, Amblygaster sirm), which not only dominate the pelagics catch in Brunei, but are also said to be dependent on climatic fluctuations (Klyashtorin, 2001; Hobday et al., 2009).

Interestingly, 3.30MTL revealed no significant long-term trend. Based on available information, regulatory changes, as well as extreme climatic condition, might be possible factors that influenced the 3.30MTL trajectory. First, due to the high discarding rate of the trawl fishery in Brunei (Kelleher, 2005; Matzaini et al., 2007), the DOF implemented a new mesh size regulation on all LS trawlers in 2002, where a diamond mesh size of 38 mm was increased to a square mesh of 51 mm on the trawl cod end (Matzaini et al., 2007). Following this management intervention, 3.30

MTL is expected to rise as catches of smaller sized, lower-TL species by trawlers

69

Chapter 3

decrease, while recruitments, and hence catches, of high-TL species increase. However, the strength of the index to capture the positive effect on the community structure as induced by this management intervention thinned out when the fishery continued to expand. This expansion beyond the initial ecosystem to stocks previously unexploited or only lightly exploited is suggested by the gradual increase in absolute FiB trend (Pauly et al., 2000), and resulted in a declining 3.30MTL index between 2005 and 2006. The second smaller peak observed in 2007, on the other hand, might have been resulted from a combined ‘bottom-up’ effect (Caddy et al., 1998; Pauly, 2010) of a strong El Niño event reported for that year, in addition to the increasing impacts of eutrophication in the coastal waters of Brunei (DOF-MIPR, 1992).

The trends behind the metric and species composition analyses are further supported by the temporal variations of TS observed. The TS have been shown to provide useful explanatory evidence underlying MTL and FiB calculations and trends (Laurans et al., 2004; Gascuel et al., 2005; Munyandorero and Guenther, 2010). In this study, the increase of high-valued large pelagics such as tuna and barracuda – species known to occupy further offshore waters – towards the end of the time series as shown in the catch TS suggested that a geographical expansion of Brunei LS fishery is the most likely explanation that influenced the 3.30MTL and FiB trajectories.

The gradual increase in absolute FiB trend would also imply that the decrease in MTL in this study was not matched by ‘ecologically correct’ increase in catch (Pauly et al., 2000). Accordingly, the absence of decreasing 3.30MTL and FiB trends could

70

Chapter 3

have been used to signify a potentially sustainable fishery, but to claim so may be misleading. Closer inspection of the composition of high-TL species catches (i.e. TL > 3.5) revealed clear shifts over the years, which may be interpreted as reflecting a shift in the underlying fish community structure in the ecosystem (Myers and Worm, 2003; Tsehaye et al., 2007). The decline in catch of relatively sedentary demersal species such as snappers was followed by an increase in the migratory pelagic fishes such as tunas and barracudas, potentially suggesting a shift in ecosystem functioning. Admittedly, species composition of catches might be influenced by the fleet type, but then again shifting gears are fishers’ adaptive response to shifted resources (Cinner et al., 2011). Particularly in the case of Brunei’s LS fishery sector, it is common for companies to own several fishing vessels of different gear types. As such, a decline in effort for the trawl and long line fleet is evident from the existing data.

In fact, given that Brunei fisheries are multi-specific, where ecological compensation phenomena are expected to occur (Jackson et al., 2001; Perez-Espana et al., 2006), the fact that a decline in MTL was evident in the data could only attest to the poor and deteriorating state of the marine ecosystem of Brunei. Recent stock assessments of the demersal resource revealed that Brunei has “joined the club” of regional decline in stock abundance, and that overfishing is apparent (DOF, 2011). Fishing is known to have indirect impacts on the whole ecosystem, and results from the analyses carried out in this study had demonstrated the marked changes in the structure of fisheries catches and fish communities from Brunei waters in the past decade. The removal of large predators from marine ecosystems would instigate cascade effects on food webs (Jackson et al., 2001) and, as impacts on key elements

71

Chapter 3

of the food webs increase, the upward transfer of production becomes impaired (Pauly et al., 2000).

Admittedly, the interpretation of trophic indicators can be subjective, and inevitably, this present study carried the caveats associated with MTL studies as pointed out by several authors (see discussions; Pauly and Watson, 2005; Cury et al., 2005b; Baeta et al., 2009; Munyandorero and Guenther, 2010; Angelini and Vaz-Velho, 2011). The dynamics of fishery resources are often caused by many overlapping factors, the effects of which are confounding and not always well documented or understood. Some of these factors are further examined in subsequent chapters of this thesis.

The decline of MTL might suggest that the fishery of Brunei waters is beginning to show signs of non-sustainability. Food web collapses might not have occurred yet because the decline in MTL has not been associated with overall declining catches (Pauly et al., 2001). Furthermore, the multi-specific and high biodiversity nature of the tropical fisheries meant that a long time period may be necessary for a collapse to become noticeable in the exploited community after over-fishing occurs (PerezEspana et al., 2006). This study indicates that the FDFW process is relatively weak in Brunei waters throughout the study period because catches of high-TL species have been maintained. However, a closer look at species shifts revealed that within a short time scale of 10 years, significant changes on the structure of the fish assemblages beyond the direct effects on target species, particularly on the demersal ecological group, had occurred. Hence, discernible trends are evident even based upon a relatively short time series. The fishery of Brunei seems to have become more dependent on pelagic stocks, particularly those of low-valued low-TL species.

72

Chapter 3

Worryingly, not only are such fisheries usually only marginally profitable (Christensen, 1998), but also pelagic stocks in general are vulnerable to environmental changes, making fisheries increasingly difficult to manage. Currently, other than restricting fishing zones and gear specifications, fisheries regulations in Brunei are relatively ineffective in limiting fishing efforts, especially from the smallscale fishery sector. There are no fishing quotas and enforcement is practically ineffective.

While attempting to capture the overall picture of Brunei marine ecosystem, this study had demonstrated a mixed outcome regarding the use of catch MTL as a tool for evaluating ecosystem health. The MTL approach is relatively data non-intensive, and the general MTL index had inferred a declining trend in Brunei’s marine ecosystem state, despite the short time period. However, considering the apparent shift in community structure that had occurred throughout the study period, MTL appears to be a relatively weak measure of the ecosystem state that fails to detect what may be an imminent collapse of the demersal stock. In many fisheries around the world, there is a need for detection methods able to identify potential problems within a time scale short enough to be able to react effectively. This chapter has highlighted the deteriorating condition of Brunei coastal demersal fisheries between 2000 and 2009. How such conditions influence the abundance of the different demersal groups in the past 10 years will be examined next in Chapter 4 (Exploring human and environmental influence on recent abundance trends of selected demersal stocks in Brunei).

73

Chapter 4 Exploring human and environmental influence on recent abundance trends of selected demersal stocks in Brunei “A population [of people] can be counted; but who knows how many fishes are in the sea? And yet it appears to me a project big with possibility…” – Hjort, 1907

Chapter Summary Recent trends in abundance of overall demersal finfish resources and 21 selected finfish families from the waters of Brunei Darussalam were evaluated. A general linear model (GLM) was used to standardise catch per unit effort (CPUE) data from the large-scale fishery sector for the years 2000 – 2009. Data were stratified temporally by year and month, and spatially by area (fishery management zones). Operational (types of gears) and environmental (local-scale sea surface temperature, chlorophyll-a concentration level, land rainfall and global-scale Southern Oscillation Index) variables were also considered in the full model. Where there are sufficient data, the total variance explained by the most parsimonious models range between 32 to 98%, with fishing zones identified as the most important factor in explaining CPUE variability. While the goal of standardisation in this study is to generate an index of relative abundance, emphasis is also being placed on the potential effect of the environment on fish stock abundance, which may act at several life history stages. In general, influence of environmental variables was not significant in explaining CPUE variation, except in mojarras (Gerreidae), bigeyes (Priacanthidae) and croakers (Sciaenidae), so cross-correlation analysis was used to further explore the lagged-effects of environmental variables on the demersal stocks. Overall demersal fishery resource abundance, as well as the abundance of several demersal zoobenthic feeders, exhibited a general decline over the study period. On the other hand, the majority of the intermediate demersal predator stocks were shown to increase. Identifying the sources of bias that exist when inferring short time series of catch-effort data enables better understanding of the complex fishery system, which in turn, may assist in formulation of fisheries management plans and can serve as hypothesis-generators for future research.

Chapter 4

4.1 Introduction Growing evidence has revealed that changes in fish stocks coincide with climateinduced changes in marine ecosystems, which are also affecting trophodynamics and community structure (Beaugrand et al., 2008; Alheit, 2009). In addition to long-term changes, however, fish stock biomass can show pronounced, unexpected fluctuations in space and time over short-term time scales that makes it difficult to propose strategies for long-term fisheries management (Hsieh et al., 2006).

In Brunei, this is further complicated by the fact that the fishery resources are exploited by both the small-scale (SS) and large-scale (LS) fisheries. Like other fisheries in the region (Silvestre and Pauly, 1997b; Garces et al., 2006), the demersal fisheries of Brunei, therefore, are multi-gear and multi-species in nature (Table 4.1).

Demersal fishery resources have been classified as over-exploited in the Southeast Asia region since the 1960s (Pauly, 1988; Silvestre and Pauly, 1997a; Christensen, 1998; Christensen et al., 2003; Abu Talib et al., 2003; Lymer et al., 2010), but the presence of offshore oil rigs in Brunei which caused a significant part of its shelf being effectively closed to fishing meant that the bulk of demersal stocks off Brunei waters were lightly fished, at least up till the year 2000 (Christensen et al., 2003; Silvestre and Garces, 2004). However, the variability of climatic factors such as sea surface temperature (SST) and precipitation may also play an important role in the demersal catches in Brunei – for instance through the closely coupled interaction between SST and chlorophyll-a concentration (chl-a) in controlling nutrient supply, and therefore food availability, for the fish stocks (e.g. Solanki et al., 2003; Qiu et al., 2010; Sartimbul et al., 2010). While there are few studies in the region which 75

Chapter 4

have discussed the response of fish production to climate variability, the existence of such link in southern South China Sea in general, and in Brunei waters in particular, is not clear from the existing literature. This is especially the case for demersal stocks.

TABLE 4.1: Break down of SS and LS fisheries by gears and principal family/group of demersal fish exploited in Brunei, adapted from Silvestre and Matdanan (1992), Ranimah (2008) and personal observations. Level of Gear class Gear type2 Principal family/group of capital demersal fish occurring in Local name English input1 the catch name Net Carangidae, Haemulidae, LS Pukat tunda Trawler Lactariidae, Nemipteridae, Priacanthidae, Sciaenidae, Serranidae, Trichiuridae, Mullidae, Lutjanidae, Leiognathidae Net Pukat lingkong Purse seine Carangidae, Trichiuridae, Leiognathidae Hook and Rawai & Bubu Long line & Haemulidae, Lethrinidae, line & Trap Serranidae, Siganidae, trap Sparidae, Lutjanidae Net Ancau Seine net Leiognathidae SS (company / Net Andang Gill/trammel Carangidae, Leiognathidae individual) net Trap Haemulidae, Lethrinidae, Bubu Fish trap (portable) Serranidae, Siganidae, Sparidae, Lutjanidae Trap Tugu/Kilong Leiognathidae, Haemulidae, Tidal / (stationary) Sciaenidae, Serranidae, Kabat/Lintau Palisade Lutjanidae trap Hook and Haemulidae, Lethrinidae, Jaul/Rawai Hand line/ line Serranidae, Siganidae, Long line Sparidae, Lutjanidae 1. Based on classification by Ranimah (2008). 2. Gears in bold specifically target demersal stocks as main catch, while any demersal fish found in the other gears occurred as incidental catch.

As a follow-up to the evaluation of the state of Brunei marine ecosystem in general (Chapter 3), this chapter draws on official fishery-dependent data to further analyse the state of several commercially-important demersal fish stocks of Brunei waters in the last decade. This involved investigating the joint responses of fish production to fishing and climate variability. While taking the limitations in the use of fishery76

Chapter 4

dependent data into account, the research questions considered in this chapter are the following: 1)

What are the trends in catch and abundance (standardised CPUE for vessel-related and seasonal effects) of any of the top 20 commerciallyimportant demersal families stocks over the years? What is their current state (in 2009)?

2)

How do non-biological factors (i.e. vessel-related effects, seasonal effects, etc.) affect the index of abundance (i.e. CPUE, or catch per unit effort) calculated for the demersal stocks?

3)

How have physical components in Brunei marine ecosystems varied in the past decade? How did demersal fish stocks respond to short-term (seasonal) and long-term changes in environmental factors?

4.2 Materials and Methods 4.2.1 Study area The study area is similar to that described in Chapter 3 (Section 3.2.1).

4.2.2 Official fishery-dependent data and data selection This study is based on the monthly catch and effort data collected from the LS fishery sector, covering the period 2000 – 2009, provided by the Department of Fishery, Brunei (DOF).

The fishery-dependent data collected included, among other things, fishing vessel’s name, gross tonnage or horse power of vessel, gear type, general fishing ground, operating fishing days, and catch by species or species group and by marketing grades. Although designed for recording catches at the species level, there were some 77

Chapter 4

species, particularly those with a relatively low catch, that were grouped together, either according to family or market grades, depending on the expertise of the enumerators dispatched to the fishing port. Hence, due to such variability, some parts of the dataset are better suited to analysis at higher taxonomic levels.

In this study, the emphasis is on the demersal finfish resources (i.e. true fishes, thus excluding other aquatic animals such as whales, crustaceans and molluscs, but including sharks and rays) (FishBase v. 08/2011) of Brunei. The catch data recorded by the DOF under the ‘mix’ categories contain an unknown proportion of demersal finfish catches. For the purpose of finding the overall demersal finfish catch for each gear in this study, it is assumed that 50% of the ‘mix’ category catch is attributed to the demersal category throughout the study period (see Pauly, 1979). However, for a better understanding of the demersal stocks abundance in greater detail, only catch data of species with associated biological information available, and contained records in the dataset for at least 7 years, were considered.

As a result, data of 20 fish families (out of 25 demersal fish families recorded) and one family of rays from the LS fishery sector dataset were deemed most suitable for further investigation on the state of demersal stocks of Brunei waters (henceforth, termed ‘demersal resources’ throughout the thesis). The catch of these families makes up about 85 – 99% of the total known demersal finfish catch by weight.

78

Chapter 4

4.2.3 Environmental data collation Environmental variables including sea surface temperature (SST), rainfall and chlorophyll-a concentration (chl-a), were selected from a number of sources. Collation and preparation of environmental data prior to analysis is described below:

4.2.3.1 Sea-surface temperature (SST) data As there was no comprehensive in-situ recording of SSTs for Brunei waters, monthly average SST data for 2000 – 2009 were obtained from the HadISST1 dataset, downloaded from the Hadley Centre for Climate Prediction and Research website1. HadISST1 dataset is a monthly SST and sea ice concentration data for 1° x 1° latitude and longitude cells from 1871 to present (Rayner et al., 2003). Including the exclusive economic zone (EEZ), the SST data for Brunei encompasses 8 cells which were extracted using a Microsoft Excel spreadsheet, executed in Microsoft Office applications using the built-in code written in VBA (Visual Basic for Applications) macro.

4.2.3.2 Chlorophyll-a concentration (chl-a) data Monthly chl-a data from 2000 to 2009 were generated from the Sea-viewing Wide Field-of-View Sensor (SeaWiFs) on Orbview-2 satellite, downloaded from the NASA Goddard Space Flight Centre (GSFC) website2. The data were at 0.1° x 0.1° latitude and longitude resolution (O'Reilly et al., 2000). Since the resolution for chl-a data were higher compared to the SST data, the average of chl-a level for the main fishing area was used. Data extraction was also executed in Microsoft Office applications using VBA macro code.

1 2

http://www.metoffice.gov.uk/hadobs/hadisst/ http://oceancolor.gsfc.nasa.gov/SeaWiFS/

79

Chapter 4

4.2.3.3 Precipitation data The monthly precipitation data set for 2000 – 2009 were provided by the Department of Civil Aviation, Brunei. The rainfall data were recorded from Brunei’s meteorological observation station at Brunei International Airport.

4.2.3.4 Indices of the Southern Oscillation (SOI) The global-scale state of ENSO is measured in a number of ways including equatorial winds, trade wind anomalies and the Southern Oscillation Index (SOI). The SOI referred to throughout this study is the equatorial SOI, which is the standardised anomaly of the difference between the area-average monthly sea level pressure in an area of the eastern Pacific (80°W – 130°W, 5°N – 5°S) and an area over Indonesia (90°E – 140°E, 5°N – 5°S). The base period used for computing the anomalies is 1981-2010. Monthly average values of the equatorial SOI were extracted from NOAA’s Climate Prediction Centre website1. Sustained high positive values of the SOI represent La Niña event and sustained high negative values an El Niño. Subsequently, the warm (or cold) episode of El Niño (or La Niña) is normally associated with prolonged dry conditions (or wetter than normal) in Brunei (Sirabaha, 2008).

4.2.4 Effort unit selection There are considerable difficulties in defining suitable measures of effort, as fishing effort is a function of many variables which fishers can manipulate (Sparre and Venema, 1998; Maunder and Punt, 2004; Xiao, 2004). Usually, a single quantity (e.g. number of days at sea, total horse power, etc.) is used with the assumption that 1

http://www.cpc.ncep.noaa.gov/data/indices/

80

Chapter 4

relative change in other variables has exactly the same effects on the catch. In general, however, one does not always know a priori what the best measure of effort is. Therefore, it is always best to assess each effort variable and choose the best one as supported by appropriate hypothesis tests (Xiao, 2004).

In this study, four types of effort (f) unit were assessed. From the official database, the following were extracted; (f1) number of active vessels per month, (f2) number of fishing days at sea per month, (f3) number of trips made per month and (f4) total amount of horsepower of all vessels per month. Although the amount of horsepower (f4) exerted by the fleet is simply a product of number of active vessels and engine type, it is the effort unit currently being employed by the DOF in their monitoring and assessment of catch and effort data, and hence, useful for comparison.

4.2.5 Standardisation of CPUE Following previous section 4.2.4, the best measure of effort was found to be the number of days at sea per month (f2), and that monthly ‘nominal’ CPUE was assumed to be best expressed as catch (metric ton) per days at sea (this study, Section 4.3.3). However, differences in gear types may create variations unrelated to fish abundance (Maunder and Punt, 2004). A generalized linear model (GLM) approach, which are most recent and commonly used in standardisation of CPUE data (review by Hilborn and Walters, 1992; Maunder and Punt, 2004; Ye and Dennis, 2009), was then used to minimize bias and enable valid comparisons across the entire fishery to be made. In addition, the influence of climate forcing and other environmental factors are also considered in the standardisation procedure. The nominal CPUE data were log-transformed to meet assumption of normality and homoscedasticity in the

81

Chapter 4

error distribution, while at the same time accounting for the multiplicative nature of the relationship between the response and explanatory variables.

Subsequently, a ‘standardised CPUE’ index was estimated for overall demersal (finfish) resources group and for each of the 21 selected families (20 demersal fish + 1 chondrichthians ray), based on the following full model:

(

) ( (

)

(

)

(

)

(

)

)

where CPUE is the nominal CPUE (catch per days at sea) in month i and year j, β0 is the intercept, β1 – β4 are the effects of gear g, zone z, month i and year j, respectively, β5 – β9 are the effects of all two-way interactions except for the interaction between year and month, β10 – β13 are the effects of the environmental variables (i.e. SST, chla, rainfall and SOI), and

is a normally distributed error term. Other interactions

were not considered since they have no intuitive biological meaning. All the explanatory variables were treated as categorical, except for the environmental variables.

Akaike’s information criterion (AIC) was used to determine the set of factors and interactions that explained a substantial amount of the nominal CPUE variability (i.e. most parsimonious model). Accordingly, the smallest value of AIC indicates best model fit.

82

Chapter 4

4.2.6 Definition of stock status based on catch Each family were assessed to see if they have been overexploited or collapsed to low population abundance using the original classification of stocks based on catch relative to the maximum catch (Cmax) of Froese and Kesner-Reyes (2002) (Table 4.2).

TABLE 4.2: Original criteria used by Froese and Kesner-Reyes (2002) for assigning exploitation stages to fisheries, based only on catch data relative to maximum catch (Cmax). Status of fishery Year C/Cmax <0.1 Undeveloped/No info Before C = Cmax 0.1-0.5 Developing Before or after C = Cmax >0.5 Fully exploited 0.1-0.5 Overexploited After C = Cmax <0.1 Collapsed

4.2.7 Statistical analyses Principal components analysis (PCA) was used to explore the relationship between the four different types of efforts and the best measure of effort was taken as the variable that accounts for the most variance in the first component.

Using standardised CPUE, the stock abundance of the overall demersal finfish resources, as well as the abundance of each of the different demersal families, was examined by analysing their trends in catch and CPUE. The trends in catch were assessed using simple linear regression against time. Changes in standardised CPUE over the years were examined using the test for linear trend in one-way ANOVA (as variable “year” was taken as categorical in the GLM procedure). In addition, time lags in standardised CPUE of up to 4 years (i.e. 48 months) behind the environmental variables were tested using cross-correlation analysis. Standard time-series cross-

83

Chapter 4

correlation function (CCF) is applied to the environmental variables as the leading indicator, and residuals of the standardised CPUE as the lagging indicator.

4.3 Results 4.3.1 Exploratory analysis: distribution by gear Although the demersal assemblage are known to be targeted by both the trawl and long line fisheries only, a small proportion was also found in the purse seine fishery as incidental catch (Figure 4.1). However, in terms of volume, contribution by both the purse seine and long line fisheries are considered negligible (Figure 4.2).

PS

19

2

LL

30

25

TW

3

45

0%

20%

7

6

40% Demersals

60% Invertebrates

21

80%

100%

Pelagics

FIGURE 4.1: Proportion of number of different types of species caught by each gear in Brunei. Numbers inside the bars represent the actual number of species/groups of species. Overall, highest diversity of species are caught by trawlers (TW), followed by purse seiners (PS) and long liners (LL).

84

Chapter 4

1200

Zone 2

1000

1000

800

800

600

600

400

400

200

200

0

0

Purse-seine

Zone 3

Long line

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

2009

2008

2007

2006

2005

2004

2003

2002

2001

Trawl

2000

DEMERSAL CATCH (MT)

1200

YEAR

FIGURE 4.2: Contribution of the different gears on the overall demersal finfish catch (i.e. assuming 50% of the ‘mix’ category is attributed to demersal fish), by zone over the studied period (2000-2009).

4.3.2 Exploratory analysis: short- and long-term changes in environmental variables Monthly variations in SST, chl-a, rainfall and long-term climate index SOI are presented in Figure 4.3. Except for SOI, all other variables revealed seasonal variations (one-way ANOVA, p < 0.05 in all cases except SOI index, where p = 0.279), which follows a monsoonal pattern. The monsoon seasons were defined as per Sirabaha (2000): Northeast monsoon (Nov – Jan), transition I (Feb – Apr), Southwest monsoon (May – Jul) and transition II (Aug – Oct). Throughout the study period and averaged over both fishing zones (i.e. Zones 2 and 3 of Brunei fishery management zonation), SST varied between 29.9 ± 0.24 °C in May and 27.1 ± 0.23 °C in February, while chl-a peaked in December around 3.03 ± 3.08 mg m-3and were lowest around August at 0.76 ± 0.61 mg m-3. Amount of rainfall recorded on land ranged between 17.8 mm (minimum recorded) and 977.0 mm (maximum recorded), and are generally high during the NE monsoon months and low during post-NE monsoon period.

85

Chapter 4

2

10

(a) SST

(b) chl-a

ZONE 2 ZONE 3

8

1

6

°C

mg g

-1

0 -1

4 2 0

-2 -3 600

-2

ZONE 2 ZONE 3

-4 1.2

(c) rainfall

0.8

400

0.4

200

0 0

YEAR

FIGURE 4.3: Seasonal variation (i.e. detrended time series) of (a) sea surface temperature (SST), (b) chlorophyll-a concentration (chl-a), (c) rainfall, and (d) the Southern Oscillation Index (SOI) in Brunei between 2000 and 2009.

After correcting for seasonal factors, none of the climate variables showed any significant trends throughout the study period (Table 4.3).

TABLE 4.3: Summary statistics of seasonally-adjusted environmental variables in Brunei over the study period, from January 2000 to December 2009. Seasonally-adjusted Monthly Area of Standard variables against change t- stats P-value reference error time (Slope b) Zone 2 0.000 0.001 -0.283 0.778 SST Zone 3 0.000 0.001 0.007 0.994 Zone 2 0.002 0.001 1.633 0.105 Chl-a* Zone 3 0.003 0.003 1.268 0.208 All 0.494 0.355 1.391 0.167 Rainfall All -0.001 0.002 -0.470 0.639 SOI * Natural log-transformation was applied on chl-a data.

86

Jan-10

Jan-09

Jan-08

Jan-07

Jan-06

Jan-05

Jan-04

Jan-03

Jan-00

Jan-10

Jan-09

Jan-08

Jan-07

Jan-06

Jan-05

Jan-04

Jan-03

-1.2 Jan-02

-400 Jan-01

-0.8 Jan-00

-200

Jan-02

-0.4

Jan-01

mm

(d) SOI

Chapter 4

4.3.3 Best measure of effort Only one principal component was extracted from the four different types of efforts, which explained 93% of the total variance. This suggests that the use of any of the four types of effort to represent a measure of ‘fishing effort’ was justified, which is further supported by the direction of the component loadings of the four effort variables in the bi-plot graph (Figure 4.4). The best measure of effort was found to be f2 (number of fishing days at sea), which has the highest loading in the extracted component (Table 4.4).

1.0

100 80

0.5 Component 2

Percent of variance

3.719

60 40

f3 f2

0.0

f1 f4

-0.5

20 0.199

0.057

0 1

2 3 Component number

0.025

4

-1.0 -0.5

0.0 0.5 Component 1

1.0

Objects (adjusted to scale of component loadings) f1 Component loadings

FIGURE 4.4: PCA outcome of four different types of efforts. (a) Scree plot of variance explained by successive components before rotation, with data points labelled with eigenvalues; and (b) Bi-plot of first two principal components, showing the equivalence of all four measures of fishing effort.

TABLE 4.4: Principal component loadings from the correlation matrix between the four different types of effort. Only one component extracted (i.e. eigenvalue > 1), but the second component is also shown for comparison. Components Component 1 Component 2 f1 (number of vessels) 0.973 -0.170 f2 (number of days at sea) 0.979 0.073 f3 (number of trips made) 0.937 0.336 f4 (amount of engine horse power) 0.967 -0.228

87

Chapter 4

4.3.4 Standardisation of CPUE Summary results from the GLM analyses of nominal CPUE were given in Table 4.5. In all analyses, the total variance explained by the most parsimonious models range between 32 to 98%. Summary statistics of these models are given in Appendix 4.1.

For the whole demersal finfish group, types of gear accounted for most of the variance in its nominal CPUE (47%). However, at family level, 15 out of 21 data sets analysed revealed that variation in nominal CPUE (family) was best explained by fishing zones (5 – 90% of overall variance), while the type of gears was the most important factor in explaining nominal CPUE (family) variation in only four families (18 – 66% of overall variance). In general, the effects of gear (where applicable), fishing zones, year, month, and all two-way interactions except between year and month, were significant (p < 0.05), unless stated otherwise.

The GLM results also revealed that, in general, the influence of environmental variables were not significant in explaining the CPUE (family) variation, except in mojarras (Gerreidae), bigeyes (Priacanthidae) and croakers (Sciaenidae), at 15%, 33% and 7%, respectively, of overall variance.

No standardised CPUE values for the sicklefishes family (Drepaneidae) were calculated, as the total variance explained in all the full and other tested models were found to be non-significant (Full model: explained variance = 55.5%, F26,50 = 1.103, p = 0.409). In the remaining 21 data sets, visual inspection of the residuals from the log-normal error models suggested a reasonable overall fit.

88

Chapter 4

TABLE 4.5: Most parsimonious models used in the standardisation procedure of nominal CPUE of overall demersal resources and of specific categories at family level for Brunei waters between 2000 and 2009 (colon [:] signifies interactions between terms). Summary statistics of these models are given in Appendix 4.1. Dependent variable (natural log of nominal CPUE) Total demersal finfish resources Ariidae, sea catfishes Balistidae, triggerfishes Carangidae, jacks Drepaneidae, sicklefishes Ephippidae, batfishes Gerreidae, mojarras Haemulidae, grunts Lactariidae, false trevallies Leiognathidae, ponyfishes Lethrinidae, emperors Lutjanidae, snappers Mullidae, goatfishes Nemipteridae, threadfin breams Priacanthidae, bigeyes Psettodidae, halibuts Sciaenidae, croakers Serranidae, groupers Siganidae, rabbitfishes Stromateidae, butterfishes Trichiuridae, hairtails Dasyatidae, rays

Most parsimonious model year + month + gear + zone + month:gear + year:gear + gear:zone + year:zone year + month + zone year + gear + zone + year:zone year + month + gear + zone + month:gear + year:gear + gear:zone + year:zone + month:zone year + month + zone month + zone + SST year + month + gear + zone + year:gear + month:zone year + month + gear + zone year + month + gear + zone + month:gear + year:gear + year:zone + gear:zone year + gear + year:gear year + month + gear + zone + month:gear + year:gear + gear:zone + year:zone year + zone year + gear + zone + year:gear + gear:zone + year:zone year + month + gear + zone + month:zone + year:zone + rainfall year + month + zone + year:zone year + month + zone + year:zone + SOI year + month + gear + zone + year:gear + gear:zone gear + zone year + month + gear + zone + month:gear year + gear + zone + year:gear year + month + gear + zone + month:gear + year:gear + year:zone

N

Variance explained (%)

515

72.4

103 73

78.6 66.4

292

71.7

54 41 61

NS 93.7 68.9

128

61.7

130

83.2

211

69.7

47

64.7

399

81.4

48

83.6

167

78.4

60

97.8

126 149

85.9 62.1

354

71.7

58 72 124

31.8 58.7 32.8

249

84.1

Full model: ln(CPUEspp) = year + month + gear + zone + month:gear + year:gear + gear:zone + year:zone + month:zone + SST + chl-a + rainfall + SOI

89

Chapter 4

4.3.5 Lagged-effects of environmental variables on stocks’ CPUE index Only values from the trawl fishery were used to explore the potential lagged-effects of environmental variables on the demersal stocks’ CPUE index, due to the following reasons:

(1) In time-series analyses such as CCF, there is increasing inherent uncertainty as the proportion of missing values embedded in a time series increase, and (2) Most, if not all, of the demersal stocks are being exploited by trawlers, as the contribution by purse seines and long lines to total demersal catch can be considered negligible.

Even then, analysis of missing data patterns only identified a handful of stocks that would be seen sufficient for CCF analyses (Appendix 4.2). As a result, CCFs of SST, chl-a and rainfall (i.e. environmental parameters) were generated only against the overall demersal resources abundance and the abundance of nine selected demersal family stocks (Appendix 4.3).

In general, a discernible cyclic pattern of the CCF plots can be seen between the CPUE and SST time series, although the strength, direction and length of period vary between the different stocks. Such pattern is less obvious in CCF plots between the CPUE and chl-a time series, and almost non-existent in CCF plots between the CPUE and rainfall time series.

Understanding the effects of climate on a fishery requires first an understanding of the biology of the fish, and secondly its life stages that are likely to be affected by the 90

Chapter 4

environmental conditions. However, this requires in-depth analysis of an integrated life cycle and climate model of the stock, usually developed based on results from previous physiological, growth and tagging studies, which is beyond the scope of this thesis. Therefore, associations identified in this study are used to generate testable hypotheses for follow-on studies instead, and this is summarised in Table 4.6. Subsequently, environmental variables labelled “+” are expected to enhance reproductive success and cohort size of the stock and lead to an increase in CPUE values a number of months later, while those labelled “–“ are expected to hinder reproductive success and cohort size.

91

TABLE 4.6: Hypotheses linking climate variability and demersal stocks CPUE generated from the CCF analyses. Months with strong significant correlation between the environmental parameters as leading indicators and residuals from CPUE standardisation for each fish stocks as lagging indicators are given in parantheses. (E.g. SST value for Zone 3 has strong positive correlation with CPUE of snappers 6 months later.) CCF plots are given in Appendix 4.3.

Category

Overall demersal resources Carangidae, jacks Sciaenidae, croakers Lutjanidae, snappers Serranidae, groupers Haemulidae, grunts

Area (fishing zone)

General hypotheses made (a) Environmental variability influence stocks (b) Primary productivity at different life stages or in different area, (i.e. chl-a) affect stocks lower in affecting CPUE a number of months later the food chain first SST Chl-a Rainfall Species guild (a)

2

+ (34)

− (36)

3

− (22)

+ (22)

− (14)

2 3 2 3 − (13)

2 3

+ (6) − (13)

2 3 2 3

Lactariidae, false trevallies

2

Leiognathidae, ponyfishes

2 3

+ (6)

− (6)

+ (1)

Nemipteridae, threadfin breams

2

− (4)

+ (6)

+ (7)

− (9)

− (10)

Dasyatidae, rays

3 + (6), − (13)

3 2

− (5)

+ (6)

3

(c) Overexploited stocks are more susceptible to environmental variability State of stock in 2009(b)

n/a

Exploited

Intermediate predator

Exploited

Intermediate predator

Exploited

Intermediate predator

Over-exploited

Intermediate predator

Exploited

Zoobenthic feeder

Exploited

Zoobenthic feeder

Exploited

Zoobenthic feeder

Over-exploited

Zoobenthic feeder

Over-exploited

Large zoobenthos

Exploited

(a) Species guild based on Ecopath trophic groups for Brunei (Silvestre and Garces, 1993) (b) Stock status based on C/Cmax (see next section)

92

4.3.6 Trends in catch and CPUE index of demersal stocks in Brunei between 2000 and 2009 The overall demersal catch by the LS fishery sector annually showed no significant trend over the years investigated (Table 4.7; Figure 4.4). The highest annual catch of 1029 MT was attained in 2001, while the lowest annual catch of 392 MT occurred in 2006. On the other hand, the trends in annual catch varied among the different fish families (Table 4.7; Figure 4.5).

When normalised for the different gears and over both fishing areas, the annual standardised CPUE for the overall demersal group showed a declining trend over the years investigated (Table 4.8; Figure 4.5). Similar to the annual catch, the trends in CPUE varied among the different fish families (Table 4.8; Figure 4.6). Notably, whereas the standardised CPUE for most of the families showed a steady or increasing trend, the standardised CPUEs for ponyfishes (Leiognathidae), threadfin breams (Nemipteridae), croakers (Sciaenidae) and rays (Dasyatidae) exhibited significant declining trends. However, it is important to note that these trends were calculated for both fishing areas, and that localised decrease (or increase) in abundance were also observed in some stocks. 1,200

-2.5

(a) Annual catch

(b) Annual standardised CPUE Log-transformed CPUE (MT/day)

CATCH (MT)

1,000 800 600 400 200 0 2000

2003

2006

2009

-3

-3.5

-4

-4.5 2000

2003

2006

2009

FIGURE 4.5: Overall demersal (a) catch (in MT) and (b) standardised YEAR YEARCPUE (MT/day) of Brunei waters between 2000 and 2009. Annual standardised CPUE is the estimated marginal mean (i.e. predicted mean) for factor “year”, adjusted for other factors. Error bars show 95% confidence intervals.

93

Chapter 4

TABLE 4.7: Trends in annual catch of specific categories of fish and in total demersal finfish catch (MT) from 2000 to 2009 were shown to vary. The summary statistics for the best fit linear regression model are given. Also shown are mean annual catch over the years and state of stocks in 2009. Bonferroni correction states that alpha (α) should be set to 0.0023 for the p-value to be significant (i.e. α/n = 0.05/21). Only significant terms are shown. Significant p-values are in bold.

Category Total demersal finfish resources Ariidae, sea catfishes Balistidae, triggerfishes Carangidae, jacks Drepaneidae, sicklefishes Ephippidae, batfishes Gerreidae, mojarras Haemulidae, grunts Lactariidae, false trevallies Leiognathidae, ponyfishes Lethrinidae, emperors Lutjanidae, snappers Mullidae, goatfishes Nemipteridae, threadfin breams Priacanthidae, bigeyes Psettodidae, halibuts Sciaenidae, croakers Serranidae, groupers Siganidae, rabbitfishes Stromateidae, butterfishes Trichiuridae, hairtails Dasyatidae, rays

Mean annual catch (MT)

Annual change (Slope b x120)

p-value

583.5

-18.29

0.024

Exploited

9.8

-2.12

0.006

Overexploited

2.4

-0.83

0.025

Overexploited

23.9

1.65

0.001

1.7

-0.35

0.257

Overexploited

0.3 4.2 2.7

0.24 -1.54 0.41

0.003 0.111 <0.001

Overexploited Collapsed Exploited

77.5

0.09

0.963

47.5

-3.10

0.001

0.7

-0.09

0.574

60.8 19.0

-6.30 -0.22

<0.001 0.900

0.008

-6.857

28.5

Overexploited Exploited

59.4

-15.45

<0.001

0.020

-6.362

27.8

Overexploited

18.8

-3.50

0.033

10.7 35.2 14.7

2.49 -2.62 -0/16

<0.001 0.005 0.470

0.7

0.05

0.450

Exploited

0.9

-0.13

0.248

Overexploited

10.8 132.5

-0.46 1.18

0.590 0.313

Collapsed Exploited

Std. error

0.004

0.001

t-stats

3.552

4.996

R2 (%)

9.7

21.0

Status in 2009(a)

Exploited

Exploited 0.008

-3.256

8.3

Overexploited Overexploited(b)

Exploited 0.003

7.713

39.5

Exploited Exploited Exploited

(a) Based on C/Cmax (b) Based on catch in 2008

94

Chapter 4

Ariidae, sea catfishes

30

Balistidae, triggerfishes

12

25

10

20

8

15

6

10

4

5

2

0

40 30 20 10

0 2000

2003

2006

0

2009

Drepaneidae, sicklefishes

7

2000

2003

2006

2009

Ephippidae, batfishes

1.6

2003

2006

2009

Gerreidae, mojarras

25

1.2

5

20

1.0

4

0.8 3

15

0.6

2

10

0.4

1

CATCH (MT)

2000 30

1.4

6

Carangidae, jacks

50

5

0.2

0

0.0 2000

2003

2006

Haemulidae, grunts

7

0

2009

2000

120

6

2003

2006

2009

Lactariidae, false trevallies

100

5

2000

100

2003

2006

2009

Leiognathidae, ponyfishes

80

80 60

4 60

3

40

40

2

20

20

1 0

0 2000

2.5

2003

2006

2009

Lethrinidae, emperors

0 2000

2003

2006

2009

Lutjanidae, snappers

160

2003

2006

2009

Mullidae, goatfishes

60

140 2.0

2000

50

120 40

100

1.5

30

80 1.0

60

20

40

0.5

10

20 0.0

0 2000

2003

2006

2009

0 2000

2003

2006

2009

2000

2003

2006

YEAR FIGURE 4.6.1 – 4.6.12: Annual catch (MT) of selected family stocks between 2000 and 2009. Dashed line is shown if trend over study period is not significant. Two horizontal lines show range of catch when stock is overexploited (below top green line, where C = 0.5 x Cmax) or collapsed (below bottom red line, where C = 0.1 x Cmax), based on Cmax recorded within the study period. Note the different scales on the y-axis.

95

2009

Chapter 4

Nemipteridae, threadfin breams

250

Priacanthidae, bigeyes

60

30

50

200

Psettodidae, halibuts

35

25

40 150

20 30 15

100

20

50 0

5

0 2000

2003

2006

2009

Sciaenidae, croakers

70

0 2000

2003

2006

2009

Serranidae, groupers

25

60

CATCH (MT)

10

10

2000

2006

2009

Siganidae, rabbitfishes

1.4 1.2

20

50

1.0

40

15

0.8

30

10

0.6

20

0.4 5

10 0

0.2

0 2000

3.0

2003

2003

2006

2009

Stromateidae, butterfishes

25

2.0

20

1.5

15

1.0

10

0.5

5

2003

2006

2009

Trichiuridae, hairtails

30

2.5

0.0

0.0 2000

2003

2006

2009

2003

2006

2009

Dasyatidae, rays

200 160 120 80 40

0 2000

2000

0 2000

2003

2006

2009

2000

2003

2006

2009

YEAR FIGURE 4.6.13 – 4.6.21: (continued) Annual catch (MT) of selected family stocks between 2000 and 2009. Dashed line is shown if trend over study period is not significant. Two horizontal lines show range of catch when stock is overexploited (below top green line, where C = 0.5 x Cmax) or collapsed (below bottom red line, where C = 0.1 x Cmax), based on Cmax recorded within the study period. Note the different scales on the y-axis.

96

Chapter 4

TABLE 4.8: Summary statistics of test for linear trend in one-way ANOVA to assess trends in annual standardised CPUE of specific categories of fish and in total demersal finfish catch (MT/day) from 2000 to 2009 in Brunei. Direction of trends are indicated by “−” and “+” for decreasing and increasing trends, respectively. “0” denotes lack of significant trend. Bonferroni correction is applied with α = 0.0023. Significant p-values are in bold.

General Trend

d.f. (numerator, denominator)

F stats

p-value

Overall demersal resources



9, 411

2.958

0.002

Ariidae, sea catfishes

0

9, 68

2.823

0.007

Balistidae, triggerfishes

0

9, 51

2.370

0.025

+

9, 190

7.874

<0.001

0

9, 44

2.995

0.091

Ephippidae, batfishes

+

9, 16

8.150

<0.001

Gerreidae, mojarras

0

9, 51

0.764

0.650

Haemulidae, grunts

+

9, 84

3.730

<0.001

Lactariidae, false trevallies

+

9, 95

4.109

<0.001

Leiognathidae, ponyfishes



9, 146

9.745

<0.001

Lethrinidae, emperors

0

5, 36

3.729

0.008

Lutjanidae, snappers

+

6, 315

6.086

<0.001

Mullidae, goatfishes Nemipteridae, threadfin breams Priacanthidae, bigeyes

0

7, 31

2.675

0.027



8, 133

6.637

<0.001

0

1, 18

1.129

0.302

Psettodidae, halibuts

+

9, 85

39.050

<0.001

Sciaenidae, croakers



9, 107

7.146

<0.001

Serranidae, groupers

+

9, 298

5.034

<0.001

Siganidae, rabbitfishes

0

7, 49

0.813

0.581

Stromateidae, butterfishes

0

9, 39

2.205

0.043

Trichiuridae, hairtails

0

9, 99

2.458

0.014

Dasyatidae, rays



9, 190

31.121

<0.001

Category

Carangidae, jacks Drepaneidae, sicklefishes

(a)

(a) Nominal CPUE values were used instead of standardised CPUE values.

97

Chapter 4

Ariidae, sea catfishes

-3.0

Balistidae, triggerfishes

-2.0

Carangidae, jacks

-3.0

-4.0 -5.0

-5.0 -6.0

-7.0

-7.0 -8.0

-9.0 2000 -5.0

2003

2006

2009

Drepaneidae, sicklefishes

-10.0 2000 -2.0

2003

2006

2009

Ephippidae, batfishes

-9.0 2000

2003

2006

2009

Gerreidae, mojarras

-5.0

-4.0 -6.0

-7.0

Log-transformed CPUE (MT/DAY)

-6.0 -7.0

-9.0 -8.0

-11.0 2000 -5.0

2003

2006

2009

Haemulidae, grunts

-10.0 2000 -5.0

2003

2006

2009

Lactariidae, false trevallies

-6.0

-6.0

-8.0 2000 -1.0

2003

2006

2009

Leiognathidae, ponyfishes

-3.0

-7.0 -7.0

-5.0 -8.0

-8.0

-9.0 2000 -4.0

-7.0

-9.0

2003

2006

2009

Lethrinidae, emperors

-10.0 2000 -3.0

2003

2006

2009

Lutjanidae, snappers

-9.0 2000 0.0

2003

2006

2009

Mullidae, goatfishes

-2.0 -6.0

-4.0

-8.0

-4.0 -6.0

-5.0

-8.0 -10.0 2000

2003

2006

2009

-6.0 2000

2003

2006

2009

-10.0 2000

2003

2006

2009

YEAR FIGURE 4.7.1 – 4.7.12: Annual standardised CPUE (MT/day) of selected family stocks between 2000 and 2009 in Brunei. Value given is the estimated marginal mean (i.e. predicted mean) for factor “year”, adjusted for other factors. Error bars show 95% confidence intervals. Note the different scales on the y-axis.

98

Chapter 4

Nemipteridae, threadfin breams

-2.0

-3.0

-3.0

Priacanthidae, bigeyes

-2.0

Psettodidae, halibuts

-4.0

-4.0

-6.0

-4.0 -5.0

-8.0

-5.0 -6.0

-6.0

Log-transformed CPUE (MT/DAY)

-7.0 2000

2003

2006

2009

Sciaenidae, croakers

-2.0

-7.0 2000 -4.0

-10.0

2003

2006

2009

Serranidae, groupers

-12.0 2000 -5.0

2003

2006

2009

Siganidae, rabbitfishes

-5.0

-4.0

-6.0 -6.0

-6.0

-8.0 2000 -2.0

2003

2006

2009

Stromateidae, butterfishes

-7.0 2000 0.0

-4.0

-2.0

-6.0

-4.0

-8.0

-6.0

-10.0

-8.0

-12.0 2000

2003

2006

2009

Trichiuridae, hairtails

-7.0 2000 -2.0

2003

2006

2009

Dasyatidae, rays

-4.0

2003

2006

2009

-10.0 2000

-6.0

2003

2006

2009

-8.0 2000

2003

2006

2009

YEAR FIGURE 4.7.13 – 4.7.21: (Continued) Annual standardised CPUE (MT/day) of selected family stocks between 2000 and 2009 in Brunei. Value given is the estimated marginal mean (i.e. predicted mean) for factor “year”, adjusted for other factors. Error bars show 95% confidence intervals. Note the different scales on the y-axis.

99

Chapter 4

The standardised CPUE indices at family levels revealed that generally, abundance is higher in the deeper waters of Zone 3. Also noted was the lack of significant difference in abundance between both fishing areas for grunts (Hamullidae), ponyfishes (Leiognathidae), butterfishes (Stromateidae) and rays (Dasyatidae) (Table 4.9). TABLE 4.9: Comparison in stock abundance between Zone 2 and Zone 3 of Brunei fishing zonation, using standardised CPUE. Bonferroni correction is applied with α = 0.0023. Significant p-values are in bold. Category Total demersal finfish resources Ariidae, sea catfishes Balistidae, triggerfishes Carangidae, jacks Drepaneidae, sicklefishes(a) Ephippidae, batfishes Gerreidae, mojarras Haemulidae, grunts Lactariidae, false trevallies Leiognathidae, ponyfishes Lethrinidae, emperors Lutjanidae, snappers Mullidae, goatfishes Nemipteridae, threadfin breams Priacanthidae, bigeyes Psettodidae, halibuts Sciaenidae, croakers Serranidae, groupers Siganidae, rabbitfishes Stromateidae, butterfishes Trichiuridae, cutlassfishes Dasyatidae, rays

Mean standardised CPUE (x 10-3MT/day) Zone 2 Zone 3

d.f.

t-stats

p-value

45.17

76.20

483

-3.845

<0.001

1.49 0.65 3.20

13.73 3.71 13.06

88 65 65.8*

-11.309 -11.146 -5.357

<0.001 <0.001 <0.001

0.45

-

-

-

-

0.83 0.59 0.75

12.57 8.64 5.42

36 59 4.1*

-4.668 -9.453 -2.327

<0.001 <0.001 0.079

15.92

0.75

116

5.204

<0.001

7.72

14.62

47.4*

-1.360

0.180

0.95 10.02 1.66

0.30 35.21 37.82

3.6* 319.2* 38

3.540 -17.618 -16.847

<0.001 <0.001 <0.001

2.87

64.48

61.8*

-17.575

<0.001

2.34 2.45 5.35 6.49 1.45

32.63 6.55 22.63 5.52 10.69

50 110 133 325.9* 55

-10.328 -3.900 -7.143 -3.766 -11.795

<0.001 <0.001 <0.001 <0.001 <0.001

0.25

1.13

62

-2.418

0.019

2.25

4.50

114

-3.213

<0.001

21.47

22.96

204.1*

-2.193

0.029

(a) Nominal CPUE values were used instead of standardised CPUE values. * Where p < 0.05 for Levene’s test for equality of variances, adjustments were made to the degrees of freedom and t-statistics were calculated without assuming equal variances between the two groups.

100

Chapter 4

4.4 Discussion and Conclusion This chapter further confirms the presence of shifts in species composition of Brunei’s demersal habitat (Chapter 3) as well as highlighting the sources of bias that can exist when using short time series of fishery-dependent catch and effort data, as it has been done here. It is worth stressing, however, that standardised CPUE calculated in this study were used to assess relative trends in biomass or abundance of the stocks and not to estimate absolute stock size or the contribution of different fleets on fishing mortality. Indeed, some have warned against the use of commercial catch and effort data as an index of abundance, especially if the fundamental assumption of direct proportionality is violated (Chapter 2, Section 2.2.2). Judicious use of catch and effort data, however, may still provide useful insights and can be beneficial in some situations, particular in data-poor fisheries such as Brunei.

4.4.1 Best measure of effort Accurate estimates of fishing effort are essential for accurate biological (e.g. stock assessment), economical (e.g. estimating profitability of a fishery) and social (e.g. designation of marine protected areas) assessments in facilitating sustainable fisheries management (as reviewed by McCluskey and Lewison, 2008). A common challenge, however, is to choose an appropriate measure of fishing effort that reflect a measure of direct effect of fishing on stocks mortality. Measuring effort is not a new problem – considerable energy has been applied by fisheries researchers to develop reliable measures of effort (Beverton and Holt, 1957; Bordalo-Machado, 2006; McCluskey and Lewison, 2008). However, the diversity of effort measures currently being used in fisheries illustrates that the choice to date has been justified

101

Chapter 4

to concentrate more on what is easiest to measure rather than what is ideal in terms of identifying a measure of effort that is correlated to fish mortality. Common measure of effort would usually describes the resources allocated to fishing such as time (days or hours fished), capital (number of vessels, length or horsepower of vessel), labour (number of crew, distance travelled, number of trips made) or gear (mesh size or number of hooks) (McCluskey and Lewison, 2008).

The most frequently used quantitative measure of effort is that of time spent fishing (e.g. Xiao, 2004; Battaile and Quinn, 2004; Ortiz de Zarate and Ortiz de Urbina, 2007; Rist, 2007). Since the total time spent on a fishing trip would usually compose of time taken to the fishing ground and time spent searching and fishing, there have been studies which argue that the use of time measures could lead to considerable biased estimates of biologically relevant fishing effort (Hilborn and Walters, 1992). Accordingly, the proportion of total time on a fishing trip that is spent fishing would decrease with increasing distance to the fishing ground. If this is unadjusted, fishing effort may be increasingly overestimated with increasing distance (Hilborn and Walters, 1992; Rist, 2007). However, considering the small area of the fishing grounds in this study, the total time spent on the fishing trip (i.e. number of operating days out in the sea per month) as a proxy for time spent actively fishing is deemed reasonable. For instance, in good weather, it may take less than 15 hours to reach the furthest point of the far fishing ground (i.e. Zone 3). Typically, in the case of the trawl fleet, fishers would leave the port in the morning and carry out trawling both during the day and at night for three consecutive days before returning to Muara FLC. Given adequate supply and maintenance, each vessel makes between four to eight trips per month on average (DOF-pers.comm.).

102

Chapter 4

4.4.2 CPUE standardisation Fishery-dependent data collection is one of the most valuable tools available to fishery managers, especially since fishery-independent data are often extremely costly or difficult to collect (Maunder and Punt, 2004). Collection of even the simplest data set can help eliminate the threat of overfishing and subsequent population collapse. To allow valid comparisons of catch rates over time or across fishing grounds and to account for changes in vessel catchability, it is essential to standardise CPUE values to adjust for the impact on catch rates of changes over time of factors other than abundance (Beverton and Holt, 1957; Hilborn and Walters, 1992; Maunder and Punt, 2004; McCluskey and Lewison, 2008).

The GLM models in this study explained between 32 – 98% of the overall variation in the official catch and effort data. Ye and Dennis (2009) remarked that the lower range fraction of explained variation is low by normal statistical standards, but is in the range that most CPUE standardisations can achieve. In fact, Ye and Dennis (2009) also pointed out that the reliability of CPUE standardisation should not be judged based on the model’s capacity to explain data variation. For instance, explained variance of models used by Battaile and Quinn (2004) for CPUE standardisation of walleye pollock (Theragra chalcogramma) in the eastern Bering Sea ranges between 20 – 70%, while standardisation used by Yamada et al. (2006) for Pacific bluefin tuna (Thunnus orientalis) explained 17 – 24% of the overall variation. In fact, the average percentage of explained variation obtained in this study (i.e. ~73%) is considered relatively high in comparison to other studies, although this is suspected to be due to the large number of model parameters resulted from the inclusion of two-way interactions between the main effects.

103

Chapter 4

In principle, the fraction of variability explained can increase substantially by including more explanatory variables (Maunder and Punt, 2004). The number of explanatory variables to be considered in CPUE standardisation often varies in literature and usually depends on the type of fishery. In this study, the number of explanatory variables considered is relatively low compared to other studies. For example, Horn (2003) considered 23 possible explanatory variables when standardising the catch and effort data for ling (Genypterus balcodes) off New Zealand, while Siddeek et al. (2012) used 10 explanatory variables to standardise CPUE data of golden king crab (Lithodes aequispinus) in the Aleutian Islands. Nevertheless, the main effects of year, month, gear types and fishing area are the standard set of explanatory variables commonly used in catch and effort standardisation (Maunder and Punt, 2004).

All two-way interactions between the main operational, temporal and spatial effects were considered in the analyses, except for the year x month interaction as there is no a priori reason to include the year x month interaction in the standardisation process.

The year x zone and month x zone interactions were significant when the yearly or monthly variation in CPUE varied in different fishing area. This suggests that there are changes in the geographical distribution of stocks, either as a result of localised overexploitation or annual and seasonal migration pattern (Hvingel et al., 2000). Seasonal migrations of some of the demersal stocks between Brunei Bay and the nearby coastal waters of Brunei have been proposed by Silvestre and Matdanan (1992). Seasonal migration between the shallow waters of fishing area Zone 2 and the deeper waters of fishing area Zone 3, however, has not been demonstrated. 104

Chapter 4

Consequently, there is a need to further clarify whether such variation is an outcome of overexploitation or natural migration pattern, especially if Brunei were to effectively manage its demersal stocks based on management zones.

This study found that the year x gear interaction was significant in about half of the data sets analysed since a different CPUE trend between the different gear types (which is essentially referring to the different fishery fleets) is to be expected. A year x gear interaction may also account for the improvements in technology or skipper experience between the different fishery fleet. A significant month x gear interaction, on the other hand, reflects the differing fishing strategy between the different fishery fleet annually. While trawlers tend to operate the whole year round, the peak of purse seiners and long liners fishing activities are mostly between March and October. The lack (or availability) of target species seems a likely reason for the fishing activity patterns of purse seiners, since pelagic stocks availability have been observed to depend on the cycle of the monsoons in South China Sea (Ruddle, 1986; Ruddle and Davis, 2011). However, the same might not be relevant to the long liners, since they also target the same finfish stocks as the trawlers. Instead, it was reported that such variation was due to operational difficulties of the long line fleet (DOF-pers.comm.). It should be noted, however, that trawlers appear to work all year round because of the non-selective nature of their gear, rather than the availability of demersal finfish stocks all year round (i.e. they also switch target species depending on stocks availability; DOF-pers.comm.).

In this study, variations within each fleet (i.e. individual vessels) were not considered, which is unlike those commonly done in some of the catch-effort 105

Chapter 4

standardisation studies (e.g. Sbrana et al., 2003; Ortiz de Zarate and Ortiz de Urbina, 2007). However, the effect of vessel characteristics may be accounted to some degree in the gear x zone interaction. Accordingly, vessels operating in fishing area of Zone 3 are bigger and have greater fishing capacity then vessels operating in Zone 2 (DOF-pers.comm.). Nevertheless, it is highly recommended that future studies on CPUE standardisation of Brunei catch-effort data should include individual vessel characteristics, especially with the varying level of experience between the skippers in the different sectors.

The GLM analyses also showed that the full model for CPUE standardisation of sicklefishes (Drepeneidae) is not significant, thus suggesting that the model is inadequate. This inadequacy may be in the relationship between CPUE and abundance or due to additional variation in CPUE not explained by the explanatory variables.

4.4.3 Potential influence of environmental variability on selected demersal stocks in Brunei GLM results from this study indicate that environmental variables are only significant in three out of 21 demersal stocks analysed and that, generally, the influence of environmental factors appear negligible on the demersal stocks abundance. Both SST and rainfall produce identical negative correlational response on mojarras (Gerreidae) and bigeyes (Priacanthidae), respectively. On the other hand, croakers (Sciaenidae) were found to be positively correlated with SOI values during the study period, which is essentially associated with La Niña phase of ENSO. Such outcomes most likely identified the intermediate effect of climate and the environment on fish catchability. For instance, mojarras (Gerreidae) are known 106

Chapter 4

zooplankton feeders in Brunei waters (Silvestre et al., 1993), and with few evidence suggesting that high temperature will accelerate planktonic larval development and reduce larval duration for many tropical fish (Munday et al., 2012), the combined effect of food availability reduction as well as fishing may have resulted in instantaneous increase in mortality (and hence, decrease in catchable population for the fishery).

However, species responses to environmental parameters do not appear to be easily generalized in the literature, especially since the way in which the environment can affect stock abundance is particularly complex as it involves mechanisms that may act at several life history stages (e.g. Beentjes and Renwick, 2001; Balston, 2007; Beaugrand and Kirby, 2010; Olsen et al., 2011). As a result, inclusion of environmental variables in CPUE standardisation procedure is rarely done (Damalas et al., 2007), especially for demersal species, as evident from the paucity of such studies in the literature. Often, these environmental variables do not have as much explanatory power as, or may be confounded with, other operational, spatial and temporal effects. It should be noted, however, that the possible presence of colinearity should not be a problem in this study since the goal of the standardisation is to generate an index of relative abundance rather than to investigate the variables that explain variation in CPUE (Maunder and Punt, 2004). Nonetheless, this study could be considered an initial step towards understanding the effect of the environment on a multi-specific demersal community in the tropics, which has been left out of focus of the scientific community up to now, perhaps due to the small range of variability of environmental factors compared to the temperate regions.

107

Chapter 4

One exploratory approach to disentangle the confounding human and environmental effects on fish stocks is to test correlation between stocks’ CPUE and the environmental parameters at various lags, based on the hypothesis that these environmental factors influence fish production by controlling nutrient supply for prey production for fishes, and that fishery catches are age-structured (Balston, 2007; Qiu et al., 2010).

Analyses from the 1989/1990 trawl survey in Brunei waters revealed that the majority of the demersal stocks, which comprised roughly 70% of total demersal biomass estimated, have relatively fast turnover rates, typical of small-sized and short-lived tropical fish species (Silvestre and Garces, 2004). Although higher-TL species such as snappers (Lutjanidae) and groupers (Serranidae) may be fully recruited to the fishery from ages 3 onwards, it is assumed that the fishes, in general, would have a dominant age-at-catch of ≤ 3 years (see Qiu et al., 2010). Hence, time lags were tested for up to four years (i.e. 48 months) to allow for time taken for the input nutrients to be circulated among the ecosystem and transferred into prey supply (Beentjes and Renwick, 2001; Balston, 2007; Qiu et al., 2010). Furthermore, it was unknown whether environmental variability has a direct effect upon spawning success or upon juvenile survival in the year following spawning (Beentjes and Renwick, 2001; Olsen et al., 2011). Lagged-effects of ENSO via SOI were not considered in this study since there are difficulties in separating spuriously high (or low) SOI values from those that occurred during a developed La Niña (or El Niño) phase.

108

Chapter 4

The CCF analyses returned a number of significant correlations between the environmental parameters, and the CPUE of overall demersal resources, as well as the nine selected demersal stocks. Such relationship, however, should be treated with caution, since relating large number of CPUE time series to environmental variables, and testing several time lags, may add to the chance of type I error in statistics. Additionally, marked periodicity in the correlation values were observed for some of the analyses, particularly for CCF plots with SST time series, which partly may be due to temporal (seasonal) correlations present in the data.

For intermediate predators (i.e. based on Ecopath trophic groups used by Silvestre et al. (1993)), such as snappers (Lutjanidae) and groupers (Serranidae), the strongest correlation with chl-a occurred at 13-month lag, corresponding to the condition one year prior to the catch. Demersal zoobenthic feeders stocks, such as ponyfishes (Leiognathidae) and threadfin breams (Nemipteridae), on the other hand, revealed its top two strongest correlation values at 6-/9- and 13-month lag. Such outcomes: (1) support the observation that annual recruitment can either be in single or dual pulses for these stocks in Brunei (Silvestre and Matdanan, 1992; Silvestre and Garces, 2004), and/or (2) suggest the hypothesis that approximately a year (or six month) is needed to transfer chl-a to intermediate predator (or demersal zoobenthic feeder) stocks (Sartimbul et al., 2010). However, while it may be reasonable to presume abundance being driven by the occurrence of such lagged-effects on short-lived species (Postuma and Gasalla, 2010; Quetglas et al., 2013), precaution is necessary for longer lived species where recruitment, and hence abundance, highs and lows can be buffered by a large number of year classes (Arreguín-Sánchez et al., 1996; Chambers and Trippel, 1997). Moreover, since little knowledge is available of the

109

Chapter 4

spawning ground location and the life-history stages for these local stocks, one can only speculate on the reasons to explain how the environmental factors relate to the stock abundance.

Also particularly noteworthy is the state of the stocks which exhibited significant CCFs with the environmental variables. Out of the eight fin-fish stocks analysed, those that were significantly correlated with an environmental factor coincidentally were in an ‘overexploited’ state during the last year of the study period (except groupers (Serranidae) but see below). Likewise, there have been several studies which demonstrated that stocks that are in an overexploited state seem to have been more susceptible to environmental variability (Hsieh et al., 2006; Perry et al., 2010). There is increasing evidence that harvested species fluctuate more than unharvested ones, which is probably due to the elevated sensitivity to environmental variability resulting from the demographic truncation caused by fishery exploitation (Hsieh et al., 2006; Anderson et al., 2008; Quetglas et al., 2013).

Based on the catch data throughout the study period, the grouper (Serranidae) stock is classified as exploited for the year 2009. However, considering that groupers (Serranidae) are K-species, with well-studied characteristics of being prone to overexploitation even under relatively light fishing effort (Arreguín-Sánchez et al., 1996; Jennings et al., 2001), it is not surprising then if the catch (and hence the calculated CPUE value) may be overestimated (i.e. hyperstability).

110

Chapter 4

4.4.4 Probable cause of non-proportionality While strict proportionality between CPUE and abundance is frequently assumed, it has long been recognised that CPUE may not accurately reflect changes in abundance (Beverton and Holt, 1957), thus resulting in either a hyperstability or hyperdepletion scenario (Chapter 2, Section 2.2.2).

For CPUE to be proportional to abundance in an area, it assumes that the effort is distributed at random within that area with respect to the stocks. However, it is almost impossible for this to be true, since fishers will go where they believe the fish stocks to be, thus resulting in effort to be concentrated on those sites of highest abundance. Consequently, different spatial allocation of effort and efficient search by the fishers can lead to hyperstability (Salthaug and Aanes, 2003), causing the CPUE to be a poor measure of population abundance. For instance, consider that two areas initially had the same level of stock abundance, where one is closer to shore while the other is far away, expensive to exploit, or for some other reason undesirable. Effort will first be spent in the nearby area, but after a while the fishers are willing to trade off the undesirable aspects of the other area for its higher CPUE. Accordingly, the CPUE may remain stable or even increase, even though the total abundance may still be declining (Hilborn and Walters, 1992; Walters, 2003). To avoid this, one common approach is to spatially disaggregate CPUE data and patterns of effort, to reveal localised declines (Walters, 2003; Maunder et al., 2006). The spatial resolution of the existing data in this study, unfortunately, is too crude to discern fine spatial changes. Nevertheless, the relatively higher CPUE values shown for fishing area Zone 3 in Brunei waters suggests that abundance may still be greater in Zone 3, even if the total abundance of overall demersal stocks for the whole of Brunei water

111

Chapter 4

may be declining. This scenario, however, will become a major concern if one considers the two fishing areas to be of different productivity and hence different level of stock abundance initially. Considering that almost all of the reefs and ‘notake’ zones are located in the coastal area of Zone 2, and that stock abundances are likely to decrease with depth (Silvestre and Matdanan, 1992; Arreguín-Sánchez et al., 1996; Silvestre and Pauly, 1997a), the lower CPUE values observed in Zone 2 would then suggest localised overexploitation of the demersal stocks. In this case, hyperdepletion is unlikely, since the assumed area of higher productivity (i.e. Zone 2) would be nearer, less expensive to exploit, and hence more desirable (Hilborn and Walters, 1992).

Interestingly, this study also identified several demersal stocks which exhibited similar level of abundance between both fishing zones. This could be due to the following: (1) level of exploitation on fish stocks is similar in both zones (i.e. for the first scenario where the initial level of abundance is assumed to be the same), or (2) stocks in area of higher productivity (i.e. Zone 2) may be overexploited to low abundance, thus closely match those in area of lower productivity (i.e. Zone 3). In the case of the ponyfish (Leiognathidae) stock, overexploitation to localised depletion is likely since a significant declining trend was observed in Zone 2 during the study period. However, similar conclusion could not be found for the grunt (Haemulidae), butterfish (Stromateidae) and ray (Dasyatidae) stocks.

Conversely, variable catchability between stocks, either through behaviours of fishers (i.e. gear saturation effect) or fish (i.e. different species respond differentially to the gear), could also cause bias in the relationship between CPUE and abundance

112

Chapter 4

towards hyperdepletion (Hilborn and Walters, 1992). Given that the main contribution to catch are those caught by trawlers in this study, if assuming that hyperdepletion exists, this is more likely due to the gear saturation effect then, as trawlers are only sensitive to size and not species type. However, hyperdepletion in fisheries rarely occurs. There are some examples, such as the south Australian rock lobster fishery (Hilborn and Walters, 1992), but hyperstability appears to be the most common relationship (Harley et al., 2001). Some of the well known fisheries collapses in the world are ascribed to hyperstability (e.g. the northern cod stock collapse (Shelton, 2005)), and hence considered to be one of the biggest problems for fisheries managers.

While there is much evidence to suggest that hyperstability frequently occurs in the relationship between CPUE and abundance, the assumption of proportionality may still be valid in some cases (e.g. Haggarty and King, 2006; Ye and Dennis, 2009). Unfortunately, lack of access to fishery-independent data available meant that the use of CPUE as abundance indicators in this study could not be evaluated. In fact, in many cases, the use of fishery-dependent data represents the only method available for abundance estimation (Maunder and Starr, 2003; Tsehaye, 2007). Nevertheless, standardised CPUE values may still provide a more accurate measure of uncertainty around the indices.

4.4.5 Spatio-temporal changes in CPUE of demersal stocks in Brunei In this study, the trends in CPUE may appear slightly more optimistic than the demersal abundance trends calculated from the trawl surveys conducted by the DOF in 1999 and in 2008 (DOF, 2011). The DOF reported that the geometric mean catch

113

Chapter 4

rate in Brunei waters (fishing zones 2 and 3 combined) in 2008 had declined by about 60%, whereby only 38.6% of the overall demersal stocks are left when compared to the abundance in 1999 (DOF-pers.comm.). Although the reduction in overall demersal resources between the start (2000) and the end (2009) of the study period is comparable to the DOF report (i.e. abundance level in 2009 is 39.2% of the abundance level in 2000), the rate of change over the years greatly varies between different stocks and fishing areas.

With few exceptions, it is interesting to note that majority of the fish families which are shown to increase in abundance are of the intermediate predator category, such as jacks (Carangidae), groupers (Serranidae), halibuts (Psettodidae) and batfishes (Ephippidae). On the other hand, demersal zoobenthic feeders such as ponyfishes (Leiognathidae) and threadfin breams (Nemipteridae) had significantly decline over the study period. Such pattern appears contradictory to the “fishing down the marine food web” phenomenon described earlier in the thesis (Chapter 3), while Gulland (1972) similarly argued that fishing would have reduced fish that are expensive and attractive to fishers first. However, such changes are, in fact, not new and have been identified previously in the overexploited demersal fishery of the Gulf of Thailand (Pauly, 1988). Specifically, Tiews et al. (1967) who first presented evidence for the massive changes in the species composition of the Gulf of Thailand demersal resources due to fishing in the 1960s noted that while some of the previously abundant groups such as the ponyfishes (Leiognathidae) had declined, there was a marked increase of snappers (Lutjanidae) and squid (Loligo duvauceli). Similar findings were also observed by Pauly (1979) in a later analysis for the Gulf of Thailand, and by Koranteng (1998) on the outcomes of several trawl surveys carried

114

Chapter 4

out in the large marine ecosystem of Ghana. It should be noted, however, that the analyses presented in this chapter have treated each family as an entity, by portraying trends in overall abundance. However, the various species in each family could have acted differently from the group behaviour as a result of possible differences in response to factors such as life history, environmental changes, and reaction to fishing gear.

4.4.6 General limitations of analyses Limitations and source of bias of using fishery-dependent data and CPUE as an index of abundance has been covered extensively in the literature (e.g. Harley et al., 2001; Maunder and Punt, 2004; Froese et al., 2012). Inherently, some of the methods used in this chapter may be open to criticism. For instance, the combination of catch data across species to monitor community abundance means that the trends perceived can be misleading as it may reflect changes in abundance of one or few dominant species. Moreover, since the fleet mainly exploit stocks that can be caught with their gears, the index of abundance are therefore indicative of the older and larger sizedstocks.

While GLMs are considered a powerful statistical technique, the fraction of overall variation in the data explained by a catch-effort standardisation eventually depend on the explanatory variables, and the key assumption that the relationship between some function of the expected value of the response variable (i.e. CPUE) and the explanatory variables is linear. Nonetheless, many different approaches have been considered in catch-effort standardisations as this is a rapidly developing field. For example, there are many studies which have used extensions of the GLM approach

115

Chapter 4

such as general additive models (GAMs) and generalized linear mixed models (GLMMs). Subsequently, GAMs enable the linear predictor in GLM to be replaced by an additive predictor, giving it a partially non-parametric aspect, while GLMMs extend the GLM approach by allowing some of the parameters in the linear predictor to be treated as random variables. Other recent approaches in catch-effort standardisation have been reviewed by Maunder and Punt (2004).

Even though great care has been taken in extracting the right data to answer the questions posed in this study, the results obtained ultimately depend upon the data provided. For instance, Brunei has been noted for its high discard rate (Kelleher, 2005; DOF-pers.comm.), while the extent of illegal, unreported and unregulated (IUU) fishing in Brunei and Southeast Asian region is of considerable concern (APFIC, 2007). Unfortunately, the fishery authorities do not appear to have suitable means to check the authenticity of the data supplied – a common setback of fisherydependent data worldwide.

In brief, while the abundance trend of overall demersal resources in Brunei waters is in decline between 2000 and 2009, the trends for the different demersal family stocks vary greatly over time and between fishing areas. Potential influences of environmental parameters, separated from fishing effect, have been identified and could serve as hypothesis-generators for future research. However, inferring changes based on short-term data can be risky, which leads some authors to tap into memories of fishers. This is further considered in the next Chapter 5 (Utilisation of local ecological knowledge to assess status of selected demersal stocks in Brunei).

116

Chapter 5 Utilisation of local ecological knowledge to assess status of selected demersal stocks in Brunei “People don’t know the past, even though we live in literate societies, because they don’t trust the sources of the past.” – Daniel Pauly, 2010

Chapter Summary Fishers’ local ecological knowledge (LEK) was collected through a rapid appraisal process using map-based semi-structured interviews on 259 currently active fishers in Brunei Darussalam. Based on fishers’ perceptions, this study examined changes in relative abundance between the 1960s and the 2000s, of four demersal stocks (red snappers Lutjanus erythropterus, groupers Epinephelus spp., ponyfishes Leiognathus spp. and rays Dasyatis spp.) and two pelagic stocks (Japanese scads Decapterus maruadsi and narrow-barred mackerels Scomberomorus commerson). Red snappers were cited more than any other species as depleted. Highly experienced fishers (>40 years) recalled greater past abundance than less experienced fishers (<10 years) and were more likely to have caught larger red snappers from shallow inshore waters than further offshore. While none of the fishers acknowledged an increase in abundance for any of the six selected stocks, the more experienced fishers’ perceptions on ponyfishes, Japanese scads, narrow-barred Spanish mackerels and rays showed a significant downward trend. This indicates that experienced fishers acknowledged a greater depletion since 1960s than less experienced fishers, and provides another compelling case of the ‘shifting baseline syndrome’ (SBS), whereby different generations of fishery stakeholders have altered perceptions of their environment. A fuzzy logic expert system was used to standardise and quantify the anecdotal evidence to produce a decadal time series of resources abundance from 1960 to present. This study also incorporated the use of geographic information system (GIS) for collecting and systematizing fishers’ LEK, which enabled identification of the preferential habitats of demersal fish stocks and the highpressure harvest zones for Brunei fisheries. Fishers’ opinions on current management practice revealed that the fishery suffers from a “trust gap” between the fishery authorities and the fishing communities. Understanding the prevalence and implications of SBS can make better use of resource user knowledge, while the extraction of LEK as a source of viable information can provide insights for the fishery authorities in the development of spatially explicit management measures in Brunei waters. 117

Chapter 5

5.1 Introduction Understanding of the full influence of fishery activities in Brunei, as elsewhere in the region, is often hindered by the short time-series of many of the datasets on fishery catch statistics. One means of overcoming constraints in data-poor fisheries, which characterise many developing countries, is to tap into the memories of fishers – an approach linked to the emerging discipline of historical ecology (Johannes et al., 2000; Silvano and Valbo-Jorgensen, 2008; Venkatachalam et al., 2010).

Besides determining past abundance patterns of target fish, long before the onset of data recording by scientists, local ecological knowledge (LEK) studies may uncover the “shifting baseline syndrome”, whereby each generation of stakeholders accepts a lower standard of resource abundance as being normal (Pauly, 1995).

This chapter describes a questionnaire developed for the collection and analysis of information derived from LEK of fishers operating in Brunei waters, and its application to four selected demersal stocks and two pelagic stocks, with special reference to the demersal red snapper (Lutjanus erythropterus). By using Geographical Information Systems (GIS) as well as a fuzzy expert system, this study aimed to systematically convert, standardise and quantify LEK into a form that is suitable for consideration and analysis by fisheries managers and scientists. Specifically, this chapter addresses the following research questions: 1) How can fishers’ LEK in Brunei be collected and analysed in a meaningful way for application in fisheries management and science? 2) Has there been discernible change in the relative abundance of selected stocks between the 1960s and the 2000s based on fishers’ perceptions? 118

Chapter 5

3) Is the “shifting baseline syndrome” (SBS) evident or prevalent in marine resource users of Brunei? 4) Where are the high-pressure harvest zones and are there any conflicts between different fishery sectors?

5.2 Materials and Methods 5.2.1 Data collection Opinions of large-scale (LS) and small-scale (SS) fishers of different ages on Brunei fisheries resources were assessed and quantified through a semi-structured mapbased interview survey. The survey was carried out in Malay, which is the local language spoken in Brunei, without the need of a translator.

5.2.1.1 Interview preparation Prior to undertaking the fieldwork in Brunei, background information was collected on the people and culture within the study area. This was achieved through internet searching and the use of published and some unpublished references on fishers’ fishing practices and lifestyles. Interview questions were developed in consultation with Professor Andrew Price (School of Life Sciences, University of Warwick) and Dayang Ranimah Haji Abdul Wahab (then Head of Marine Fisheries Development and Management Division, DOF) and were pilot-tested on 65 fishers during an initial fieldtrip in Brunei in May 2010. Issues that had arisen from the pilot survey were identified and interview design and procedures were refined accordingly. This enabled better preparations to be made for the second fieldtrip which took place over a longer time-frame between March and May 2011.

119

Chapter 5

The survey was carried out with the help of two research assistants. Prior to undertaking the survey, the research assistants were briefed on the research objectives as well as the interview procedures to be employed during the survey. Ethical and technical standards were followed in accordance with (Bunce et al., 2000).

5.2.1.2 Interview design and procedures The interview questions developed are modifications made on questionnaires from previous studies focusing on the use LEK to detect SBS (Saenz-Arroyo et al., 2005b; Ainsworth et al., 2008; Venkatachalam et al., 2010). As few technical terms as possible were used to make the questions comprehensible to respondents with varying education levels. In addition, local fishery officers were consulted to help with appropriate word usage and clarify any technical or difficult terms. Interviews generally lasted between twenty minutes and one hour.

During the oral portion of interviews, the researchers would first chat informally with respondents to help establish a relaxed atmosphere conducive to openness and frankness. Interview questions were asked either as part of the dialogue during the conversation or as direct questions once a good rapport had been established. Whenever possible, interviews were conducted in private to ensure respondents were not influenced by responses of others (Figure 5.1).

120

Chapter 5

Figure 5.1: Obtaining local ecological knowledge from fishers in Brunei. Whenever possible, interviews were conducted in private, either on a one-on-one or two-on-one arrangement, to ensure respondents were not influenced by responses of other fishers.

In this survey, the questions were divided into three main sections (Appendix 5.1). The first section of questions was designed to be easy to answer, to allow the respondents and researchers to settle into the interview dialogue. Examples of information requested in this section included age, number of years the respondent had fished for and type of fishing gears commonly employed in their fishing operations. The second section involved questions on perception of changes of Brunei marine resources since the respondents first started fishing. During pilot interviews, respondents were originally asked an open-ended question of naming all the fish species that they believed to have been depleted by fishing and the associated year or time period. However, it became apparent that while verbally fluent respondents provided very specific answers, the majority of respondents provided minimal responses and often replied with “I don’t know” or “All species”. Consequently, feedback was better when the question was structured to be closedended, although the varieties of species studied were compromised. Hence, in addition to naming all the fish species they believed to have been depleted throughout their fishing career, the respondents were also asked to characterise the abundance of six selected stocks (i.e. red snappers, groupers, ponyfishes, scads, mackerels and rays) into one of the three categories (high, medium, low) for each 121

Chapter 5

time period: 1960s, 1970s, 1980s, 1990s and 2000s. These stocks were selected as they can be easily identified by all fishers, whether old or young. Respondents were also asked to score their perceived changes in general fish abundance and fish size throughout their whole fishing career. In the last section, questions were structured to extract LEK specifically related to the red snapper stocks, to further explore the extent of SBS in Brunei. Questions in this section included maximum size and best day’s catch of red snapper.

The survey also incorporated map-based interviews (Close and Hall, 2006), where respondents recorded on an individual hardcopy map, with the help of the researcher, their approximate most frequently visited fishing site and previous harvesting locations for red snappers. To check for logical consistency of fishing areas drawn, fishers were asked again to describe the fishing location verbally and in relation to notable oil fields, islands, shorelines or reefs nearby (Appendix 5.2). A nautical chart of Brunei waters was provided by the DOF to be used in the map-based portion of the interview.

122

Chapter 5

5.2.1.3 Survey sampling strategy The survey can be divided into two types; (i) LS fishery survey, and (ii) SS fishery survey. The LS fishery survey was conducted over two weeks, from 28th March to 8th April 2011, while the SS fishery survey was carried out between 21st March and 5th May 2011. In general, all respondents were cooperative as the researchers encountered only two refusals.

(i)

LS fishery survey Skippers of LS vessels (trawlers, purse seiners and long-liners) harboured at Muara FLC were identified with the help of DOF staff, Awang Bidin bin Suru. For 2011, there were a total of 43 LS vessels (21 trawlers, 12 purse seiners and 10 long liners) registered with DOF. However, during the two-weeks surveying period, only twenty trawlers, six purse seiners and three long liners were actively operating, and only 23 skippers managed to be interviewed.

(ii)

SS fishery survey A stratified convenience sampling approach, either by roving at accesspoints or by contacting key informants identified from the official registration list for the whole of Brunei (provided by DOF), was used to sample the SS fishermen population. Four access-points were selected out of several known landing sites, based on site inspection visits during the first fieldtrip. The four access-points were Jerudong landing, Pengkalan Sibabau landing, Kuala Tutong landing and Kuala Belait landing (Figure 5.2). 123

Chapter 5

FIGURE 5.2: Map showing four access-points (big red circles) which were selected out of several known landing sites (small yellow circles) for the interview survey of fisheries in Brunei. Villages in which the respondents currently reside are shaded in pink.

To ensure the interview survey was representative and generated data with sufficient statistical power, a minimum effective sample size was calculated (Raosoft, 2004). Based on the total number of fishermen engaged in the capture fishery in 2010 (c. 2527 full-time and part-time fishermen; DOF pers. comm.), 246 respondents were required for the fishermen population to be representative at 90% confidence interval and an accepted error of 5%. However, as the fieldwork progressed, it became apparent that having a valid fishing gear license does not guarantee an active fishing operation by the license holder. Allowing for non-compliance, a total of 236 respondents were obtained. Therefore, the sample size was representative at 90% confidence interval with an accepted error of 5.1%.

124

Chapter 5

The total number and distribution of SS fishers interviewed for the whole of Brunei are summarised in Table 5.1. TABLE 5.1: Distribution of small-scale fishers interviewed between March and

May 2011 in Brunei. District - Villages Brunei Muara -

Tutong -

Batu Marang Bebatik Kilanas Bengkurong Jerudong Kapok Kiulap Lambak Kanan Mata-mata Mentiri Muara Pandai Besi A Pengkalan Sibabau Pudak Serasa Serdang Setia A & B Sungai Besar Sungai Bunga Sungai Kebun Sungai Siamas Tamoi Tengah Tungku Bukit Beruang Danau Keramut Kuala Tutong Penanjong Sengkarai Telisai

Small-scale companies

Small-scale individuals

Total SS fishers

53

85

138

10

25

35

-

13

13

17

33

50

80

156

236

Temburong -

Belait -

Amo A BatangTuau Belais Menengah Negalang Ering Rataie Kuala Belait Lumut Mumong RPK Pandan Sungai Liang Sungai Teraban

Total SS fishers

125

Chapter 5

5.2.2 Data assembly and analyses Data assembly and analyses were done using Microsoft Excel, SPSS v.19 and ArcGIS 10.1 software. Results from the interview survey were organised into two main groups, forming either the attribute dataset or spatial dataset. Quantitative attribute data collected were tested and analysed using standard non-parametric tests where necessary, as used in previous LEK studies (Saenz-Arroyo et al., 2005b; Ainsworth et al., 2008; Venkatachalam et al., 2010). In addition, a subset of these data was subjected to fuzzy logic analysis using methodology adapted from Ainsworth et al. (2008). Qualitative attribute data were transcribed, arranged by key themes and visually assessed. The spatial dataset, which contains the map portion of the interviews or, more precisely, specific spatial information from the interview, were input from the hardcopy maps as ArcView GIS shapefiles using “heads-up” digitizing as described by Close and Hall (2006).

5.2.2.1 Fishers’ perceptions of fish depletion and changes in indicator species (Lutjanus erythropterus) – descriptive approach Respondents were asked the year they started fishing actively, which was later placed into one of the three ‘experience years’ groups (<10, 11 – 39 and >40 years) for comparison. Differences in variables (other than the red snapper catch variables), between the groups were determined using non-parametric Kruskal-Wallis test. Each of the red snapper catch response variables (i.e. best day’s catch, weight of biggest fish ever caught, and distance offshore and water depth associated with best day’s catch) was regressed against fishing experience years using linear regression analysis by using transformed data to meet the assumptions of normality and heterogeneity of variance tested. Besides comparing by fishing experience, separate analysis was

126

Chapter 5

undertaken on the catch response variables to determine strength of association with years.

Reported abundance categories were converted into the same numerical interpretation of responses where “low”, “medium” and “high” score as -1, 0 and 1, respectively. Trends in perceived change of stock abundance from when fishers started fishing and present were then assessed – a logistic regression analysis was used to model the probability of fishers perceiving ‘no change’ or ‘some change’ against their fishing experience.

5.2.2.2 Estimating past abundances from LEK - Fuzzy logic approach Fishers’ responses were subjected to fuzzy logic analysis using methodology adapted from Ainsworth et al. (2008), whereby perceptions of fish abundance were categorized into fuzzy datasets so that the threshold between abundances categories is not crisp, but instead adjusted to a gradation of membership in several abundance categories.

When fishers scored the abundance as ‘high’, this corresponded to a numeric abundance score of 1, ‘medium’ corresponded to 0.5 and ‘low’ corresponded to 0. The average score was taken for all the contributing interviews, and partial memberships in five fuzzy set linguistic abundance categories (i.e. low, medium-low, medium, medium-high and high) were determined using the membership functions such as that shown in Figure 5.3. Agreement between fishers was also used as an indicator of data quality, by changing the shape of the membership functions according to the variance among fishers’ scores. If fishers agreed well with each 127

Chapter 5

other, the membership functions more closely resembled Figure 5.3a. When there was a high variance among fishers’ abundance scores, a wider form of membership functions as in Figure 5.3b was used. This enabled the membership of the abundance score in as many as four overlapping abundance categories. Consequently, this approach enabled the influence of a small number of interview comments that contradicted with majority response to be ignored.

Variance (V) among abundance scores assumed a binomial distribution by having the majority abundance category as marked by fishers (low, medium or high) as ‘correct’ and all other estimates as ‘incorrect’. Variance was calculated as V = np(1-p), where p is the fraction of responses indicating the ‘correct’ response and n is the total number of responses. Membership for the fuzzy abundance categories was calculated following Ainsworth et al. (2008) given below:

FIGURE 5.3: Fuzzy set membership for fisher abundance scores. Memberships in the low (L) and high (H) linguistic abundance categories are defined by trapezoidal functions; memberships in medium-low (ML), medium (M) and medium-high (MH) categories are defined by triangular functions. X-axes represent average abundance scores from interviews for time period and species. As variance among fisher responses increases, the subtended angle increases from a minimum (left) to a maximum (right). Taken from Ainsworth et al. (2008).

128

Chapter 5

The partial memberships for each species group and time periods are then combined through the ‘de-fuzzification’ process – a process by which the range of possible conclusions for a particular species group-time period element is reduced to a single point output representing the relative abundance. This was done using the centroid weighted average approach (Cox, 1999). The weightings (partial memberships) were multiplied with the centroid of the corresponding abundance category in the defuzzification membership function – which is identical to Figure 5.3 – to obtain a weighted average of the abundance output categories. The lower and upper error range was established by multiplying the weightings in each abundance category by the lower and upper values of the membership function, respectively, as opposed to the centroid for the weighted average.

129

Chapter 5

5.2.2.3 LEK assessment using spatial information technology – GIS approach Each fishing area drawn by fishers onto the individual hardcopy map was constructed using careful on-screen digitization and stored as ArcGIS shape files, following Close and Hall (2006). Fishing area may be identified as points or lines on the base (hardcopy) map. Thus, the construction of point and line layers in ArcGIS reflected two forms of fishing activity. Lines were used to represent fishing paths along shorelines, between oil platforms or over reefs, where a fisher would start fishing from one end and work his way to the other end, or move from location to location in a linear fashion, often drifting with the current and/or wind. Points represent a particular area of fishing activity. Attribute entry that were acquired indirectly from the spatial dataset (i.e. distance and depth of location associated with best day’s catch of red snappers) were linked to the main attribute dataset.

Raw spatial data (Appendix 5.3) and associated attribute data were then converted into a more meaningful form that fisheries managers and planners could use for planning. To achieve this, arbitrary ‘likelihood’ surfaces were created using a combination of GIS-based map overlaying and buffering of features on individual fishers’ map layers (Close and Hall, 2006). In this study, the ‘red snapper harvesting location’ (points feature class) and ‘most frequently visited location’ (lines feature class) data were used for the construction of two types of surface classifications, namely the ‘red snapper distribution classification’ and ‘high-pressure zone classification’. These two classifications were used to illustrate red snapper distribution across the study area and areas that receive a high degree of fishing pressure. The general analysis sequence for the two classifications is shown in Figure 5.4. 130

Chapter 5

GIS database  Harvest_points  Frequent_lines

Query ‘Fishery sector’

SS fishery  Harvest_points  Frequent_lines

LS fishery  Harvest_points  Frequent_lines

Buffer [email protected] m

Buffer [email protected] m

Union

Harvest_points

Frequent_lines

Red snapper distribution classification

High-pressure zone classification

KEY Single data layer

Multiple data layer

Pre-defined process

FIGURE 5.4: Process flow for the construction of the two classifications using GIS, namely the red snapper distribution classification and the high-pressure zone classification. ‘Harvest_points’ refer to red snappers best day’s catch harvesting sites which are points feature class. ‘Frequent_lines’ refer to fishers’ most frequently visited area when the go out in the sea, which are lines feature class.

131

Chapter 5

The multi-buffer approach was employed to adjust for map bias and boat drift (Close and Hall, 2006). Although there is no scientific basis for determining the number of buffers and their absolute or relative widths, seven buffers of 100 m and 250 m, for SS small boats and LS large vessels, respectively, were used in this study. These values were based on the observed maximum distance of lateral boat drift of typical fishing boats used by the fishers (i.e. boats are ~8 m long for SS fishery and ~20 m long for LS fishery), Brunei’s generally moderate weather (Panaga, 2007) with slight wave and gentle breeze (except during the Northeast monsoon), and the relatively shallow bottom depth of main fishing area. Furthermore, expert opinions of the DOF officers were consulted, and published literatures were reviewed, to ensure that the buffers yield a realistic representation of the harvest area. The cumulative buffer ‘likelihood’ values, or ‘scores’, were summed in the union of individual fisher’s maps (224 layers for the red snapper harvest location and 236 layers for the most frequently visited sites) such that the highest possible score would occur in locations with the highest chance of red snapper being present and locations of the highest fishing pressure.

Proportional symbol maps were also used to uncover the spatiotemporal variation in fishers’ best day’s catch.

132

Chapter 5

5.3 Results 5.3.1 Exploratory analysis: distribution of small-scale fishers by gears Most (77%) of SS fishers own either one or two gears (Figure 5.5a) which usually include gill net and/or hand line. Major gears operated by the SS sector are fish gill net (andang ikan), hand line (jaul) and portable fish trap (bubu) (Figure 5.5b). Accordingly, most of the gears are passive and selective, and about 40.9% of SS fishers employ gears specifically targeted at the demersal stocks.

Number of type of gears

SS(company) SS(individual)

Lintau (trap) 0.92%

Kabat (trap) 0.46%

Kilong (trap) 0.23%

Bubu (trap) 12.47% Bintur (trap) 1.85%

5 4 3

Rawai (line) 2.54%

2

Jaul (line) 24.71%

1 60 50 40 30 20 10 0 10 20 30 40 50 60 70 Number of SS fishers

Rantau (net) 1.39%

Tugu (trap) 0.92% Others 1.85%

Ancau (net) 3.23%

Andang ikan (net) 41.34%

Andang Rambat udang (net) (net) 0.69% 7.39%

FIGURE 5.5: Distribution of gears among small-scale fishers interviewed in Brunei between March and May 2011 (a) Frequency of fishers that own between one and five different types of gears, and (b) Proportion of the different types of gears operated by overall SS fishers.

133

Chapter 5

5.3.2 Perceptions of decline in indicator species (Lutjanus erythropterus) and changes in selected stocks from fishers’ interviews The majority of fishers interviewed (90.7%) agreed that fishing had led to depletion or loss of some species. Altogether, respondents cited a total of 80 species which they perceived had been depleted in their lifetime, with 18 species being cited by all groups (Appendix 5.4).

Red snapper was cited more than any other species as depleted, with the less experienced fishers being less likely (77%) than more experienced fishers (91%) to have caught one (Kruskal-Wallis test; χ2 (d.f.) = 9.40 (2), p = 0.009). Highly experienced fishers were more likely to have caught larger and more red snappers on a trip nearer to the shore, where it was shallower, than further offshore (Table 5.2, Figure 5.6; linear regression analysis, p < 0.05 in all cases).

Because experienced fishers would likely have a greater chance of landing larger and bigger catches than the newer/younger fishers, catch parameters were then assessed against actual years (Table 5.3, Figure 5.7). Both best catch and weight of largest red snappers caught have decreased according to fishers’ perceptions. Additionally, both the depth and distance offshore associated with fishers’ best catch have been greater in recent years (linear regression analysis, p < 0.05 in all cases).

134

Chapter 5

Log-transformed weight of largest fish (kg)

Log-transformed best catch (kg)

8 6 4 2 0 0

20

40

60

2 1 0 -1

80 Log-transformed depth of best catch (m)

6 Log-transformed distance from shore of best catch (km)

3

4 2 0

-2 -4 0

20

40 Fishing experience (years)

60

80

0

20

0

20

40

60

80

40

60

80

5 4 3 2 1 0 Fishing experience (years)

FIGURE 5.6: Linear regressions of interview data showing (a) best day’s catch, (b) weight of largest red snapper, (c) distance offshore associated with best day’s catch, and (d) water depth associated with best day’s catch, against fisher experience. TABLE 5.2: Results of linear regression analyses of four different red snappers catch variables as a function of fishing experience, as obtained from fishers interviews. Significant p-values are in bold. Attribute (log-transformed) N Slope b (S.E.) t-test stats p-value Best day’s catch (kg) 226 0.011 (0.006) 2.036 0.043 Largest red snapper ever caught (kg) 225 0.010 (0.002) 4.143 <0.001 Distance offshore associated with best day’s catch (km) 224 -0.014 (0.005) -2.693 0.008 Depth associated with best day’s catch (m) 206 -0.012 (0.004) -3.268 0.001

135

8

3

Log-transformed weight of largest fish (kg)

Log-transformed best catch (kg)

Chapter 5

6

2 1

4

0

2

-1

0 0

2

3

0

4

Log-transformed distance from shore of best catch (km)

Log-transformed depth of best cathc (m)

6

1

4 2 0

-2 -4

1

2

3

4

5 4 3 2 1 0

0 1 2 3 4 1 2 3 4 Log-transformed years before present Log-transformed years before present FIGURE 5.7: Linear regressions of interview data showing (a) best day’s catch, (b) weight of largest red snapper, (c) distance offshore associated with best day’s catch, and (d) water depth associated with best day’s catch, against reported year. 0

TABLE 5.3: Results of linear regression analyses of four different red snappers catch variables as a function of actual years, as obtained from fishers interviews. Significant p-values are in bold. Attribute (log-transformed) N Slope b (S.E.) t-test stats p-value Best day’s catch (kg) 226 0.365 (0.079) 4.619 <0.001 Largest red snapper ever caught (kg) 225 0.110 (0.034) 3.218 0.001 Distance offshore associated with best day’s catch (km) 224 -0.214 (0.078) -2.759 0.006 Depth associated with best day’s catch (m) 206 -0.117 (0.054) -2.154 0.032

136

Chapter 5

Figure 5.8 shows the distribution of perceived changes for the two depletion indicators (abundance decrease and size reduction) by fishing experience categories. In general, the majority of fishers agreed that overall fish abundance had declined (79.9% of all fishers), but with no perceived reduction in fish size (72.2% of all fishers) since fishing began.

100%

100%

+ 50%

0 -

0% < 10 years 11 - 39 years > 40 years

Proportion of fishers

(b) Fish size

Proportion of fishers

(a) Fish abundance

+

50%

0 -

0% < 10 years 11 - 39 years > 40 years

FIGURE 5.8: Proportion of fishers reporting changes in depletion indicators; (a) fish abundance, and (b) fish size, since when they first started fishing. Dark blue bars indicate decrease (-), blue bars indicate no change (0) and light blue bars indicate increase (+).

The perceived abundance change from when fishers started fishing and present using the reported abundance scores (“low”, “medium” and “high” score as -1, 0 and 1, respectively) for the six species were tested against experience years (Figure 5.9). The Y-axis represents the probability of fishers perceiving changes in the selected stocks. All fishers acknowledged abundance to either decrease or remain the same, when present abundance was compared to the abundance when fishers first started fishing. However, only Japanese scads, narrow-barred Spanish mackerels and rays showed a significant change in perceptions (Table 5.4), whereby experienced fishers acknowledged a greater depletion since when they first started fishing, compared to less experienced fishers. A significant trend in this figure, however, does not indicate an abundance decrease, only that there is disagreement between fishers of differing fishing experiences. 137

Chapter 5

Red snappers

Groupers

Probability of fishers perceiving changes in the selected stock (0 – no change, 1 – some or a lot of change)

1

1

0

0 0

20

40

60

80

Ponyfishes

0

1

1

0

0 0

20

40

60

Narrow-barred Spanish mackerels

1

0

40

60

80

Rays

0

80

20

20

40

60

80

60

80

Japanese scads

1

0 0

20

40

60

80

0

20

40

Fishing experience years

FIGURE 5.9: Evidence of shifting cognitive baselines in Brunei fishers. Y-axis shows the probability of fishers perceiving some change in abundance between when they first started fishing and present; X-axis shows fisher experience. Solid lines represent significant logistic regression model. ‘Some or a lot of change’ corresponds to a reduction from ‘high’ to ‘low’, ‘high’ to ‘medium’ or ‘medium’ to ‘low’ category of abundance in the interview, and presented as 1 in the logistic regression. TABLE 5.4: Effect of fishers’ experience (β) on their perceived change in abundance between the time when they first started fishing and present, quantified using logistic regression analysis. Significant p-values are in bold. βexperience Standard Species group coefficient Wald χ2 test d.f. p-value errors in model Red snappers 0.018 0.012 2.313 1 0.128 Groupers 0.009 0.010 0.769 1 0.381 Ponyfishes 0.012 0.008 1.910 1 0.167 Rays 0.024 0.008 8.910 1 0.003 Narrow-bared Spanish 0.032 0.010 9.498 1 0.002 mackerel Japanese scads 0.041 0.011 12.840 1 <0.001

138

Chapter 5

5.3.2 Change in abundance of selected stocks from fuzzy logic output Change in relative abundance determined from the fuzzy expert system for all six species groups are presented in Figure 5.10. Despite the wide confidence intervals, there is a consistent trend of sudden decline in relative abundance in the 2000s. This is in agreement with published surveys which have shown that Brunei fishery was lightly exploited up till end of the 1990s (see Discussion, Section 5.4.2). Abundance scores from the fuzzy expert system revealed that the abundance of both snappers and groupers species declined the most compared to other species (relative score in 2000s < 0.3). Red snappers

Groupers

1

1

0.5

0.5

0

0

-0.5

-0.5

-1

-1 1960s

1970s

1980s

1990s

2000s

1960s

1970s

1980s

1990s

2000s

1990s

2000s

Rays

Ponyfishes 1

1

0.5

0.5

0

0

-0.5

-0.5

-1

-1 1960s

1970s

1980s

1990s

2000s

1960s

Narrow-barred Spanish mackerels

1980s

Japanese scads

1

1

0.5

0.5

0

0

-0.5

-0.5

-1

1970s

-1 1960s

1970s

1980s

1990s

2000s

1960s

1970s

1980s

1990s

2000s

FIGURE 5.10: Relative abundance of selected stocks in Brunei over five time periods from fuzzy logic algorithm. Zero on the Y-axis corresponds to the centroid of the ‘medium’ abundance category in the defuzzification membership function. Error bars show weighted average based on upper/lower range of membership function, rather than centroid.

139

Chapter 5

5.3.3 GIS outcomes as indicators of spatial and temporal patterns of red snapper distributions and fishers’ fishing activities The distribution of red snapper based on fishers’ LEK is shown in Figure 5.12, which revealed that patches of high presence possibility were associated with areas of higher complexity, where reefs, shipwrecks and oil platforms are found. These areas are mainly present in the shallow waters of Brunei, which corresponded to Zone 2 of Brunei fishing zonation. Accordingly, these reported harvesting sites are located at depths between 1 m and 111 m, and were significantly clustered (average nearest neighbour test; nearest neighbour ratio = 0.516, z-score = -13.85, p-value = <0.001) throughout the study area. The majority (9.7%) of the reported red snapper harvesting locations occurred within 3 km of Lumut reefs (Figure 5.11).

20 18

Number of fishers

16 14 12 10 8 6 4 2 Lumut Patches Victoria Patches Champion Shoal Browne Patch Pulau Pelumpong Takat tiruan 3 Mampak Patches Two Fathom Pelong Rock Australian Wreck Takat tiruan 2 Takat tiruan 1 Ampa Patches Brunei Patches Colombo Shoal Abana Rock American Wreck Fairley Patches Littledale Shoal Chearnley Shoal Iron Duke Shoal Otterspool Rock Scout Rock Cunningham Patch Pulau Punyit Silk Rock

0

Nearby reef/shipwreck

FIGURE 5.11: Nearby major natural / artificial reefs or shipwrecks associated with the reported red snapper harvesting location in Brunei waters.

140

Chapter 5

FIGURE 5.12: Mapping LEK. Distribution of red snapper presence scores based on fishers’ LEK in Brunei (i.e. harvest location of best day’s catch).

141

Chapter 5

FIGURE 5.13: Spatiotemporal pattern of fishers’ best day’s catch in Brunei for: (a) all data combined; (b) 1960s-1970s; (c) 1980s-1990s; (d) 2000s; and (e) 2010-2011.

142

Chapter 5

Red snapper distribution can be grouped into four main areas of unequal size: (1) Brunei Bay, (2) waters off Brunei-Muara district to just east of Kuala Tutong, (3) waters off Belait district as far as Kuala Tutong, and (4) deeper waters off the continental shelf (Zone 3). Mean best day’s catch differed between the four main areas over the years (two-way ANOVA; F8,215 = 5.915, p < 0.001), with a relatively higher best day’s catch occurring in waters off Belait district (Figure 5.13; post-hoc LSD test, p < 0.05 for Area (3) against all other areas). Eight fishers recalled catching ≥ 1000 kg of red snappers on their best day’s catch – the majority (62.5%) of this occurred near reef patches found in the waters off Belait district.

Figure 5.14 shows the results of the high-pressure harvest area (HPA) analysis. Although fishers operate over a wide-ranging area, the fishery essentially focuses on the eastern side of Brunei waters. The calculated potential harvest area for fishers interviewed in this study was 4 588.5 km2, encompassing 53.4% of Brunei’s continental shelf area of 8 600 km2 (or 11.9% of Brunei’s total marine territory area of 30 000 km2).

143

Chapter 5

FIGURE 5.14: Distribution of high pressure harvest area (HPA) based on 236 fishers’ most frequently visited sites. The proposed marine protected area (MPA) will be a ‘no take’ MPA, to be implemented soon. (Note: Based on a press statement released by DOF in September 2011, the MPA would be fully operational by January 2012, although currently (April 2013) it has yet to be implemented).

144

Chapter 5

Summary statistics of HPA scores are shown in Table 5.5 and Figure 5.15. The distribution is right-skewed, with more values in the lower range, suggesting a relatively dispersed pattern of fishing activity, as would be expected in a mixed fishery deploying multiple gears. If the fishery had only a very small number of locations where fishers visited frequently, then a left-skewed distribution of HPA scores would have been expected, since fishers would be visiting the same locations. Median and upper quartiles of HPA scores were used as a threshold to identify the highest-pressure harvest zones, which covered an area of 233.29 km2 for scores >36 (HPA36) (2.7% of Brunei’s continental shelf area) or 77.72 km2 for scores >54 (HPA54) (0.01% of Brunei’s continental shelf area).

69.1% (828 km2) of the potential harvest area was found to fall within the newly proposed no-take marine protected areas (MPAs) of Brunei waters, of which 8.8% and 2.6% were HPA36 and HPA54, respectively.

Frequency of value occurrence

14000

TABLE 5.5: Basic statistics of HPA scores

12000

Measure: Range Mean Median Mode Std. dev.

10000 8000 6000

1-153 41.42 36 21 26.03

4000 2000 0 1

20

39

58

77

96

115

134

153

Range of summed values from buffer overlay

FIGURE 5.15: Range of HPA scores generated for Brunei waters using 236 fishers’ most frequently visited sites.

145

Chapter 5

5.3.4 Qualitative assessment of fishers’ opinions on current fishery management in Brunei Fishers’ comments on current management practices for Brunei marine resources can be divided into two main groups; approximately 20% of fishers gave either a short positive comment (e.g. “everything OK”) or provided elaborate positive viewpoints (e.g. “it is good that the local authority have deployed artificial reefs in our waters”). The majority (65.6%), however, had negative comments on the local fishery authority and current management practices. Fisher comments can be further divided into eight themes, summarised in Table 5.6. TABLE 5.6: Fishers’ negative comments on the local fishery authority and current management practices in Brunei were assessed qualitatively and divided into eight themes. Note that the proportion of interviewed fishers will not add up to 100%, since a fisher may comment on more than one issue.

Themes Gear conflicts Poor enforcement Lack of communication (difficulty in communicating with DOF) Unfair treatments (for SS fishers) Poorly managed artificial reefs Poorly maintained/Lack of landing sites Licensing issues Others: Lack of fish price control Conflicts near borders Demand subsidies

Proportion of interviewed fishers (out of all 259 fishers) 33.2% 18.1% 14.3% 14.3% 11.2% 8.9% 2.7% 3.5%

146

Chapter 5

5.4 Discussion and Conclusion This study presents the first quantified evidence of the SBS based on fishers’ perceptions in Brunei. It underlines the importance of tapping into LEK, which can make a valuable contribution to the body of scientific knowledge concerning marine stock status. The systematic collection of LEK of fishers in Brunei has generated useful guidelines for compiling such knowledge, and has also demonstrated a perceived decline in abundance of marine resources in Brunei, especially in the last decade. More importantly, this study also serves as a digital archive for an important knowledge resource which would otherwise be lost if it was never assembled (Johannes, 1998).

The sample size (259 fishers) used in the interview survey is considered to have provided a fair representation of fishers views in Brunei, representing just over 10% of the total number of recorded fishers in Brunei in 2010.

5.4.1 Extracting Brunei’s local ecological knowledge on fishery In this study, level of fishing experience (measured in years spent fishing), rather than fishers’ age is used as an explanatory variable. Although both variables are highly inter-correlated, fishers’ experience was considered a better indicator, especially in the context of Brunei’s fishing community structure. In Brunei, the majority of young fishers are foreign labourers employed specifically to fish, while some of the older fishers are civil servant/army retirees who may have only started fishing recently, may not have a great level of fishing experience and/or fishing may not always have been their livelihood. 147

Chapter 5

The Code of Conduct for Responsible Fisheries, formulated in 1995 by FAO, highlights the need to investigate and document traditional fishery knowledge and to assess its application to the sustainable conservation, management and development of fisheries (Article 12.12). Brunei has a long fishing tradition and European records as far back as the sixteenth century attest to the significance of fisheries as a socioeconomic activity in Brunei (Silvestre and Matdanan, 1992). However, due partly to the favourable economic circumstances of a high GDP ever since the 1970s, the total number of young local fishers is declining due to more comfortable, land-based alternatives (Silvestre and Matdanan, 1992). In an earlier study of LEK, Pilgrim (2006) pointed out that as level of economic development increases and resource dependence of a community decreases, the rate of knowledge acquisition tends to diminish, thus resulting in a progressive loss of LEK in the younger generations. Therefore, in areas where few written records are kept, important knowledge about natural resources may die with each generation unless someone records it, since conventional biological field research is unlikely to recover this information.

One example of such knowledge is on the types and numbers of species mentioned as depleted by the fishers. Just as in Rodrigues, Mauritius, where there were some species of fish about which the younger generations in the fishery community had little or no knowledge (Bunce et al., 2008), about half of all species in Brunei mentioned as depleted in this study had not been identified by the least experienced fishers. These findings, in accordance with other studies (Johannes and Yeeting, 2001; Dulvy and Polunin, 2004; Saenz-Arroyo et al., 2005b), demonstrate that interviews with older or more experienced fishers can be a useful means of identifying declining species and confirming the disappearance of exploited species, 148

Chapter 5

potentially in time for conservation action. Also extracted from the interview surveys were traces of traditional management systems mentioned by experienced, older fishers. These fishers, all from the traditional fishing village of Kampung Setia B, commented on the disappearance of the customary practice of ‘bagi makan laut’ (literally means ‘letting the sea eat’) usually carried out by a sharif (a hereditary title for Arab descendants, traditionally believed to possess great religious knowledge and hence, the expertise to perform rituals to safeguard the sea from bad omens). The time to bagi makan laut was usually determined by village elders who would have extraordinary traditional knowledge of biophysical and biological resource characteristics. During this period, fishers were prohibited to fish, and such rules were easily enforced, as fishers believed them to be sacred. The underlying myth was in fact based on scientific rationales, as seasonal closures are known to enhance fish stocks (Hilborn and Walters, 1992). Although the practices may not be as explicit as those observed in many oceanic island communities, such as in North Lombok, Indonesia (Arif, 2007) and in Vanuatu (Hickey, 2007), the practical outcomes of these beliefs and values may have induced an effective system for the sustainable management of marine resources. An important point, however, is that such wisdom and knowledge had been well-maintained because it was handed down through generations. However, with the loss of traditional LEK, such practices are becoming rapidly eroded, and in some places stocks have dwindled (Johannes, 1981a; Mulrennan, 2007).

Interview results from this study also highlight potential difficulties in composing suitable questions for extracting LEK by interviewing fishers. First, the interview design had to be simplified because, generally, the initial question sequence was too 149

Chapter 5

long to retain interest and attention of respondents. Moreover, some of the fishers were approached just as they arrived from the sea and, while it may be preferable to conduct the interviews at times that are most convenient to the fishers (e.g. at fishers’ home), logistical constraints precluded more extensive survey. In any interview survey, trade-offs between number of questions asked, number of interviewees consulted and survey duration are inevitable (Bunce et al., 2000; Venkatachalam et al., 2010).

Secondly, a common problem when conducting interviews is that fishers may interpret questions quite differently from the way they were intended by the interviewer (Meeuwig et al., 2007). In this study, fish stocks that were ecologically classified as demersals were poorly defined, so when fishers were asked about the changes in abundance of demersal fish specifically, they were referring to changes in abundance of fish in general. The results can be easily misinterpreted if care is not taken to consider the answers in the light of fishers’ perceptions and experience.

Consequently, it is also important to cross-reference interview results against other sources of information. For instance, one important body of LEK, particularly in tropical waters, concerns reef fish spawning aggregations (Morris et al., 2000). Although not part of the key questions, and hence not probed further during the interviews, the outcomes pertaining to fishers’ harvest sites for best day’s catch of red snapper in the interview had implied a spawning aggregation in the waters northeast of Southwest Ampa oilfield, where fishers reported red snappers catch of up to 2000 kg in this area (Figure 5.12). Extreme reports of high catches may have created suspicion of exaggeration by the respondents. However, when the data were 150

Chapter 5

presented as number of red snappers (e.g. assuming that the total weight of 2000 kg could result from 505 aggregating individual snappers at 3.96 kg (mean from this study) each), the fishers’ answers fell within the range previously reported for Brunei (see Chou et al., 1992). Adding to this belief may be the type of gear used when these catches were made – fishers that reported their best day’s catch of ≥1000 kg all recounted using either a purse seine (pukat lingkong) or a ring net (ancau); both are ‘surrounding nets’ based on FAO’s definition and classification, a roughly rectangular net without a distinct bag that is set vertically in water to surround the school of fish, although generally of pelagic nature. Possibly, these fishers had encountered red snapper spawning aggregation, since reef fish species such as snappers and groupers are well known to form spawning aggregation at the same location, season and moon phase each year (Arreguín-Sánchez et al., 1996). Like other small, local stocks, many of these aggregations are highly vulnerable to rapid depletion, or complete elimination (Johannes et al., 1999; Morris et al., 2000). Considering the generally limited knowledge on fish reproduction in the scientific literature, clues from fishers concerning spawning periods and potential aggregation sites could potentially be useful for either defining periods of closed fishing activities or designing MPAs (Johannes and Neis, 2007). Further research could be undertaken in order to confirm these potential spawning sites, especially since these sites are not included in the newly proposed MPA.

In fact, research on the spatial heterogeneity of ecosystems for fishery management has typically focused on biological processes of the environment. The network of MPAs for Brunei proposed by DOF just over a year ago resulted from three comprehensive marine biodiversity assessments (DOF-pers.comm.). Meeting the 151

Chapter 5

challenge of place-based ecosystem management such as MPAs, however, will require an improved understanding of the human use patterns in the marine environment, by highlighting areas of intense use or areas where multiple activities are occurring (Moreno-Báez et al., 2012). This study uncovers potential fishing ‘hot spots’, visualized in map form and this information is neither extremely difficult nor time consuming to produce, yet serves as a potentially important complement to the collection of scientific data in the area.

This study also confirms the occurrence of territorial and gear conflicts between fishers of different areas and sectors, as predicted by Beales et al. (1982) and identified by Silvestre and Matdanan (1992) and the national fishery authorities (Ranimah, 2008). However, of particular significance, are the potential conflicts between fishers, and scientific, or bureaucratic, perspectives on marine resources, as more than half of the total area legally established for the new no-take MPAs is used frequently by current fishers.

Although differences in perspective are common in fisheries (Gray et al., 2008), the current study revealed that fishers had more optimistic perceptions of stock abundance than suggested from analysis of official scientific data. While this is not surprising considering the spatial and temporal dynamics of fishers’ exploitation activities and the way these influenced their observations and related perceptions (Murray et al., 2008), existence of such divergent view itself may create problems whereby fishers may not be supportive of introducing effort or catch controls, or the limited entry to the fishery. Lack of trust between fishers and the national fishery institution was certainly evident from the interview survey – one of the main themes 152

Chapter 5

identified from qualitative analysis of interview data was a lack of effective communication between fishers and the fisheries authority. Furthermore, fishers were more hesitant to talk to researchers, perhaps for fear of reprimand or other repercussions (Close, 2003), if they thought that the researchers were working for the fishery department, despite assurances of anonymity and confidentiality beforehand. Evidently, fishers were more willing to freely share their knowledge and voice their views when they were notified that the interviewers were independent student researchers. Just as study of fishers’ knowledge of resource complexities is seen as high priority, and potential means of reducing the likelihood of conflicts, equally important is the public outreach of scientific data to allow fishers and their knowledge to become effective and integrated counterparts in fisheries science and management (Johannes and Neis, 2007). It has been shown in Newfoundland, Canada how fishers began to see how a more ‘scientific’ approach to record keeping could improve their own fishing performance (Murray et al., 2006). Similarly, fishing-rights owners in Vanuatu had willingly sacrificed considerable short-term catch benefits by subscribing to extended moratoria of up to 15 years on fishing for certain species. They were assured that the benefits of making such a sacrifice would accrue to them directly, despite the risk linked to accepting such a long term view (Adams, 1998). Needless to say, such collaboration will depend very much on the degree of trust between fishers and fisheries authorities, including scientists who are expected to fill in the gaps in knowledge about past, present and future state of the fishery (Hilborn and Peterman, 1995).

153

Chapter 5

5.4.2 Evaluating past abundance of Brunei fishery resources One significant advantage of a fuzzy logic approach is that it provides a transparent method for interpreting linguistic descriptions of abundance as perceived by the fishers (Ainsworth et al., 2008). Inevitably in this study, the fuzzy logic algorithm was only used as a method of deriving numerical trends from ordinal LEK data, despite its capacity to integrate various data streams to reconstruct a composite time series of species abundances (Ainsworth, 2011). While the trends shown are relative, at least two other studies had demonstrated that it was possible to develop absolute trends by scaling the relative outputs to match the available partial time series from scientific sampling (see Ainsworth et al., 2008; Lozano-Montes et al., 2008). This was not feasible in the current study, due to the lack of reliable survey information specific to the study area, as well as the lack of permission to carry out a more detailed study using unpublished data collected by the DOF.

By synthesising data contained in anecdotes and fishers’ memories, this study has reaffirmed the relatively high abundance of Brunei fishery resources, at least up to 2000, in the literature. Silvestre and Garces (2004) reported that Brunei fishery resources were still lightly exploited based on the demersal trawl surveys carried out between 1989 and 1990. Using regression modelling, Christensen et al. (2003) has shown that the predicted fish biomass of trophic level ≥ 3.0 in Brunei in 2000 was roughly 80% of its predicted biomass in 1960. In comparison, the predicted fish biomass in the neighbouring countries in 2000 was calculated to be between 7 – 33% of its predicted biomass in 1960 (Christensen et al., 2003).

154

Chapter 5

Estimates of relative abundance of the different species group for the 2000s suggested that large demersal stocks (i.e. snappers and groupers) had declined the most, followed by large pelagics (narrow-barred Spanish mackerel), small pelagics (Japanese scads), small demersal (ponyfishes) and rays. These outcomes seem logical and consistent with the life history information for the different species groups in the literature. For example, the perceived decline in Japanese scads and ponyfishes are less substantial than that of large demersal stocks (i.e. red snappers and groupers), perhaps because both scads and ponyfishes are known small-bodied rspecies which have short generation times that favour rapid adaptation (Munday et al., 2012; FishBase, 2012). On the other hand, the substantial decline in snappers and groupers may reflect the fact that these K-species groups, which generally have restricted breeding ground, are highly vulnerable to fishing and thus could be rapidly over-exploited over a short period of time.

For red snappers, this is further supported in this study by the significant correlation of both the best day’s catch and largest fish ever caught, not only with the period of fishers’ experience, but also with actual years. Worryingly, the real scale of these reductions could also be much greater than reported since large catches remain possible so long as there are still unexploited or lightly fished areas (Saenz-Arroyo et al., 2005a; Saenz-Arroyo et al., 2005b). Remarkably, this trend seems inconsistent with the scientific outcome from official fishery statistics (Chapter 4), which do not point to any discernible resource decline. Yet, there is clear evidence of resource decline using information reported by fishers, which strongly suggests that harvests were higher in earlier decades. Invariably, and perhaps understandably, the fishery authorities rely mainly on fishery catch statistics to inform decisions about stock size 155

Chapter 5

and state of fishery. As this and other studies (Saenz-Arroyo et al., 2005a; Venkatachalam et al., 2010) have shown, however, patterns derived from short time series can be inaccurate. Fishery management decisions that augment scientific data with records drawn on long-term information, even if the only source available is reports from fishers, may still be better and more robust that decisions, actions or outcomes reliant on short-term scientific information alone.

5.4.1 State of ‘shifting baseline syndrome’ in Brunei Although relatively inexpensive and easy to carry out, care must be taken in collation and use of opinion based knowledge (Venkatachalam et al., 2010; Daw et al., 2011). Of particular significance is the potential for recall bias, which remain a challenge when using memory-based perceptions of trends, particularly in the absence of records or repeated surveys.

Few studies on uncovering the cognitive processes of memory and recall bias currently exist in fishers’ LEK literature (e.g. Saenz-Arroyo et al., 2005b; Bunce et al., 2008; Venkatachalam et al., 2010). All these studies, although claimed to have found field evidence for Pauly’s ‘shifting baseline syndrome’ (Pauly, 1995), had, in fact, demonstrated ‘generational amnesia’, a mechanism in which individuals setting their perceptions from their own experience, and failing to pass their experience on to future generations, and hence, as these observers leave the system, the population’s perceptions of normality updates and past conditions are forgotten (Papworth et al., 2009).

156

Chapter 5

Essentially, a range of different mechanisms exist which can either mask or exaggerate perceived trends at a community or individual level. However, for SBS to occur, in addition to the presence of age- or experience-related differences in perception, biological change must be present in the system (Papworth et al., 2009). In this study, fishers’ perceptions of trends in the abundance of the six selected stocks had revealed three mixed results on the state of SBS for Brunei;

(1) Fishers’ perceptions of the narrow-barred Spanish mackerel and ray stocks showed a classic example of SBS where the age- or experience-related perceptual change corresponded with decreasing catches from official fishery data (see Chapter 4 for ray, Appendix 5.5 for mackerel).

(2) In fishers’ perceptions of Japanese scads abundance, the results suggested that declining trends may be exaggerated or incorrectly perceived, especially since the official catch data revealed that recent catches were larger (Appendix 5.5). Influential memory of extremely good catches in earlier years may have created ‘memory illusion’, whereby fishers would inaccurately remember past conditions and recall change where there was none, or vice versa (Papworth et al., 2009). Thus, it is likely that SBS is absent in fishers’ perceptions of these stocks.

(3) Fishers’ perceptions of the demersal stocks (ponyfishes, red snappers and groupers) implied absence of SBS as there were no significant differences of perceived changes between fishers of various experience years.

157

Chapter 5

Knowledge of SBS is important considering that the use of LEK for assessment of system state and dynamics is becoming more common. If SBS does occur, one may need to take into account the inherent bias associated with these sorts of data, especially since biases are expected to increase with the length of time over which respondents are asked to remember (Daw et al., 2011). In addition, SBS can present a problem when setting conservation goals for ecosystem or fish stocks regeneration, as perceptions of past change may influence target setting, particularly when biological data are not readily available (Saenz-Arroyo et al., 2005a; Venkatachalam et al., 2010).

As a result, it may be interesting to note from this study that fishers’ perceptions of how abundant the three demersal stocks were in the past are likely to be more accurate than may have commonly been thought. This seems contradictory to previous studies on fishers’ SBS which suggested that the true magnitude of stock decline is always not well appreciated by those who rely on the resources (SaenzArroyo et al., 2005b; Bunce et al., 2008; Ainsworth et al., 2008; Venkatachalam et al., 2010; Ainsworth, 2011). Except for rays, all the other stocks are targeted species which fishers’ depend on for their livelihood. Perhaps if fishers are more aware of the creeping disappearance of resources, such as in the case of demersal stocks, transfer of LEK across generations may be more effective. In contrast, it may be that due to the migratory nature of the pelagic stocks, the fishers are less likely to acknowledge decline if they believe that the stocks ‘have just gone somewhere else, and will be back’ (Johannes and Neis, 2007). Undeniably, further research is required to test this, as well as to investigate factors which could bias trend perceptions.

158

Chapter 5

5.4.2 Concluding remarks This study addresses the need for more inclusion of fishers’ LEK in the management of fisheries, and represents the first attempt to process and apply LEK data to elucidate abundance trends of the six species groups in Brunei. The study also demonstrates that LEK can provide an invaluable and rapid means of highlighting issues that need further consideration and research. Although such knowledge may not equate to scientific “truth” – as it is based mainly upon perceptions – LEK may contain clues to aid fisheries management in relatively data-poor countries such as Brunei. More importantly, however, this study supports the suggestion that Brunei marine ecosystem seems to have experienced rapid significant changes in the last 10 years.

159

Chapter 6 General Discussion – Towards sustainable management of the demersal fisheries of Brunei Darussalam “The world is beautiful and verdant, and verily Allah, be He exalted, has made you His stewards in it, as He sees how you acquit yourselves." – Prophet Muhammad (PBUH)

Chapter Summary To ensure continuity of its demersal fishery resources, sustained effort to effectively manage the fisheries of Brunei is necessary. At present, the overall goal of Brunei’s fisheries management is the “sustainable development of (coastal) capture fisheries… towards optimization of benefits to the nation”. To assist in achieving this goal, the current legal framework in Brunei has covered most aspects of fisheries, including jurisdiction, management, development and delegation of responsibilities. Despite the good intention of the fisheries laws, and in light of the findings set out in this thesis, there remain issues in implementation to achieve the sustainability of fish resources. Particularly, in relation to the continued open access nature of the fisheries, as well as the inadequate capacity of the institution both in enforcing regulation and carrying out timely applied research. Improved monitoring, assessment and management approaches are required in order to move progressively towards sustainable demersal fishery development.

160

Chapter 6

6.1 Introduction With an ever increasing human population, the issue of resource scarcity has caught the attention of policy makers, academics and the general public. Due to their potential to regenerate, fishery resources have been categorized as renewable resources, similar to forest and grazing pasture. However, they cannot regenerate under all circumstances and their potential to regenerate is limited and conditional on how much of the resource is taken and how much is left in the water. It is this basic idea that underlines the need to understand the resources and factors influencing their variability, both natural and man-made.

Assessment of fisheries has evolved through different phases over time, in parallel to the changing perceptions of the resources. A century or more ago, it was widely believed that the resources of the sea are infinite and that fishing could go on indefinitely (Costanza, 1999). Now, many have acknowledged that most major stocks of the oceans are declining and exhibit serious depletion problems (Myers and Worm, 2003; FAO, 2012b). Many different fishery assessment tools have been developed and it is likely more will be developed in the future, to address specific questions, pertinent at the time of their development. However, these assessment approaches develop not solely out of the questions asked, but also depend on the resources available to accomplish the task and their applicability to the circumstances on which they are to be used. Accordingly, it is within such context that the demersal fishery resources of Brunei Darussalam were assessed in this thesis.

In this chapter, findings from the study presented in the preceding chapters are synthesised together with other information to explore possible approaches that 161

Chapter 6

should be taken in the management of demersal fishery resources, and their implications on the sustainability of the fisheries in Brunei. The current institutional framework and fishery management objectives are reviewed, some of the key issues are identified and several options and/or recommendations are presented.

6.2 Managing the demersal fishery resources of Brunei Darussalam Through an objective analysis of the stock situation with regard to demersal catches from Brunei marine ecosystem, it has been shown in this thesis that significant changes have occurred to the structure of the fish community, particularly since 2000, with an underlying decreasing trend for the overall demersal stocks abundance. Although some of the issues concerning demersal stocks are not new to the fishery authorities in Brunei, they may have not been presented as comprehensively as done here, and were rather speculative in the past. To ensure continuity of the demersal fishery resources, it is essential that sustained effort is made to manage the demersal fishery of Brunei.

6.2.1 Legal and institutional framework As briefly mentioned in Chapter 1, the Department of Fisheries (DOF) is the line agency responsible for a variety of functions which includes fisheries research, development and management. The current fisheries law which defines the authority and functions of DOF is the Fisheries Order of 2009, which recently repealed the Fisheries Act of 1972 and section 4 of the Brunei Fishery Limits Act of 1983, and amended the Fisheries Regulations of 1973.

162

Chapter 6

According to the Order, the organisational structure of DOF is headed by a Director who prepares fisheries plans designed to ensure the optimum utilisation of fishery resources. The Order has also established, among others, the framework for (1) the licensing system, (2) the establishment of lobster fishing areas, marine reserves and marine parks and restrictions related to these areas, (3) the development and management of inland fisheries, and (4) the types of offences, penalties and powers of enforcement officers.

6.2.2 Objectives of fisheries management in Brunei One of the essential requirements for effective fisheries management is the general arrangements of goals and objectives, to allow development of strategies in achieving the goals and objectives, rational allocation of human and resources, and prioritization, weighting and balancing of choices when there are conflicts (Barber and Taylor, 1990). In Brunei, the overall goal of fisheries management is the “sustainable development of (coastal) capture fisheries… towards optimization of benefits to the nation” (Silvestre and Matdanan, 1992; DOF, 2011).

Nevertheless, fisheries management objectives are usually different from more general fisheries goals, whereby they are more verifiable, specific and quantifiable. They often have a performance measure attached by which the management agency can evaluate progress and effectiveness in meeting the stated objectives (Barber and Taylor, 1990). In the 1990s, DOF had identified and adopted the following specific objectives for fishery management in Brunei (Silvestre and Matdanan, 1992):

163

Chapter 6

(1) Increased production and efficient utilisation of capture fishery resources towards greater self-sufficiency. (2) Upliftment of the socioeconomic status or well-being of participants in capture fisheries; and (3) Provision of a stable supply of fish and fishery products at reasonable prices

However, these objectives were formulated then, based on the available assessments and studies of variable scope and reliability. More recently, while DOF still uphold the attempt to optimize productivity or efficiency of the fisheries exploitation regime as one of its management objectives (i.e. objective number 1), they have replaced the other two earlier objectives. Currently, the other two objectives of fishery management in Brunei are (DOF, 2011):

(1) To ensure that the benefits of production or improved productivity are distributed equitably, and (2) To ensure that the productivity generated results in minimum damage to the resource base and natural environment.

One of the strategies to achieve these objectives is to improve the knowledge base of the fisheries and its resources, which this study seeks to provide with respect to the demersal stocks of Brunei.

164

Chapter 6

6.2.3 Current aspects of demersal fishery management At present, management of demersal fishery is mainly directed towards limiting fishing efforts of the LS sector (especially trawlers) although the demersal stocks are also exploited by the SS sectors (Table 4.1, Chapter 4). Some of the regulations governing the demersal fishery include:

(1) Gear controls: the mesh size in the cod-end of bottom trawls must not be less than 51 mm, while the use of spear gun, dynamite and chemicals for fishing is prohibited. (2) Fishing zone specifications: fishing vessels from LS sectors are prohibited to operate in Zone 1 (0 – 3 nm from shoreline). In addition, fishing vessels of inboard engine > 350 horse power is not permitted to operate in Zone 2 (3 – 20 nm from shoreline). (3) Restrictions on entry though licensing: moratorium on new fishing licenses for LS trawlers in Zone 2 and SS fisheries in Zone 1, have been imposed since 2000 and 2008, respectively.

Although there have been improvements, compliance with these regulations is poor. For instance, fishers operating without licenses are the norm rather than the exception in the SS sector. Banned fishing practices (i.e. use of cyanides, etc.) continue to operate freely, and LS vessels are often seen fishing in the shallow waters of Zone 1, much to the annoyance and detriment of SS fishers (Chapter 5).

165

Chapter 6

6.3 Key issues There is a wide range of issues facing Brunei’s fisheries industry in general and the demersal fishery in particular. However, many of them are beyond the scope of this study. Therefore, the key issues presented below are only those of which are relevant to the findings of this study.

6.3.1 Overexploitation of the fisheries resources Trawl surveys carried out by DOF in 2006 revealed that the exploitation rate (E) (see Silvestre and Garces, 2004) of overall demersal stocks was E = 0.50, which was approximately the same as the maximum acceptable limit (Emax = 0.54) although higher than the biological optimum (E0.1 = 0.36) (DOF, 2011). Despite operating around the maximum sustainable yield, stock abundance decline is evident in the past decade. The total catches of overall demersal resources remained stagnant between 2000 and 2009, while the CPUE index of abundance revealed a decreasing trend (Chapter 4). The fishery administrators claimed that the increase in both trawl and SS effort had apparently caused stock abundance to decline (DOF, 2011), yet no significant increase in trawl effort is evident over the study period (Figure 3.3, Chapter 3). This highlights the importance of choosing the right effort indicator as supported by appropriate hypothesis tests (Chapter 4). Like most fisheries worldwide, a larger proportion of the DOF effort is spent on regulating the trawlers, which are often seen to cause more harm to the resources by being relatively wellequipped with modern electronic devices, and heavier and bigger fishing gears. However, it is obvious that while an eight-meter fibreglass fishing boat used by the SS sector is considered “small-scale”, its fishing capacity may be fairly “high” with 166

Chapter 6

two, or even three, outboard motors, and portable electronic devices to aid fishing. Additionally, their ability to easily manoeuvre over reef patches and between oil structures mean that they can simply concentrate their effort on highly productive fishing ground which may be important nursery, spawning or feeding grounds for the fish. This non-random effort distribution was apparent from the interviews with the SS fishers, which shows clustering of high-pressure harvest area (Chapter 5).

At present, the fishery authorities are rapidly developing the aquaculture industry in Brunei, which is erroneously perceived to be able to solve some of the problems posed by declining stocks, by meeting the increasing demand for seafood. However, aquaculture sometimes aggravates the problem of fisheries decline as the feed for farmed (carnivorous) fishes comes from marine ecosystems (Pauly et al., 2002). Therefore, it would be advisable for the fishery management efforts to shift its focus to the SS sector instead, especially the fleet operated (SSc) sub-sector, to regulate the latent excess fishing capacity in Brunei fisheries.

6.3.2 Inadequate information on the fisheries resources The small fishing area and unisectorally oriented legal and institutional framework in Brunei enable relatively good quality of catch data to be collected from the LS sector. Under-reporting of catches may not be an issue, which is often faced by other LS fisheries in the region. For instance, other than to compile catch statistics, catch data are also used by district governments in Indonesia to determine the amount of tax or levy that must be paid by fishing operators or vessel owners. This relationship between the catches reported by fishers and the tax payable increases the likelihood of under-reporting of catches (Prisantoso, 2011). 167

Chapter 6

In spite of this, there is generally lack of consistent time series of fisheries data that are essential for making reliable stock assessments. Perhaps one contributing factor for data and information gathering to be poorly implemented is the lack of clear mandate and understanding of responsibilities on part of the DOF. The present formal top-down system could mean that at the lower levels, nobody is clearly accountable for the accuracy of data collected.

From this study, it is evident that the existing fisheries data available at national level is designed primarily for providing production statistics and not for providing data suited to science-based stock assessments. For instance, catches of several species recorded under the ‘mix’ categories may include catches of juveniles with a separate category for the adult stock. Also noted was the discrepancy in the local and scientific species categories in the current catch-effort data collection system between the different gears/fleet. As an example, catches of obtuse barracuda, Sphyraena obtusata is recorded with a local name of ‘alu-alu’ in the dataset obtained from the purse seine and long line fisheries, but the same catch is categorised under ‘titir’ in the trawl fishery dataset. On the other hand, both the purse seine and long line fisheries dataset had considered the local name of yellowtail barracuda, Syphyraena langsar as ‘titir’.

While such confusion is common even within academia (e.g. Senou, 2001), it limits the usefulness of the catch-effort data collected. For the purpose of this study, objective reasoning led the dataset to be analysed at higher taxonomic levels. In order to improve the assessment of demersal fish resources in Brunei, the fishery authorities should attempt to improve the accuracy of existing fisheries statistics. 168

Chapter 6

Refining the inconsistency within the LS catch-effort data collection system accordingly may be the first step towards improvement.

6.3.3 Open access to the fisheries resources In general, the fishery resources in Brunei may be considered to be managed in an open access manner. The resources are open to all, only requiring one to pay for the annual fishing licence and registering the vessel being used, but even these minimum criteria for entry are seldom adhered to (e.g. many fishers operating onshore are without a fishing licence). What is particularly concerning is the common misconceptions among locals that fishing licences are only for full-time fishers operating out in the sea.

In the absence of control, open access systems, whether in fisheries or in any other free-range resources, invariably become overexploited, hence leading to declining returns for all participants. Such outcome has commonly been dubbed the “Tragedy of the Commons” (Hardin, 1968). Where there is control of overall exploitation, the resources may be protected but serious social and economic distortions commonly still arise (Christie, 2004; Sanchirico et al., 2002). Open access fisheries are characterised by the race to fish, whereby all participants strive to catch as much of the resource as they can, before their competitors do so. Thus, to maintain catch in a situation where resources are declining has encouraged some fishers in Brunei to use illegal fishing techniques. From the interviews conducted in this study, few of the respondents from Kuala Tutong area claimed to have witnessed others using explosives over reef patches, while a number of respondents from Temburong district reported others who may have used fish poison such as rotenone found in 169

Chapter 6

roots of Derris plant (locally known as ‘tuba’), as they can sometimes smell it even after four or five months. These are typical short term measures fishers feel they are obliged to adopt even though there is widespread awareness amongst them of the damage this is inflicting on the resource base and market. In fact, measures to spread awareness such as road shows and dissemination of information through leaflets and other media have been done by the fisheries authorities (DOF-pers.comm.), but there are fishers who argued that communication between fishers and the fisheries authorities is still considerably poor (Table 5.6, Chapter 5), thus leading to a lack of legitimacy of the latter (and therefore their regulations) in the eyes of the user, and hence to poor compliance and cooperation.

At present, while regulations exist for lobster stocks and aquatic mammals in Brunei (Section 24 – 25, and Section 32 of Fisheries Order of 2009), no restrictions are being imposed on the demersal fish stocks.

6.3.4 Inadequate monitoring, control and surveillance capacity Non-compliances are notorious in fisheries, and currently, enforcement is still beyond the capacity of the fishery authorities in Brunei. With the dispersed nature of the fisheries (i.e. numerous landing sites along the rivers and coast), huge numbers of fisheries personnel would be required to enforce regulations under the existing top down management system. Although various Government departments such as the Royal Brunei Marine Police unit and the Immigration Department often coordinate with the DOF in enforcement activities, such operations tend to focus more on patrolling Brunei waters against illegal foreign fishing vessels than enforcing regulations locally. 170

Chapter 6

As a result, the second most common theme identified from the fishers’ interviews in this study relates to the incompetency of DOF to implement fisheries regulations properly (Table 5.6, Chapter 5). Such dissatisfaction is particularly apparent among local SS fishers who perceived the foreign labour fishers as having no or little sense of long-term responsibility to the resources and fishery. Consequently, they felt that the lack of enforcement would weaken even the locals’ moral obligation to comply. Indeed, as moral obligation and social influence are weakened, compliance tends to erode among those who would normally comply with the regulations. These were already found to be the case in a study on enforcement and compliance with fisheries regulations in Malaysia, Indonesia and the Philippines (Viswanathan et al., 1997).

6.3.5 Regulatory ambiguity on recreational (sport) fishery Presence of regulatory ambiguity may complicate fisheries management, as evident from other fisheries worldwide (e.g. Prisantoso, 2011; Njiru et al., 2008). Such issue was noted during the course of this study when discrepancy was detected on the number of fishers as provided by the fishery authorities and personal observation during the interview survey.

Based on the Fisheries Order of 2009, no regulations are currently in place for recreational (sportfishing) fishers in Brunei, except for the licence they have to obtain when holding or organizing a sportfishing event or tournament (Section 23, Fisheries Order of 2009). The definition of sportfishing based on Section 2 of the Order means that it intuitively falls under the part-time SS category according to the fishery authorities, where fishing is done as a secondary activity either for family consumption, for part-time business or for pleasure (Ranimah, 2008). 171

Chapter 6

Ambiguity arises from the definition of sportfishing in the Order, which is taken as “fishing for sport or pleasure only”, without further clarification, on the fishing gears used for instance. Presently, part-time fishers that employ nets and/or traps are being imposed by the fishery authorities to only operate with valid licenses for their gears, while regulations on those that possess hook and line gears (especially pole and line, or ‘joran’) remain unclear. In the licensing database collected by DOF, few records exist of part-time SS fishers that own hook and line gears only, leading to the assumption that these fishers may fish for reasons other than “for sport or pleasure”. However, there is no guarantee that those who own hook and line gears, but do not have a valid licence, may be fishing for pleasure only, and majority would agree that their intent to fish also includes own or family consumption. Subsequently, part-time fishers who own nets and/or traps may also argue that they are “fishing for pleasure” and avoid having to pay for licenses if they are aware of the rights they may assume under the Order. In fact, recreational fishing in some countries such as Canada, Finland and Sweden, is practised with gears that were predominantly designed for commercial purposes such as gill-nets and traps (Cowx, 2002). Hence, there is vagueness in deciding on the situation where a gear licence is required or not for the part-time fishers. Such ambiguity further questions the legitimate objective of the licensing system in Brunei fisheries, whether the system being applied is used to control the number of fishers and/or fishing vessels operating in Brunei waters or used to generate revenue only. In addition, the DOF also acknowledged to using the registered licenses as their source of effort data, so any assessment procedures could potentially produce underestimated results if care is not taken.

172

Chapter 6

The minimal impact on fish stocks and negligible contribution of recreational fishery to overall fisheries productivity is often used as an argument against legitimate management of this user group. However, the fishing effort exerted by recreational fishers can be massive, and can account for a significant component of the catch for the country. This has been shown in South Africa where the total contribution of recreational fishery to the total landings in the marine inshore fishery was over 30% (Cockcroft et al., 2000), while in the US, Ihde et al. (2011) revealed that 71% of species for which harvest data were available showed an increase in the magnitude of recreational harvests relative to commercial harvest from 1981 to 2006.

6.3.6 Consideration of climate change and environmental variability Regional warming trends for Southeast Asia are not above the global average and will more gradually manifest than in other regions such as the Middle East and Central Asia (Carius and Maas, 2009). This supports the lack of significant long-term changes in the environmental variables between 2000 and 2009 in Brunei as assessed in this study (Table 4.3, Chapter 4). However, climate variability is likely to increase and will become a major challenge due to its associated unpredictability (IPCC, 2007; Carius and Maas, 2009).

While the fishery authorities have acknowledged climate change as a contributing factor that aggravated the state of demersal stocks in Brunei (DOF, 2011), variability in climate and other environmental parameters are not being considered in the management of the Brunei fisheries in general. Perhaps because little reliable data is available on concrete impacts of climate change, even at regional level (Carius and Maas, 2009). But with better understanding on how climate influences the variability 173

Chapter 6

seen in the catch of commercially-important species, fisheries managers will have the opportunity to improve the sustainable harvest of fish in response to climatic conditions. In Australia, terrestrial farming systems have been managed in response to forecasts of rainfall variability for quite some time now, and results have been both reductions in financial risk and improved sustainability (Meinke and Hochman, 2000).

Impacts on the fishery as a result of anthropogenic climate change may alter the relationship between environmental parameters and demersal stock catches identified in this study. As ecosystems are altered, extreme events become more common (IPCC, 2007) and the mean state of various environmental parameters will need to be adjusted.

6.3 Monitoring, assessment and/or management recommendations Earlier reports often highlighted how Brunei, compared to other countries in the region, is in a “favourable position”, particularly in the context of development and management of coastal and marine resources which are considered to be relatively underexploited (e.g. DOF-MIPR, 1992, Silvestre and Matdanan, 1992, Silvestre and Garces, 2004). Yet, management of demersal fishery resources in Brunei still suffers from various weaknesses, as highlighted in previous section, which to some extent are related to the lack of information required for, and lack of capacity to implement, successful management. While there is no sign of systematic failure, there is indeed plenty of room for improvement. Suggestions for better monitoring, assessment and/or management approaches are summarised as follows.

174

Chapter 6

6.3.1 Licensing requirement In order to achieve the improvement of enforcement and compliance via the licensing system, the fishery authorities should impose licenses to all fishery entities without any exception, including recreational (sportfishing) fishers whether they operate inland (rivers and lakes), inshore or offshore. This can reduce IUU fishing, particularly with the ever increasing number of part-time fishers (Figure 1.5, Chapter 1). Except for the fish landing complexes (FLC) where strict monitoring is already in place, enforcement of the licensing requirement at main landing sites (especially Jerudong, Kuala Tutong and Kuala Belait) should be enhanced and plausibly linked to a reporting obligation for data collection. Involving community in the enforcement of licensing system should also be explored, where encouragement approaches, with incentives to enhance compliance and discouragement approach to reduction of noncompliance can be introduced.

6.3.2 State-of-the-art monitoring system While tremendous efforts on monitoring, control and surveillance has been noted on illegal foreign fishing vessels in recent years, the fishery authorities should attempt to improve control over the encroachment of LS fishing vessels at sea by utilizing the fishing vessel monitoring system (VMS) technology. VMS is nowadays a standard tool of fisheries monitoring and control worldwide (FAO, 2005b), but it has yet to be implemented in Brunei fishery. Admittedly, putting VMS into practice is a costly and complex procedure that requires careful research and planning, and more importantly, acceptance by the industry. High-level fishery authorities are already aware of this, with experimental installation of VMS on two purse-seine vessels several years ago for a regional study on pelagic resources (Matzaini et al., 2005). 175

Chapter 6

However, the objective then was not on the practical use of VMS, and while the DOF had resolved on the possibility of using VMS as stated in a regional workshop report (FAO, 2005b), the project never took off.

6.3.3 Access rights The biological characteristics of the fish stocks are often used as arguments against fine-scale local management, in that the restriction of access would be ineffective if the management area did not encompass the full distribution range of the stocks. However, there is a consensus of long-term site attachment in demersal species in all environments, especially reef fishes (Jones et al., 2008). Therefore, restricting access can lead to specific management arrangements for individual stocks. For instance, one of the more effective cases of fisheries management in Indonesia relates to the Bali Straits’ sardine stock where participation is limited to fishers from two areas only (Prisantoso, 2011).

However, a particular concern that arises in moving from a system of open access to one of limited access is determining which of the present users should be granted access and to whom it should be denied. Moreover, the social conflicts generated may be considerable, even if only for a short-term. Nevertheless, access rights may encourage a sense of ownership in the user, which should lead to a greater sense of long-term responsibility to the resources and fishery, leading to more responsible fishing (Shotton, 2000). In fact, the right to fish carries with it the obligation to conserve and manage the resources effectively, under the United Nation Convention on the Law of the Sea (UNCLOS III) and FAO International Code of Conduct for Responsible Fisheries, which Brunei had signed to adhere. 176

Chapter 6

However, a policy on the possible restriction of access should only be made after consideration of relevant factors and wide public discussion. This leads to the next point which considers co-management or participatory approach in fisheries management.

6.3.4 Co-management and the participatory approach Many have acknowledged community-based co-management as the only realistic solution for many of the problem facing global fisheries (Berkes et al., 2007; Gutiérrez et al., 2011). Co-management has been noted to prevent the tragedy of the commons because cooperative management by fishers, managers and scientists often results in sustainable fisheries (Gutiérrez et al., 2011). However, in countries where conservation awareness has not caught on well with people, fisheries management is often seen as a governmental responsibility, and to some extent, this may be the case for Brunei, as evident from the interview survey carried out during the course of this study.

Co-management is “an approach to management in which the government share certain authority, responsibilities and functions of managing the fisheries with resources users as partners” (SEAFDEC, 2006). Co-management is seen as an important tenet in modern fisheries management with aspects that address most of the key issues pointed out earlier, such as greater sensitivity to local socioeconomic and ecological constraints, improved management through use of LEK, collective ownership by users in decision making, increased compliance with regulations through peer pressure and better monitoring, control and surveillance by fishers (Berkes et al., 2007). 177

Chapter 6

Currently, there is no co-management system in place for Brunei, although in June 2011 at the ASEAN-SEAFDEC Conference on Sustainable Fisheries for Food Security Towards 2020, the Brunei Government had resolved to promote comanagement aimed at increasing the socio-economic benefits of the fisheries to all stakeholders. The regional guidelines for co-management in Southeast Asia (SEAFDEC, 2006) suggest the local government to work in partnership with resource user organisation at the community level for a more effective management of fisheries resources. However, such organisation is non-existent in Brunei. In fact, formation of local fishing associations in Brunei has been proposed in the report by Silvestre and Matdanan (1992), but twenty years later this idea has yet to be implemented.

Data collected from the interview survey in this study had identified few outstanding individuals and this could serve as a starting point towards the establishment of an appropriate community-level organisation, on the basis that leadership is critical for successful co-management of fisheries (Gutiérrez et al., 2011). Accordingly, the presence of at least one singular individual with entrepreneurial skills, highly motivated, respected as a local leader and making a personal commitment to the comanagement implementation process, was required. As a result, the authors further suggested that additional resources should be spent on efforts to identify community leaders and build social capital rather than only imposing management tactics without users’ involvement (Gutiérrez et al., 2011).

178

Chapter 6

6.3.5 National fishery statistics Decision-making in fisheries management is ideally based on the scientific evaluation of fish stock characteristics, but these high standards will be more difficult to achieve if the minimum requirement of an appropriate fishery data collection system is barely met. Few options to improve the quality and usability of the existing catch and effort data include:



Improvement of current LS data collection system in terms of accuracy and reliability, by routinely collecting and updating the data from time to time, perhaps by replacing monthly log-books submission into weekly or fortnightly.



Fisheries observers and inspectors should be on board LS vessels as often as possible.



Logbooks should be introduced on the SSc fleet, as this could help in assessment of fishing effort exerted by this sub-sector.



Conditions for license renewals for both LS and SSc sectors should include submission of accurate statistics of all operations.



Appoint full-time enumerators to collect data, not only from the LS sector, but also from the SS sector.

6.4 Concluding remarks Although the main data used in this study were acquired from the LS sector, the trends in CPUE may still reflect the trends of actual abundance level of demersal stocks entering the fisheries of Brunei within the study area due to the non-selectivity nature of the gears used by the LS sector (particularly trawlers). Most, if not all, of 179

Chapter 6

the SS fishers operate gears that are passive and selective, and only about 40.9% employ gears targeted at the demersal stocks (Chapter 5).

On a national level, despite the relatively small fishing area and population size of Brunei, management of its demersal fishery resources appears to be ineffective in addressing the issue of declining stocks. Hence, there is a need to improve the present capacity of the local fishery institution. Assessment of fisheries resources and their exploitation needs to be continually updated, while monitoring, control and surveillance must be strengthened to enable DOF to enforce rules and regulations.

While the national fishery statistics still require improvement, other related research efforts such as exploratory fishing and socio-economic aspect of fisheries, and in particular independent resource surveys, should be undertaken at the same time as well. Close cooperation with fishers, as well as with the industry bodies is also important in understanding overall issues in fisheries. In the context of sustainability of demersal fishery resources in Brunei, data collection from both fishery-dependent and fishery-independent sources should be given due priority, where emphasis is on the coordination and sharing of common data amongst various agencies and stakeholders.

180

Chapter 7 General conclusions and suggestions for future work

181

7.1 Conclusions Statistically assessing the demersal fishery resources, as carried out in this study, was challenging, because of the need to deal with non-experimental and observational data in which random treatments and control groups are not present. Nonetheless, this study has shed some light on the nature and types of changes that have occurred in the marine ecosystem of Brunei. In particular, it has been shown that substantial changes occurred in recent period between 2000 and 2009.

Mean Trophic Level (MTL) of large-scale fishery catches revealed a declining trend in Brunei’s marine ecosystem state, despite the short study period (2000 – 2009), at a rate of 0.08 trophic levels (TL) per decade. In order to focus on changes in relative abundance of the more threatened, higher TL finfish only, MTL was recalculated as 3.30

MTL. Subsequently, food web collapses as implied from a “fishing down the

marine food web” process may not have occurred yet in Brunei waters as (1) the decline in MTL has not been associated with overall declining catches, (2) no significant decline in 3.30MTL detected and (3) catches of high-TL species have been maintained. However, closer inspection of species shifts revealed that within a short time scale of 10 years, significant changes in underlying fish assemblage beyond the direct effects on target species, particularly on the demersal ecological group, had occurred. The implied poor and deteriorating state of Brunei marine coastal ecosystem is further supported by the increasing Fishing-in-Balance (FiB) index and temporal patterns of catch trophic spectra (TS), suggesting expansion of the fishery effort offshore by the end of the study period.

182

In-depth evaluation of the demersal finfish stocks, based on the Catch-Per-UnitEffort (CPUE) index which was standardised for other factors not related to abundance, further corroborated the occurrence of shifts in species composition of Brunei’s demersal habitat. In general, abundance of overall demersal resources had declined, even when total catches remained stationary between 2000 and 2009. Influence of climate and the environment seems relatively minimal, although additional studies to confirm this are in order, especially because potential effects of the environment may be implicit. From this study, trends in CPUE appeared to vary among the different demersal fish families. Generally, abundance of several demersal zoobenthic feeders exhibited decline, while the majority of the intermediate demersal predator stocks were shown to increase over the study period.

Considering the long fishing tradition in Brunei, this study may be open to criticism because the full influence of fishery activities may have been obscured by the short time-series of the official fishery datasets. By using fishers’ local ecological knowledge (LEK), this study has reaffirmed the relatively high abundance of Brunei fishery resources up till the 2000s reported in the literature.The study also revealed the “Shifting Baseline Syndrome” (SBS) among currently active fishers in Brunei, in respect of resource decline. Of particular significance is that fishers’ perception of how abundant the demersal stocks were in the past are likely to be more accurate than commonly suggested in the literature, especially with several authors having raised concerns over the potential problem of SBS in setting conservation goals or other fishery management programmes. However, with special reference to the red snappers stock, this study has also shown that patterns derived from short time series may potentially be wide of the mark, as there is strong evidence of red snapper 183

abundance decline using fishers’ LEK, which compellingly suggests that harvest were higher in earlier decades, before the onset of official fishery data.

While a number of factors have been identified as contributing to falling catch rates of some species, assessment of the resources in Brunei is still relatively insufficient for informing fishery management in a timely manner. However, fisheries scientists have become increasingly aware of the inherent unpredictability of fishery systems, the non-biological complexities of fisheries and the need for alternative complementary perspectives. Thus management needs to be adaptive in nature, guided by the precautionary approach, if sustainability of the fishery is to be ensured.

7.2 Suggestions for future work There are a number of research gaps that have been identified during the course of this study that will require attention for a better academic understanding of exploited demersal stocks in a tropical multi-specific and multi-gear fishery, as well as more comprehensive assessment of the demersal fishery in Brunei: 

The outcomes from this study should be supplemented with information obtained from fishery-independent sources, especially to validate CPUE values as index of relative abundance. Furthermore, incorporation of LEK with both fishery-dependent and fishery-independent data using fuzzy logic should be explored further.



This study has concentrated on changes at the community and family levels, but very little was done at species level. In addition, the analyses could be improved by using more spatially refined catch data, perhaps to the scale of 184

chl-a resolution (0.1° x 0.1° latitude and longitude) which falls within the lower home range of most demersal stocks. 

Inclusion of lagged environmental variables in GLMs of stocks’ CPUE would facilitate extraction of environmental influence from fishing and other effects on stocks’ abundance variability. It could have also benefited from a more comprehensive age-/size-structured catch data, as there are still limited understanding on the influence of multi-gear fishing and environment on stocks’ life history.



A more systematic analysis of fishers’ qualitative LEK data collected in this study has not been explored. Fishers’ LEK could provide an opportunity to highlight anomalies and generate hypotheses related to the fishery resources.

185

REFERENCES ABU TALIB, A., TAN, G. H. & ABD. HAMID, Y. 2003. Overview of the national fisheries situation with emphasis on the demersal fisheries off the West Coast of Peninsular Malaysia. In: SILVESTRE, G., GARCES, L. R., STOBUTZKI, I., AHMED, M., VALMONTE-SANTOS, R. A., LUNA, C., LACHICA-ALINO, L., MUNRO, P., CHRISTENSEN, V. & PAULY, D. (eds.) Assessment, Management and Future Directions for Tropical Coastal Fisheries in Asian Countries. WorldFish Center Conference Proceedings. ADAMS, T. 1998. The interface between traditional and modern methods of fishery management in the Pacific Islands. Ocean & Coastal Management, 40, 127-142. AINSWORTH, C. 2011. Quantifying species abundance trends in the Northern Gulf of California using local ecological knowledge. Marine and Coastal Fisheries, 3, 190218. AINSWORTH, C. H., PITCHER, T. J. & ROTINSULU, C. 2008. Evidence of fishery depletions and shifting cognitive baselines in Eastern Indonesia. Biological Conservation, 141, 848-859. ALHEIT, J. 2009. Consequences of regime shifts for marine food webs. International Journal of Earth Sciences, 98, 261-268. ALLEN, G. R. 2008. Conservation hotspots of biodiversity and endemism for Indo-Pacific coral reef fishes. Aquatic Conservation: Marine and Freshwater Ecosystems, 18, 541-556. AMES, T. 2007. Putting fishers' knowledge to work: Reconstructing the Gulf of Maine cod spawning grounds on the basis of local ecological knowledge. In: HAGGAN, N., NEIS, B. & BAIRD, I. G. (eds.) Fishers' Knowledge in Fisheries Science and Management. Paris: UNESCO. ANDERSON, C. N., HSIEH, C.-H., SANDIN, S. A., HEWITT, R., HOLLOWED, A., BEDDINGTON, J., MAY, R. M. & SUGIHARA, G. 2008. Why fishing magnifies fluctuations in fish abundance. Nature, 452, 835-839. ANGELINI, R. & VAZ-VELHO, F. 2011. Ecosystem structure and trophic analysis of Angolan fishery landings. Scientia Marina, 75, 309-319. APFIC 2007. Managing fishing capacity and IUU fishing in the Asian region. Bangkok: FAO, Regional Office for Asia and the Pacific. ARANCIBIA, H. & NEIRA, S. 2005. Long-term changes in the mean trophic level of Central Chile fishery landings. Scientia Marina, 69, 295-300. ARIF, S. 2007. Sawen: Institution, Local Knowledge and Myths in Fisheries Management in North Lombok, Indonesia. In: HAGGAN, N., NEIS, B. & BAIRD, I. G. (eds.) Fishers’ Knowledge in Fisheries Science and Fisheries Management. . Paris: UNESCO. ARREGUÍN-SÁNCHEZ, F., MUNRO, J. L., BALGOS, M. & PAULY, D. 1996. Biology, Fisheries, and Culture of Tropical Groupers and Snappers: Proceedings of an EPOMEX/ICLARM International Workshop on Tropical Snappers and Groupers, Held at the University of Campeche, Campeche, Mexico, 26-29 October 1993, The WorldFish Center. BAETA, F., COSTA, M. J. & CABRAL, H. 2009. Changes in the trophic level of Portuguese landings and fish market price variation in the last decades. Fisheries Research, 97, 216-222. BALSTON, J. M. 2007. An examination of the impacts of climate variability and climate change on the wild barramundi (Lates calcarifer): A tropical estuarine fishery of north-eastern Queensland, Australia. PhD, James Cook University. BARANOV, F. I. 1918. On the question of the biological basis of fisheries. BARBER, W. E. & TAYLOR, J. N. 1990. The importance of goals, objectives, and values in the fisheries management process and organization: a review. North American Journal of Fisheries Management, 10, 365-373.

186

BATTAILE, B. C. & QUINN, T. J. 2004. Catch per unit effort standardization of the eastern Bering Sea walleye pollock (Theragra chalcogramma) fleet. Fisheries Research, 70, 161-177. BEALES, R. 1982. Investigations into demersal resources accessible to trawling in Brunei waters. In: R. BEALES, D. CURRIE & LINDLEY, R. (eds.) Investigations into fisheries resources in Brunei. Brunei Museum, Brunei Darussalam. BEAUGRAND, G., EDWARDS, M., BRANDER, K., LUCZAK, C. & IBANEZ, F. 2008. Causes and projections of abrupt climate‐driven ecosystem shifts in the North Atlantic. Ecology letters, 11, 1157-1168. BEAUGRAND, G. & KIRBY, R. 2010. Climate, plankton and cod. Global Change Biology, 16, 1268-1280. BEDB. 2012. Brunei Economic Development Board, Prime Minister's Office [Online]. http://www.bedb.com.bn/why_ndp.html. [Accessed October 2012. BEENTJES, M. P. & RENWICK, J. A. 2001. The relationship between red cod, Pseudophycis bachus, recruitment and environmental variables in New Zealand. Environmental biology of fishes, 61, 315-328. BEGG, G. A., FRIEDLAND, K. D. & PEARCE, J. B. 1999. Stock identification and its role in stock assessment and fisheries management: an overview. Fisheries Research, 43, 1-8. BEGG, G. A. & WALDMAN, J. R. 1999. An holistic approach to fish stock identification. Fisheries Research, 43, 35-44. BERGMANN, M., HINZ, H., BLYTH, R., KAISER, M., ROGERS, S. & ARMSTRONG, M. Using expert knowledge to identify possible 'Essential Fish Habitats'. In: HAGGAN, N., BRIGNALL, C. & WOOD, L., eds. Putting Fishers' Knowledge to Work: Conference Proceedings, 2003. BERKES, F. 2006. From community-based resource management to complex systems: the scale issue and marine commons. Ecology and Society, 11, 45. BERKES, F., BERKES, M. K. & FAST, H. 2007. Collaborative integrated management in Canada's north: The role of local and traditional knowledge and community-based monitoring. Coastal Management, 35, 143-162. BEVERTON, R. J. & HOLT, S. J. 1957. On the dynamics of exploited fish populations, Springer. BHATHAL, B. & PAULY, D. 2008. 'Fishing down marine food webs' and spatial expansion of coastal fisheries in India, 1950-2000. Fisheries Research, 91, 26-34. BLANCHARD, J. L., JENNINGS, S., HOLMES, R., HARLE, J., MERINO, G., ALLEN, J. I., HOLT, J., DULVY, N. K. & BARANGE, M. 2012. Potential consequences of climate change for primary production and fish production in large marine ecosystems. Philosophical Transactions of the Royal Society B: Biological Sciences, 367, 2979-2989. BORDALO-MACHADO, P. 2006. Fishing effort analysis and its potential to evaluate stock size. Reviews in Fisheries Science, 14, 369-393. BRANCH, T. A., JENSEN, O. P., RICARD, D., YE, Y. & HILBORN, R. 2011. Contrasting global trends in marine fishery status obtained from catches and from stock assessments. Conservation Biology, 25, 777-786. BRANCH, T. A., WATSON, R., FULTON, E. A., JENNINGS, S., MCGILLIARD, C. R., PABLICO, G. T., RICARD, D. & TRACEY, S. R. 2010. The trophic fingerprint of marine fisheries. Nature, 468, 431-435. BRANDER, K. M. 2007. Global fish production and climate change. Proceedings of the National Academy of Sciences, 104, 19709-19714. BRODNIG, G. & MAYER-SCHÖNBERGER, V. 2000. Bridging the Gap: The role of Spatial Information Technologies in the integration of Traditional Environmental Knowledge and Western Science. Journal on Information Systems in Developing Countries, 1, 1 - 16.

187

BUNCE, L., TOWNSLEY, P., POMEROY, R. & POLLNAC, R. 2000. Socioeconomic manual for coral reef management, Queensland, Australia, Australian Institute of Marine Science and GCRMN. BUNCE, M., RODWELL, L. D., GIBB, R. & MEE, L. 2008. Shifting baselines in fishers' perceptions of island reef fishery degradation. Ocean & Coastal Management, 51, 285-302. BURKE, L., SELIG, E. & SPALDING, M. 2002. Reefs at Risk in Southeast Asia. World Resource Institute. CADDY, J. & GULLAND, J. 1983. Historical patterns of fish stocks. Marine Policy, 7, 267278. CADDY, J. F., CSIRKE, J., GARCIA, S. M. & GRAINGER, R. J. R. 1998. How pervasive is "fishing down marine food webs"? Science, 282, 1383. CADDY, J. F. & GARIBALDI, L. 2000. Apparent changes in the trophic composition of world marine harvests: the perspective from the FAO capture database. Ocean & Coastal Management, 43, 615-655. CARIUS, A. & MAAS, A. 2009. Climate Change and International Security: Two Scenarios for South East Asia. London: Technical Report prepared for DG RELEX of the European Commission. . CBD 2004. The 2020 biodiversity target: a framework for implementation. In: DECISIONS FROM THE SEVENTH MEETING OF THE CONFERENCE OF THE PARTIES OF THE CONVENTION ON BIOLOGICAL DIVERSITY IN KUALA LUMPUR, 9 - 10 AND 27 FEBRUARY 2004 (ed.). Montreal: Secretariat of the CBD. CHAMBERS, R. C. & TRIPPEL, E. A. 1997. Early life history and recruitment in fish populations, Springer. CHEUNG, W. W. L., PITCHER, T. J. & PAULY, D. 2005. A fuzzy logic expert system to estimate intrinsic extinction vulnerabilities of marine fishes to fishing. Biological Conservation, 124, 97-111. CHOAT, J. H. & ROBERTSON, D. R. 2002. Age-based studies on coral reef fishes. Coral Reef Fishes: Dynamics and Diversity in a Complex Ecosystem (San Diego: Academic Press, 2002). CHOU, L., LIM, G. & LENG, C. Fish communities in natural reef and artificial habitats in the coastal waters of Brunei Darussalam. ICLARM Conference Proceedings, 1992. ICLARM. CHRISTENSEN, V. 1998. Fishery-induced changes in a marine ecosystem: insight from models of the Gulf of Thailand. Journal of Fish Biology, 53, 128-142. CHRISTENSEN, V., GARCES, L. R., SILVESTRE, G. & PAULY, D. Fisheries impact on the South China Sea Large Marine Ecosystem: a preliminary analysis using spatially-explicit methodology. In: SILVESTRE, G., GARCES, L. R., STOBUTZKI, I., AHMED, M., VALMONTE-SANTOS, R. A., LUNA, C., LACHICA-ALINO, L., MUNRO, P., CHRISTENSEN, V. & PAULY, D., eds. ICLARM Conference Proceedings, 2003. CHRISTENSEN, V. & PAULY, D. 1992. Ecopath II - a software for balancing steady-state ecosystem models and calculating network characteristics. Ecological Modelling, 61, 169-185. CHRISTENSEN, V. & WALTERS, C. J. 2011. Ecosystem Approaches to Fisheries, Cambridge University Press. CHRISTIE, P. Marine protected areas as biological successes and social failures in Southeast Asia. American Fisheries Society Symposium, 2004. 155-164. CINNER, J. E., FOLKE, C., DAW, T. & HICKS, C. C. 2011. Responding to change: Using scenarios to understand how socioeconomic factors may influence amplifying or dampening exploitation feedbacks among Tanzanian fishers. Global Environmental Change, 21, 7-12. CLOSE, C. 2003. Integrating Local Knowledge and Spatial Information Technologies for Marine Species Management: A Case Study in the Turks and Caicos Islands. Master of Environmental Studies, University of Waterloo.

188

CLOSE, C. H. & HALL, G. B. 2006. A GIS-based protocol for the collection and use of local knowledge in fisheries management planning. Journal of Environmental Management, 78, 341-352. COCKCROFT, A., GRIFFITHS, M. & TARR, R. 2000. Marine recreational fisheries in South Africa: status and challenges. In: PITCHER, T. J. (ed.) Evaluating the benefits of recreational fisheries. Vancouver BC: University of British Columbia, Fisheries Centre Reports. CONN, P. B., HILBORN, R., WILLIAMS, E. H. & SHERTZER, K. W. 2010. When can we reliably estimate the productivity of fish stocks? Canadian Journal of Fisheries and Aquatic Sciences, 67, 511-523. COOPER, A. B., WEIR, K., PROGRAM, U. O. N. H. S. G., PROGRAM, N. S. G. C., OCEANIC, U. S. N. & ADMINISTRATION, A. 2006. A guide to fisheries stock assessment: from data to recommendations, University of New Hampshire, Sea Grant College Program. COSTANZA, R. 1999. The ecological, economic, and social importance of the oceans. Ecological Economics, 31, 199-213. COWX, I. G. 2002. Recreational Fishing. In: HART, P. J. B. & REYNOLDS, J. D. (eds.) Handbook of Fish Biology and Fisheries. Blackwell Science Ltd. COX, E. 1999. The Fuzzy Systems Handbook: a Practitioner's Guide to Building, Using and Maintaining Fuzzy Systems, San Diego, AP Professional. CURY, P. M., MULLON, C., GARCIA, S. M. & SHANNON, L. J. 2005a. Viability theory for an ecosystem approach to fisheries. ICES Journal of Marine Science: Journal du Conseil, 62, 577-584. CURY, P. M., SHANNON, L. J., ROUX, J. P., DASKALOV, G. M., JARRE, A., MOLONEY, C. L. & PAULY, D. 2005b. Trophodynamic indicators for an ecosystem approach to fisheries. ICES Journal of Marine Science, 62, 430-442. CUSHING, D. H. 1983. Key Papers on Fish Populations, IRL Press. DAAN, N., GISLASON, H., POPE, J. G. & RICE, J. C. 2011. Apocalypse in world fisheries? The reports of their death are greatly exaggerated. ICES Journal of Marine Science: Journal du Conseil, 68, 1375-1378. DAMALAS, D., MEGALOFONOU, P. & APOSTOLOPOULOU, M. 2007. Environmental, spatial, temporal and operational effects on swordfish (< i> Xiphias gladius) catch rates of eastern Mediterranean Sea longline fisheries. Fisheries Research, 84, 233-246. DAVIS, A., HANSON, J. M., WATTS, H. & MACPHERSON, H. 2004. Local ecological knowledge and marine fisheries research: the case of white hake (Urophycis tenuis) predation on juvenile American lobster (Homarus americanus). Canadian Journal of Fisheries and Aquatic Sciences, 61, 1191-1201. DAVIS, A. & WAGNER, J. R. 2003. Who knows? On the importance of identifying “experts” when researching local ecological knowledge. Human ecology, 31, 463489. DAW, T. M., ROBINSON, J. & GRAHAM, N. A. J. 2011. Perceptions of trends in Seychelles artisanal trap fisheries: comparing catch monitoring, underwater visual census and fishers' knowledge. Environmental Conservation, 38, 75-88. DE PONTUAL, H., BERTIGNAC, M., BATTAGLIA, A., BAVOUZET, G., MOGUEDET, P. & GROISON, A.-L. 2003. A pilot tagging experiment on European hake (Merluccius merluccius): methodology and preliminary results. ICES Journal of Marine Science, 60, 1318-1327. DELA CRUZ, C., CHOU, L. M. & DE SILVA, M. W. R. N. 1987. Land Use and Development. In: CHUA, T. E., CHOU, L. M. & M, M. S. (eds.) The Coastal Environmental Profile of Brunei Darussalam: Resource Assessment and Management Issues. Fisheries Department, Ministry of Development, Brunei Darussalam and International Center for Living Aquatic Resources Management (ICLARM), Manila, Philippines.

189

DOF-MIPR 1992. The Integrated Management Plan for the Coastal Zone of Brunei Darussalam, ICLARM Technical Report 29 (Department of Fisheries, Ministry of Industry and Primary Resources, Brunei Darussalam). DOF-PERS.COMM. RE: Personal communication with fisheries officers at Department of Fisheries, Brunei Darussalam from 2009 to 2012. DOF. 1999. Takat tiruan: Kawasan alternatif bagi nelayan kecil-kecilan Negara Brunei Darussalam [Online]. Available: http://www.fisheries.gov.bn/publications/leaflets/takat_tiruan.htm [Accessed 15 Nov 2009. DOF. 2009. Department of Fisheries, Ministry of Industries and Primary Resources [Online]. Available: www.fisheries.gov.bn 2010]. DOF 2011. Brunei Darussalam Marine Capture Fisheries: Demersal Resources, Status and Management. Brunei's World Oceans Days Seminar 2011. Bandar Seri Begawan, Brunei Darussalam. DULVY, N., POLUNIN, N. V., MILL, A. & GRAHAM, N. A. 2004. Size structural change in lightly exploited coral reef fish communities: evidence for weak indirect effects. Canadian Journal of Fisheries and Aquatic Sciences, 61, 466-475. DULVY, N. K. & POLUNIN, N. V. 2004. Using informal knowledge to infer human‐ induced rarity of a conspicuous reef fish. Animal Conservation, 7, 365-374. ECOPATH. 2011. Ecopath with Ecosim: No fish is an island [Online]. Available: http://www.ecopath.org/ January 2013]. ESSINGTON, T. E., BEAUDREAU, A. H. & WIEDENMANN, J. 2006. Fishing through marine food webs. Proceedings of the National Academy of Sciences of the United States of America, 103, 3171-3175. FAO 1995. Code of Conduct for Responsible Fisheries, Rome, FAO. FAO. 2005a. Fisheries and Aquaculture topics: Other international organizations - Topics Fact Sheets (Text by S.M. Garcia) [Online]. Rome: FAO Fisheries and Aquaculture Department Available: http://www.fao.org/fishery/topic/2897/en March 2013]. FAO 2005b. Reports of the regional vessel monitoring systems workshops: Southwest Indian Ocean, the Caribbean, Central America and Southeast Asia, Rome, Food and Agriculture Organization of the United Nations. FAO 2012a. Fishery and Aquaculture Statistics 2010, Rome, FAO. FAO 2012b. The State of World Fisheries and Aquaculture 2012, FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS, Rome. FISHBASE. 2012. FishBase: A Global Information System on Fishes [Online]. www.fishbase.org. Available: www.fishbase.org [Accessed September 2009 December 2012. FOGARTY, M. J. & POWELL, T. M. 2002. An overview of the US GLOBEC Program. Oceanography, 15, 4-12. FOLKE, C. 2004. Traditional knowledge in social-ecological systems. Ecology and Society, 9, 7. FREIRE, K. M. F. & PAULY, D. 2010. Fishing down Brazilian marine food webs, with emphasis on the east Brazil large marine ecosystem. Fisheries Research, 105, 57-62. FRIEDLAND, K., REDDIN, D., MCMENEMY, J. & DRINKWATER, K. 2003. Multidecadal trends in North American Atlantic salmon (Salmo salar) stocks and climate trends relevant to juvenile survival. Canadian Journal of Fisheries and Aquatic Sciences, 60, 563-583. FROESE, R. & KESNER-REYES, K. Impact of fishing on the abundance of marine species. 2002. ICES Council Meeting Report CM. FROESE, R. & PAULY, D. 2000. FishBase 2000: concepts, design and data sources. Los Baños, Laguna, Philippines: ICLARM. FROESE, R., STERN-PIRLOT, A. & KESNER-REYES, K. 2009. Out of new stocks in 2020: A comment on" Not all fisheries will be collapsed in 2048". Marine Policy, 33, 180-181.

190

FROESE, R., ZELLER, D., KLEISNER, K. & PAULY, D. 2012. What catch data can tell us about the status of global fisheries. Marine Biology, 159, 1283-1292. FULTON, E. A., SMITH, A. D. & PUNT, A. E. 2005. Which ecological indicators can robustly detect effects of fishing? ICES Journal of Marine Science: Journal du Conseil, 62, 540-551. GALLUCCI, V. F., SAILA, S. B., GUSTAFSON, D. J. & ROTHSCHILD, B. J. 1996. Stock Assessment: Quantitative Methods and Applications for Small-scale Fisheries, CRC Press, Inc. GARCES, L. R., STOBUTZKI, I., ALIAS, M., CAMPOS, W., KOONGCHAI, N., LACHICA-ALINO, L., MUSTAFA, G., NURHAKIM, S., SRINATH, M. & SILVESTRE, G. 2006. Spatial structure of demersal fish assemblages in South and Southeast Asia and implications for fisheries management. Fisheries Research, 78, 143-157. GARCIA, S., ZERBI, A., ALIAUME, C., DO CHI, T. & LASSERRE, G. 2003. The ecosystem approach to fisheries. . FAO Fisheries Technical Paper. FAO, Rome, Italy. GARCIA, S. M. & COCHRANE, K. L. 2005. Ecosystem approach to fisheries: a review of implementation guidelines. ICES Journal of Marine Science, 62, 311-318. GARCIA, S. M. & DE LEIVA MORENO, I. 2003. Global Overview of Marine Fisheries. GARCIA, S. M. & STAPLES, D. J. 2000. Sustainability reference systems and indicators for responsible marine capture fisheries: a review of concepts and elements for a set of guidelines. Marine and Freshwater Research, 51, 385-426. GASCUEL, D., BOZEC, Y.-M., CHASSOT, E., COLOMB, A. & LAURANS, M. 2005. The trophic spectrum: theory and application as an ecosystem indicator. ICES Journal of Marine Science: Journal du Conseil, 62, 443-452. GAYANILO, F. C. 1997. Fisat: Fao-Iclarm Stock Assessment Tools, Reference Manuals, Food and Agriculture Organization of the United Nations. GAYANILO, J. F. C., SPARRE, P., PAULY, D., CENTER, W., FOOD & NATIONS, A. O. O. T. U. 2005. FAO-ICLARM Stock Assessment Tools II: FiSAT II : User's Guide, WorldFish Center. GOVERNMENT OF BRUNEI DARUSSALAM 2009. Fisheries Order. In: ORDER MADE UNDER ARTICLE 83(3) OF CONSTITUTION OF BRUNEI DARUSSALAM (ed.). Brunei Darussalam Government Gazette. GRAY, T., HATCHARD, J., DAW, T. & STEAD, S. 2008. New cod war of words: ‘Cod is God’ versus ‘sod the cod’ — Two opposed discourses on the North Sea Cod Recovery Programme. Fisheries Research, 93, 1-7. GULLAND, J. 1972. Some notes on the demersal resources of Southeast Asia. Indo-Pac. Fish. Counc, 13, 51-60. GUTIÉRREZ, N. L., HILBORN, R. & DEFEO, O. 2011. Leadership, social capital and incentives promote successful fisheries. Nature, 470, 386-389. HAGGAN, N., NEIS, B. & BAIRD, I. G. 2007. Fishers' knowledge in fisheries science and management, UNESCO Publishing. HAGGARTY, D. R. & KING, J. R. 2006. CPUE as an index of relative abundance for nearshore reef fishes. Fisheries research, 81, 89-93. HALL, G. B. & CLOSE, C. H. 2007. Local knowledge assessment for a small-scale fishery using geographic information systems. Fisheries Research, 83, 11-22. HALL, G. B., MOORE, A., KNIGHT, P. & HANKEY, N. 2009. The extraction and utilization of local and scientific geospatial knowledge within the Bluff oyster fishery, New Zealand. Journal of environmental management, 90, 2055-2070. HARDIN, G. 1968. The Tragedy of the Commons. Science, 162, 1243-1248. HARLEY, S. J., MYERS, R. A. & DUNN, A. 2001. Is catch-per-unit-effort proportional to abundance? Canadian Journal of Fisheries and Aquatic Sciences, 58, 1760-1772. HARRIS, M. 1998. Lament for an Ocean: The Collapse of the Atlantic Cod Fishery: A True Crime Story. McClelland Sc Stewart. Inc., Toronto.

191

HART, P. J. B. & REYNOLDS, J. D. 2002a. Handbook of Fish Biology and Fisheries, Wiley. HART, P. J. B. & REYNOLDS, J. D. 2002b. Handbook of Fish Biology and Fisheries, Wiley. HEATH, M. & BRANDER, K. 2001. Workshop on gadoid stocks in the North Sea during the 1960s and 1970s: the 4th ICES/GLOBEC backward-facing workshop. ICES Cooperative Research Report, 244. HICKEY, F. 2007. Traditional marine resource management in Vanuatu: World views in transformation. . In: HAGGAN, N., NEIS, B. & BAIRD, I. G. (eds.) Fishers' Knowledge in Fisheries Science and Management. Paris: UNESCO. HILBORN, R. 2010. Pretty good yield and exploited fishes. Marine Policy, 34, 193-196. HILBORN, R. & PETERMAN, R. 1995. The development of scientific advice with incomplete information in the context of the precautionary approach. Precautionary approach to fisheries. Rome: FAO. HILBORN, R. & WALTERS, C. J. 1992. Quantitative Fisheries Stock Assessment: Choice, Dynamics and Uncertainty, Chapman and Hall, Inc. HJORT, J. 1926. Fluctuations in the year classes of important food fishes. Journal du Conseil International pour L'Exploration de la Mer, 1, 3-38. HOBDAY, A. J., GRIFFITHS, S. & WARD, T. 2009. Pelagic fishes and sharks. In: POLOCZANSKA, E. S., HOBDAY, A. J. & RICHARDSON, A. J. (eds.) A Marine Climate Change Impacts and Adaptation Report Card for Australia 2009. NCCARF Publication. HONEY, K., MOXLEY, J. & FUJITA, R. 2010. From Rags to Fishes: Data-Poor Methods for Fishery Managers. Managing Data-Poor Fisheries: Case Studies, Models & Solutions, 1, 159 - 184. HORN, P. L. 2003. CPUE from commercial fisheries for ling (Genypterus blacodes) around the North Island, New Zealand: an evaluation of series for LIN 1, LIN 2, and Cook Strait. New Zealand: Ministry of Fisheries. HSIEH, C.-H., REISS, C. S., HUNTER, J. R., BEDDINGTON, J. R., MAY, R. M. & SUGIHARA, G. 2006. Fishing elevates variability in the abundance of exploited species. Nature, 443, 859-862. HUNTINGTON, H. P. 2000. Using traditional ecological knowledge in science: methods and applications. 10, 1270 - 1274. HUTCHINGS, J. A. 1996. Spatial and temporal variation in the density of northern cod and a review of hypotheses for the stock's collapse. Canadian Journal of Fisheries and Aquatic Sciences, 53, 943-962. HVINGEL, C., LASSEN, H. & PARSONS, D. G. 2000. A biomass index for northern shrimp (Pandalus borealis) in Davis Strait based on multiplicative modelling of commercial catch-per-unit-effort data (1976-97). Journal of Northwest Atlantic Fishery Science, 25-36. IBRAHIM, S., MUHAMMAD, M., AMBAK, M. A. & ZAKARIA, M. Z. 2004. Distribution of Food Items of Six Commercially Important Demersal Fishes in the South China Sea. Pertanika Journal of Tropical Agricultural Science, 27, 31-38. ICES 2000. Report of the Working Group on Ecosystem Effects of Fishing Activities. ICES CM2000/ACME:02 Ref:ACFM+E ed. ICES, Copenhagen: ICES CM2000/ACME:02 Ref:ACFM+E. IHDE, T. F., WILBERG, M. J., LOEWENSTEINER, D. A., SECOR, D. H. & MILLER, T. J. 2011. The increasing importance of marine recreational fishing in the US: Challenges for management. Fisheries Research, 108, 268-276. IPCC-SAR 1996. Climate Change 1995: Impacts, Adaptations and Mitigation of Climate Change: Scientific-Technical Analyses. Contribution of Working Group II to the Second Assessment Report of the Intergovernmental Panel on Climate Change. IPCC 2007. Climate Change: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.

192

JACKSON, J. B. C., KIRBY, M. X., BERGER, W. H., BJORNDAL, K. A., BOTSFORD, L. W., BOURQUE, B. J., BRADBURY, R. H., COOKE, R., ERLANDSON, J., ESTES, J. A., HUGHES, T. P., KIDWELL, S., LANGE, C. B., LENIHAN, H. S., PANDOLFI, J. M., PETERSON, C. H., STENECK, R. S., TEGNER, M. J. & WARNER, R. R. 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science, 293, 629-638. JENNINGS, S., REYNOLDS, J. D. & POLUNIN, N. V. 2001. Predicting the vulnerability of tropical reef fishes to exploitation with phylogenies and life histories. Conservation Biology, 13, 1466-1475. JOBLING, M. 2008. Environmental Factors and Rates of Development and Growth. Handbook of Fish Biology and Fisheries. Blackwell Publishing Ltd. JOHANNES, R. & NEIS, B. 2007. The value of anecdote. In: HAGGAN, N., NEIS, B. & BAIRD, I. G. (eds.) Fishers' Knowledge in Fisheries Science and Management. Paris: UNESCO. JOHANNES, R. & YEETING, B. 2001. I-Kiribati knowledge and management of Tarawa’s lagoon resources. Atoll Res. Bull, 489, 24. JOHANNES, R. E. 1981a. Words of the lagoon: fishing and marine lore in the Palau district of Micronesia, University of California Press. JOHANNES, R. E. 1981b. Working with fishermen to improve coastal tropical fisheries and resource management. Bulletin of Marine Science, 31, 673-680. JOHANNES, R. E. 1989. Introduction. In: JOHANNES, R. E. (ed.) Traditional Ecological Knowledge: A Collection of Essays. Gland, Switzerland: World Conservation Union. JOHANNES, R. E. 1998. The case for data-less marine resource management: examples from tropical nearshore finfisheries Trends in Ecology and Evolution, 13, 243 - 246. JOHANNES, R. E., FREEMAN, M. M. R. & HAMILTON, R. J. 2000. Ignore fishers’ knowledge and miss the boat. Fish and Fisheries, 1, 257 - 271. JOHANNES, R. E., SQUIRE, L., GRANAM, T., SADOVY, Y. J. & RENGUUL, H. 1999. Spawning aggregations of groupers (Serranidae) in Palau. Marine Conservation Research Series Publication #1. The Nature Conservancy. JONES, G. P., MUNDAY, P. L. & JULIAN CALEY, M. 2002. Chapter 4 - Rarity in Coral Reef Fish Communities. In: PETER, F. S. (ed.) Coral Reef Fishes. San Diego: Academic Press. JONES, K. M. M., FITZGERALD, D. G. & SALE, P. F. 2008. Comparative Ecology of Marine Fish Communities. Handbook of Fish Biology and Fisheries. Blackwell Publishing Ltd. JPKE 2010. Brunei Darussalam Statistical Yearbook 2009. In: DEPARTMENT OF ECONOMIC PLANNING AND DEVELOPMENT, P. M. S. O. (ed.). Brunei Darussalam. KAWASAKI, T. 1992. Mechanisms governing fluctuations in pelagic fish populations. South African Journal of Marine Science, 12, 873-879. KELLEHER, K. 2005. Discards in the world's marine fisheries: An update. FAO Fisheries Technical Paper. Rome: FAO. KHARTINI, H. M. 2004. Status of Biodiversity Researches in Brunei Darussalam. Department of Agriculture, Ministry of Industry and Primary Resources. KHOO, H. W., SELVANATHAN, S. & ABDUL HALIDI, M. S. 1987. Economic Sector Capture Fisheries. In: CHUA, T. E., CHOU, L. M. & SADORRA, M. S. M. (eds.) The Coastal Environmental Profile of Brunei Darussalam: Resource Assessment and Management Issues. Department of Fisheries, Ministry of Development, Brunei Darussalam and ICLARM. KLEISNER, K., ZELLER, D., FROESE, R. & PAULY, D. 2012. Using global catch data for inferences on the world’s marine fisheries. Fish and Fisheries. KLYASHTORIN, L. B. 2001. Climate changes and long-term fluctuations of commercial catches: the possibility of forecasting, Rome, FAO Fisheries Technical Paper No. 410.

193

KORANTENG, K. A. 1998. The impacts of environmental forcing on the dynamics of demersal fishery resources of Ghana. PhD, University of Warwick. LAURANS, M., GASCUEL, D., CHASSOT, E. & THIAM, D. 2004. Changes in the trophic structure of fish demersal communities in West Africa in the three last decades. Aquatic Living Resources, 17, 163-174. LAVIDES, M. N., POLUNIN, N. V., STEAD, S. M., TABARANZA, D. G., COMEROS, M. T. & DONGALLO, J. R. 2009. Finfish disappearances around Bohol, Philippines inferred from traditional ecological knowledge. Environmental Conservation, 36, 235. LEHODEY, P., CHAI, F. & HAMPTON, J. 2003. Modelling climate‐related variability of tuna populations from a coupled ocean–biogeochemical‐populations dynamics model. Fisheries Oceanography, 12, 483-494. LINK, J. S. 2005. Translating ecosystem indicators into decision criteria. ICES Journal of Marine Science: Journal du Conseil, 62, 569-576. LOZANO-MONTES, H. M., PITCHER, T. J. & HAGGAN, N. 2008. Shifting environmental and cognitive baselines in the upper Gulf of California. Frontiers in Ecology and the Environment, 6, 75-80. LYMER, D., FUNGE-SMITH, S. & MIAO, W. 2010. Status and potential of fisheries and aquaculture in Asia and the Pacific 2010, FAO Regional Office for Asia and the Pacific. MACKINSON, S. 2001. Integrating local and scientific knowledge: an example in fisheries science. Environmental Management, 27, 533-545. MACKINSON, S. & NOTTESTAD, L. 1998. Points of view: combining local and scientific knowledge. Reviews in Fish Biology and Fisheries, 8, 481-490. MATSUNAGA, H. & NAKANO, H. 2006. Analysis of longline CPUE of major pelagic shark species collected by Japanese research and training vessels in the Pacific Ocean. Western and Central Pacific Fisheries Commission. MATZAINI, J., CINCO, E. A., RANIMAH, W. & HALIDI, S. 2007. Introduction. Fishing gears and methods in Southeast Asia: Brunei Darussalam Department of Fisheries, Brunei Darussalam and SEAFDEC, Training Department. MATZAINI, J., RANIMAH, W. & CINCO, E. A. 2005. National Country Report for Brunei Darussalam. In: RAJA BIDIN, R. & KU KASSIM, K. (eds.) Information collection for sustainable pelagic fisheries in the South China Sea. Kuala Terengganu, Malaysia: SEAFDEC. MAUNDER, M. N. & HOYLE, S. D. 2006. Standardisation of yellowfin and bigeye CPUE data from Japanese longliners, 1975-2004. IATTC Working Group on Stock Assessments, 6th Meeting, SAR-7-07. MAUNDER, M. N. & PUNT, A. E. 2004. Standardizing catch and effort data: a review of recent approaches. Fisheries Research, 70, 141-159. MAUNDER, M. N., SIBERT, J. R., FONTENEAU, A., HAMPTON, J., KLEIBER, P. & HARLEY, S. J. 2006. Interpreting catch per unit effort data to assess the status of individual stocks and communities. ICES Journal of Marine Science: Journal du Conseil, 63, 1373-1385. MAUNDER, M. N. & STARR, P. J. 2003. Fitting fisheries models to standardised CPUE abundance indices. Fisheries Research, 63, 43-50. MAURSTAD, A., DALE, T. & BJØRN, P. A. 2007. You Wouldn’t Spawn in a Septic Tank, Would You? Human Ecology, 35, 601-610. MCCLUSKEY, S. M. & LEWISON, R. L. 2008. Quantifying fishing effort: a synthesis of current methods and their applications. Fish and Fisheries, 9, 188-200. MD NURUL ISLAM, G., MOHD NOH, K. & YEW, T. S. 2011. Measuring productivity in fishery sector of Peninsular Malaysia. Fisheries Research, 108, 52-57. MEEUWIG, J., SAMOILYS, M., EREDIANO, J. & KOLDEWEY, H. 2007. Fishers' perceptions of the seahorse fishery in the central Philippines: Interactive approaches and an evaluation of results. In: HAGGAN, N., NEIS, B. & BAIRD, I. G. (eds.) Fishers' knowledge in Fisheries Science and Management. Paris: UNESCO.

194

MEINKE, H. & HOCHMAN, Z. 2000. Using seasonal climate forecasts to manage dryland crops in northern Australia–experiences from the 1997/98 seasons. In: HAMMER, G., NICHOLLS, N. & MITCHELL, C. (eds.) Applications of seasonal climate forecasting in agricultural and natural ecosystems-The Australian Experience The Netherlands: Kluwer Academic. MILESSI, A. C., ARANCIBIA, H., NEIRA, S. & DEFEO, O. 2005. The mean trophic level of Uruguayan landings during the period 1990-2001. Fisheries Research, 74, 223231. MONTEVECCHI, W. & MYERS, A. 1996. Dietary changes of seabirds indicate shifts in pelagic food webs. Sarsia, 80, 313-322. MORALES-NIN, B. 1992. Determination of growth in bony fishes from otolith microstructure. FAO Fisheries Technical Paper No. 322. Rome: FAO. MORATO, T., WATSON, R., PITCHER, T. J. & PAULY, D. 2006. Fishing down the deep. Fish and Fisheries, 7, 24-34. MORENO-BÁEZ, M., CUDNEY-BUENO, R., ORR, B. J., SHAW, W. W., PFISTER, T., TORRE-COSIO, J., LOAIZA, R. & ROJO, M. 2012. Integrating the spatial and temporal dimensions of fishing activities for management in the Northern Gulf of California, Mexico. Ocean & Coastal Management, 55, 111-127. MORENO, G., DAGORN, L., SANCHO, G. & ITANO, D. 2007. Fish behaviour from fishers' knowledge: the case study of tropical tuna around drifting fish aggregating devices (DFADs). Canadian Journal of Fisheries and Aquatic Sciences, 64, 15171528. MORRIS, A. V., ROBERTS, C. M. & HAWKINS, J. P. 2000. The threatened status of groupers (Epinephelinae). Biodiversity and Conservation, 9, 919-942. MULLON, C., FREON, P. & CURY, P. 2005. The dynamics of collapse in world fisheries. Fish and Fisheries, 6, 111-120. MULRENNAN, M. E. 2007. Sustaining a small-boat fishery: Recent developments and future prospects for Torres Strait Islanders, Northern Australia. In: HAGGAN, N., NEIS, B. & BAIRD, I. G. (eds.) Fishers' knowledge in fisheries science and management. Paris: UNESCO. MUNDAY, P. L., CHEAL, A., GRAHAM, N. A., MEEKAN, M., PRATCHETT, M., SHEAVES, M., SWEATMAN, H. & WILSON, S. 2012. Tropical Coastal Fish. In: POLOCZANSKA, E. S., HOBDAY, A. J. & RICHARDSON, A. J. (eds.) A Marine Climate Change Impacts and Adaptation Report Card for Australia 2012. MUNRO, J. L. 2011. Ecosystem Approaches to Fisheries, Cambridge University Press. MUNYANDORERO, J. & GUENTHER, C. B. 2010. A Six-Decade Portrait of Florida Marine Fisheries via Landings-Based Trophodynamic Indicators. North American Journal of Fisheries Management, 30, 259-280. MURRAY, G., NEIS, B. & JOHNSEN, J. P. 2006. Lessons learned from reconstructing interactions between local ecological knowledge, fisheries science, and fisheries management in the commercial fisheries of Newfoundland and Labrador, Canada. Human Ecology, 34, 549-571. MURRAY, G., NEIS, B., PALMER, C. T. & SCHNEIDER, D. C. 2008. Mapping cod: fisheries science, fish harvesters’ ecological knowledge and cod migrations in the northern Gulf of St. Lawrence. Human Ecology, 36, 581-598. MYERS, R. A., BRIDSON, J. & BARROWMAN, N. J. 1995. Summary of worldwide stock and recruitment data. Canadian Technical Report of Fisheries and Aquatic Science. MYERS, R. A., HUTCHINGS, J. A. & BARROWMAN, N. J. 1997. Why do fish stocks collapse? The example of cod in Atlantic Canada. Ecological applications, 7, 91106. MYERS, R. A. & WORM, B. 2003. Rapid worldwide depletion of predatory fish communities. Nature, 423, 280-283. NEIS, B., SCHNEIDER, D., FELT, L., HAEDRICH, R., FISCHER, J. & HUTCHINGS, J. 1999. Northern cod stock assessment: What can be learned from interviewing resource users? Canadian Journal of Fisheries and Aquatic Sciences, 56, 1949-1963.

195

NIELSEN, J. R., DEGNBOL, P., REEVES, S. & HOVGAARD, H. 2001. Indicators as a basis for robust and acceptable fisheries management. Regional Technical Consultation on Indicators for Sustainable Fisheries Management in ASEAN Region. Haiphong, Vietnam. NISHIDA, T., KAILOLA, P. J. & HOLLINGWORTH, C. E. (eds.) 2001. Proceedings of the First International Symposium on GIS in Fishery Science Japan: Fishery Research Group. NJIRU, M., KAZUNGU, J., NGUGI, C., GICHUKI, J. & MUHOOZI, L. 2008. An overview of the current status of Lake Victoria fishery: Opportunities, challenges and management strategies. Lakes & Reservoirs: Research & Management, 13, 1-12. O'REILLY, J. E., MARITORENA, S., SIEGEL, D. A., O'BRIEN, M. C., D.TOOLE, CHAVEZ, F. P., STRUTTON, P., COTA, G. F., HOOKER, S. B., MCCLAIN, C. R., CARDER, K. L., MULLER-KARGER, F., HARDING, L., MAGNUSON, A., PHINNEY, D., MOORE, G. F., AIKEN, J., ARRIGO, K. R., R.LETELIER & CULVER, M. 2000. Ocean Chlorophyll a Algorithms for SeaWiFS, OC2, and OC4: Version 4. In: O'REILLY, J. E., AND 24 CO-AUTHORS (ed.) SeaWiFS Postlaunch Calibration and Validation Analyses, Part 3. NASA Tech. Memo. OLSEN, E. M., OTTERSEN, G., LLOPE, M., CHAN, K.-S., BEAUGRAND, G. & STENSETH, N. C. 2011. Spawning stock and recruitment in North Sea cod shaped by food and climate. Proceedings of the Royal Society B-Biological Sciences, 278, 504-510. OLSSON, P. & FOLKE, C. 2001. Local ecological knowledge and institutional dynamics for ecosystem management: a study of Lake Racken watershed, Sweden. Ecosystems, 4, 85-104. ORTIZ DE ZARATE, V. & ORTIZ DE URBINA, J. M. 2007. Standardized fishing effort of albacore, Thunnus alalunga, caught by the Spanish troll fleet in the Northeast Atlantic, 1981-2005. Collective Volume of Scientific Papers, 60, 404-414. OTTERSEN, G., HJERMANN, D. Ø. & STENSETH, N. C. 2006. Changes in spawning stock structure strengthen the link between climate and recruitment in a heavily fished cod (Gadus morhua) stock. Fisheries Oceanography, 15, 230-243. OTTERSEN, G., LOENG, H. & RAKNES, A. Influence of temperature variability on recruitment of cod in the Barents Sea. ICES marine science symposia. Copenhagen[ICES MAR. SCI. SYMP.]. 1994., 1994. PANAGA. 2007. Brunei Weather [Online]. Available: http://www.panagaclub.com/divisions/kbbc/diving/panagadivers/PANAS/tat.htm [Accessed September 2009. PAPWORTH, S. K. 2007. Shifting baseline syndrome: An investigation. MSc, Imperial College London. PAPWORTH, S. K., RIST, J., COAD, L. & MILNER-GULLAND, E. J. 2009. Evidence for shifting baseline syndrome in conservation. Conservation Letters, 2, 93 - 100. PARSONS, T. 1996. Taking stock of fisheries management. Fisheries Oceanography, 5, 224-226. PATERSON, B. & PETERSEN, S. L. 2010. EAF implementation in Southern Africa: Lessons learnt. Marine Policy, 34, 276-292. PAULY, D. 1979. Theory and management of tropical multispecies stocks: A review, with emphasis on the Southeast Asian demersal fisheries. ICLARM Studies and Reviews. Manila: ICLARM. PAULY, D. 1988. Fisheries research and the demersal fisheries of Southeast Asia. In: GULLAND, J. A. (ed.) Fish Population Dynamics. Chichester: Wiley Interscience. PAULY, D. 1995. Anecdotes and the Shifting Baseline Syndrome of Fisheries. Trends in Ecology & Evolution, 10, 430-430. PAULY, D. 2010. 5 Easy Pieces: The Impact of Fisheries on Marine Ecosystems, Island Press. PAULY, D., CHRISTENSEN, V., DALSGAARD, J., FROESE, R. & TORRES, F. 1998. Fishing down marine food webs. Science, 279, 860-863.

196

PAULY, D., CHRISTENSEN, V., GUENETTE, S., PITCHER, T. J., SUMAILA, U. R., WALTERS, C. J., WATSON, R. & ZELLER, D. 2002. Towards sustainability in world fisheries. Nature, 418, 689-695. PAULY, D., CHRISTENSEN, V. & WALTERS, C. 2000. Ecopath, Ecosim, and Ecospace as tools for evaluating ecosystem impact of fisheries. ICES Journal of Marine Science, 57, 697-706. PAULY, D. & CHUENPAGDEE, R. 2003. Development of fisheries in the Gulf of Thailand Large Marine Ecosystem: Analysis of an unplanned experiment. In: HEMPEL, G. & SHERMAN, K. (eds.) Large Marine Ecosystems of the World. Amsterdam: Elsevier Science. PAULY, D., HILBORN, R. & BRANCH, T. A. 2013. Fisheries: Does catch reflect abundance? Nature, 494, 303-306. PAULY, D., MORGAN, G. R., MANAGEMENT, I. C. F. L. A. R. & AL-ʻILMĪYAH, M. A.-K. L.-A. 1987. Length-Based Methods in Fisheries Research, International Center for Living Aquatic Resource Management and Ma'had al-Kuwayt lil-Abḥāth al-'Ilmīyah. PAULY, D. & PALOMARES, M. 2010. An empirical equation to predict annual increases in fishing efficiency. Fisheries Centre Working Paper, 7. PAULY, D., PALOMARES, M. L., FROESE, R., SA-A, P., VAKILY, M., PREIKSHOT, D. & WALLACE, S. 2001. Fishing down Canadian aquatic food webs. Canadian Journal of Fisheries and Aquatic Sciences, 58, 51-62. PAULY, D. & WATSON, R. 2005. Background and interpretation of the 'Marine Trophic Index' as a measure of biodiversity. Philosophical Transactions of the Royal Society B-Biological Sciences, 360, 415-423. PAULY, D. & ZELLER, D. 2003. The global fisheries crisis as a rationale for improving the FAO's database of fisheries statistics. Fisheries Centre Research Reports, 11, 1-9. PENNINO, M. G., BELLIDO, J. M., CONESA, D. & LÓPEZ–QUÍLEZ, A. 2011. Trophic indicators to measure the impact of fishing on an exploited ecosystem. Animal Biodiversity and Conservation, 34, 123-131. PEREZ-ESPANA, H., ABARCA-ARENAS, L. G. & JIMENEZ-BADILLO, M. D. 2006. Is fishing-down trophic web a generalized phenomenon? The case of Mexican fisheries. Fisheries Research, 79, 349-352. PERRY , R. I. 2011. Potential impacts of climate change on marine wild capture fisheries: an update. The Journal of Agricultural Science, 149, 63-75. PERRY, R. I., CURY, P., BRANDER, K., JENNINGS, S., MÖLLMANN, C. & PLANQUE, B. 2010. Sensitivity of marine systems to climate and fishing: concepts, issues and management responses. Journal of Marine Systems, 79, 427-435. PG.KHAIRUL-RIJAL, P. H. A. R. 2007. Pembangunan Perusahaan Perikanan Tangkapan Di Negara Brunei Darussalam (Development of Capture Fisheries Industry in Brunei Darussalam). International Conference on Southeast Asia (ICONSEA). Kuala Lumpur, Malaysia. PIKITCH, E., SANTORA, C., BABCOCK, E., BAKUN, A., BONFIL, R., CONOVER, D., DAYTON, P., DOUKAKIS, P., FLUHARTY, D. & HENEMAN, B. 2004. Ecosystem-based fishery management. Science, 305, 346-347. PILGRIM, S. E. 2006. A Cross-Cultural Study into Local Ecological Knowledge. PhD, University of Essex. PILLING, G. M., APOSTOLAKI, P., FAILLER, P., FLOROS, C., LARGE, P. A., MORALES-NIN, B., REGLERO, P., STERGIOU, K. I. & TSIKLIRAS, A. C. 2009. Assessment and Management of Data-Poor Fisheries. Advances in Fisheries Science. Blackwell Publishing Ltd. PINNEGAR, J., JENNINGS, S., O’BRIEN, C. & POLUNIN, N. 2002. Long‐term changes in the trophic level of the Celtic Sea fish community and fish market price distribution. Journal of applied ecology, 39, 377-390.

197

PINNEGAR, J. K., POLUNIN, N. V. C. & BADALAMENTI, F. 2003. Long-term changes in the trophic level of western Mediterranean fishery and aquaculture landings. Canadian Journal of Fisheries and Aquatic Sciences, 60, 222-235. PITCHER, T. J. 2008. A Bumpy Old Road: Size-Based Methods in Fisheries Assessment. Handbook of Fish Biology and Fisheries. Blackwell Science Ltd. POSTUMA, F. A. & GASALLA, M. A. 2010. On the relationship between squid and the environment: artisanal jigging for Loligo plei at São Sebastião Island (24° S), southeastern Brazil. ICES Journal of Marine Science: Journal du Conseil, 67, 13531362. PRINCE, J. 2010. Managing Data-Poor Fisheries: Solutions from Around the World. PRISANTOSO, B. I. 2011. Assessment and management of the demersal fisheries of the Java Sea, Indonesia. Master of Maritime, University of Wollongong. QIU, Y., LIN, Z. & WANG, Y. 2010. Responses of fish production to fishing and climate variability in the northern South China Sea. Progress in Oceanography, 85, 197-212. QIU, Y., WANG, Y. & CHEN, Z. 2008. Runoff-and monsoon-driven variability of fish production in East China Seas. Estuarine, Coastal and Shelf Science, 77, 23-34. QUETGLAS, A., ORDINES, F., HIDALGO, M., MONSERRAT, S., RUIZ, S., AMORES, Á., MORANTA, J. & MASSUTÍ, E. 2013. Synchronous combined effects of fishing and climate within a demersal community. ICES Journal of Marine Science: Journal du Conseil, 70, 319-328. RAHMAN, A. 2000. Development of an Integrated Traditional and Scientific Knowledge Base: A Mechanism for Accessing, Benefit-Sharing and Documenting Traditional Knowledge for Sustainable Socio-Economic Development and Poverty Alleviation. United Nations Conference on Trade and Development. Geneva, Switzerland. RANIMAH, H. A. W. 2008. National Synthesis on Small-scale fisheries in Brunei Darussalam. In: DEPARTMENT OF FISHERIES, M. O. I. A. P. R. (ed.). Brunei Darussalam. RANIMAH, H. A. W. & CINCO, E. A. 2006. Status and Potential of Offshore Resources in Brunei Darussalam. Department of Fisheries, Ministry of Industry and Primary Resources. RANJITH, M. W. & DE SILVA, N. 1998. The Coastal Resources Management Program in Brunei Darussalam. Ocean & Coastal Management, 38, 147-160. RAOSOFT. 2004. Raosoft, Inc. Sample Size Calculator [Online]. Raosoft, Inc. Available: http://www.raosoft.com/samplesize.html [Accessed September 2009. RAYNER, N. A., PARKER, D. E., HORTON, E. B., FOLLAND, C. K., ALEXANDER, L. V., ROWELL, D. P., KENT, E. C. & KAPLAN, A. 2003. Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. Journal of Geophysical Research-Atmospheres, 108. RICHARDS, L. J. & MAGUIRE, J. J. 1998. Recent international agreements and the precautionary approach: new directions for fisheries management science. Canadian Journal of Fisheries and Aquatic Sciences, 55, 1545-1552. RIST, J. 2007. Bushmeat Catch per Unit Effort in space and time: a monitoring tool for bushmeat hunting. PhD, Imperial College London, University of London. RUDDLE, K. 1986. The supply of marine fish species for fermentation in Southeast Asia. Bulletin of the National Museum of Ethnology, 11, 997-1036. RUDDLE, K. & DAVIS, A. 2011. What is “ecological” in local ecological knowledge? Lessons from Canada and Vietnam. Society & Natural Resources, 24, 887-901. SADOVY, Y. J. 2005. Trouble on the reef: the imperative for managing vulnerable and valuable fisheries. Fish and Fisheries 6, 167 - 185. SAENZ-ARROYO, A., ROBERTS, C. M., TORRE, J. & CARINO-OLVERA, M. 2005a. Using fishers' anecdotes, naturalists' observations and grey literature to reassess marine species at risk: the case of the Gulf grouper in the Gulf of California, Mexico. Fish and Fisheries, 6, 121-133. SAENZ-ARROYO, A., ROBERTS, C. M., TORRE, J., CARINO-OLVERA, M. & ENRIQUEZ-ANDRADE, R. R. 2005b. Rapidly shifting environmental baselines

198

among fishers of the Gulf of California. Proceedings of the Royal Society BBiological Sciences, 272, 1957-1962. SALTHAUG, A. & AANES, S. 2003. Catchability and the spatial distribution of fishing vessels. Canadian Journal of Fisheries and Aquatic Sciences, 60, 259-268. SANCHIRICO, J. N., COCHRAN, K. A. & EMERSON, P. M. 2002. Marine protected areas: economic and social implications, Washington, DC, Resources for the Future SARTIMBUL, A., NAKATA, H., ROHADI, E., YUSUF, B. & KADARISMAN, H. P. 2010. Variations in chlorophyll-a concentration and the impact on Sardinella lemuru catches in Bali Strait, Indonesia. Progress in Oceanography, 87, 168-174. SARY, Z., MUNRO, J. L. & WOODLEY, J. D. 2003. Status report on a Jamaican reef fishery: Current value and the costs of non-management. SBRANA, M., SARTOR, P. & BELCARI, P. 2003. Analysis of the factors affecting crustacean trawl fishery catch rates in the northern Tyrrhenian Sea (western Mediterranean). Fisheries research, 65, 271-284. SCUBAZOO. 2009. Brunei Rapid Marine Bio-Diversity Survey [Online]. Available: http://www.scubazoo.com/updates/news-pr/brunei-rapid-marine-bio-diversitysurvey/ 2013]. SEAFDEC 2006. Supplementary Guidelines on Co-management using Group User Rights, Fishery Statistics, Indicators and Fisheries Refugia, Bangkok, Thailand, Southeast Asian Fisheries Development Center. SENOU, H. 2001. Sphyraenidae. Barracudas. In: CARPENTER, K. E. & NIEM, V. (eds.) FAO species identification guide for fishery purposes. The living marine resources of the Western Central Pacific. Rome: FAO. SHELTON, P. A. 2005. Did over-reliance on commercial catch rate data precipitate the collapse of northern cod? ICES Journal of Marine Science: Journal du Conseil, 62, 1139-1149. SHEPPARD, C. 1995. The shifting baseline syndrome. Marine Pollution Bulletin, 30, 766767. SHOTTON, R. 2000. Use of Property Rights in Fisheries Management, Rome, Food and Agriculture Organization of the United Nations. SIDDEEK, M., ZHENG, J. & PENGILLY, D. 2012. Standardization of CPUE data from the Aleutian Islands Golden King Crab Fishery: Observer and Fish Ticket Data. www.alaskafisheries.noaa.gov: Alaska Fisheries Science Center. SILVANO, R. A., UDVARDY, S., CERONI, M. & FARLEY, J. 2005. An ecological integrity assessment of a Brazilian Atlantic Forest watershed based on surveys of stream health and local farmers' perceptions: implications for management. Ecological Economics, 53, 369-385. SILVANO, R. A. M. & VALBO-JORGENSEN, J. 2008. Beyond fishermen's tales: contributions of fishers' local ecological knowledge to fish ecology and fisheries management. Environment Development and Sustainability, 10, 657-675. SILVESTRE, G. & PAULY, D. 1997a. Management of Tropical Coastal Fisheries in Asia: An Overview of Key Challenges and Opportunities. In: SILVESTRE, G. & PAULY, D. (eds.) Status and Management of Tropical Coastal Fisheries in Asia. ICLARM Conference Proceedings. SILVESTRE, G. & PAULY, D. 1997b. Status and management of tropical coastal fisheries in Asia. ICLARM Conference Proceedings, 53, 208. SILVESTRE, G., SELVANATHAN, S. & SALLEH, A. Preliminary trophic model of the coastal fisheries resources of Brunei Darussalam, South China Sea. Trophic models of aquatic ecosystems. ICLARM Conference Proceedings, 1993. 300-306. SILVESTRE, G. T. & GARCES, L. R. 2004. Population parameters and exploitation rate of demersal fishes in Brunei Darussalam (1989-1990). Fisheries Research, 69, 73-90. SILVESTRE, G. T. & MATDANAN, H. J. H. 1992. Brunei Darussalam Capture Fisheries: A Review of Resources, Exploitation and Management. In: SILVESTRE, G., MATDANAN, H. J. H., SHARIFUDDIN, P. H. Y., SILVA, M. W. R. N. D. &

199

CHUA, T. E. (eds.) The coastal resources of Brunei Darussalam: status, utilization and management. ICLARM Conference Proceedings. SIRABAHA, H. S. 2008. Climate of Brunei Darussalam. In: SERVICE, B. M. (ed.). Department of Civil Aviation, Ministry of Communications. SOLANKI, H. U., DWIVEDI, R. M., NAYAK, S. R., SOMVANSHI, V. S., GULATI, D. K. & PATTNAYAK, S. K. 2003. Fishery forecast using OCM chlorophyll concentration and AVHRR SST: validation results off Gujarat coast, India. International Journal of Remote Sensing, 24, 3691-3699. SPARRE, P. & VENEMA, S. C. 1998. Introduction to Tropical Fish Stock Assessment Part 1:Manual, Rome, Food and Agriculture Organization of the United Nations. STAPLES, D. 2009. Status and potential of offshore resources in South and Southeast Asia. Assessment and management of the offshore resources of South and Southeast Asia. Thailand: APFIC. STERGIOU, K. I., CHRISTOU, E. D. & PETRAKIS, G. 1997. Modelling and forecasting monthly fisheries catches: Comparison of regression, univariate and multivariate time series methods. Fisheries Research, 29, 55-95. STERGIOU, K. I. & TSIKLIRAS, A. C. 2011. Fishing down, fishing through and fishing up: fundamental process versus technical details. Marine Ecology Progress Series, 441, 295-301. STOBUTZKI, I. C., SILVESTRE, G. T. & GARCES, L. R. 2006. Key issues in coastal fisheries in South and Southeast Asia, outcomes of a regional initiative. Fisheries Research, 78, 109-118. SWARTZ, W., SALA, E., TRACEY, S., WATSON, R. & PAULY, D. 2010. The spatial expansion and ecological footprint of fisheries (1950 to present). PloS one, 5, e15143. SYAH, I. 2007. Hydrobiological Survey in the waters of Brunei Darussalam. MSc Marine Environmental Management, University of York. SYLAIOS, G. K., KOUTROUMANIDIS, T. & TSIKLIRAS, A. C. 2010. Ranking and classification of fishing areas using fuzzy models and techniques. Fisheries Management and Ecology, 17, 240-253. THE NATURE CONSERVANCY 2004. Delineating the Coral Triangle, its ecoregions and functional seascapes: Report on an expert workshop, held at the Southeast Asia Center for Marine Protected Areas, Bali, Indonesia (April 30 - May 2, 2003). TIAN, Y., KIDOKORO, H. & FUJINO, T. 2011. Interannual-decadal variability of demersal fish assemblages in the Tsushima Warm Current region of the Japan Sea: Impacts of climate regime shifts and trawl fisheries with implications for ecosystem-based management. Fisheries Research, 112, 140-153. TIEWS, K., SUCONDHAMARN, P. & ISARANKURA, A. 1967. On the changes in the abundance of demersal fish stocks in the Gulf of Thailand from 1963/1964 to 1966 as consequence of the trawl fisheries development. Marine Fisheries Laboratory, Bangkok, 8, 39. TSEHAYE, I. 2007. Monitoring fisheries in data-limited situations: A case study of the artisanal reef fisheries of Eritrea. PhD thesis, Wageningen University, the Netherlands. TSEHAYE, I., MACHIELS, M. A. & NAGELKERKE, L. A. 2007. Rapid shifts in catch composition in the artisanal Red Sea reef fisheries of Eritrea. Fisheries research, 86, 58-68. TURAK, E. & DEVANTIER, L. 2011. Field Guide to Reef-building Corals of Brunei Darussalam, Fisheries Department, Ministry of Industry and Primary Resources. VENEMA, S. C., CHRISTENSEN, J. M. & PAULY, D. 1988. Training in tropical fish stock assessment: a narrative of experience. Rome: FAO Fisheries Department. VENKATACHALAM, A. J., PRICE, A. R. G., CHANDRASEKARA, S. & SELLAMUTTU, S. S. 2010. Risk factors in relation to human deaths and other tsunami (2004) impacts in Sri Lanka: the fishers'-eye view. Aquatic ConservationMarine and Freshwater Ecosystems, 19, 57-66.

200

VERWEIJ, M., VAN DENSEN, W. & MOL, A. 2010. The tower of Babel: Different perceptions and controversies on change and status of North Sea fish stocks in multistakeholder settings. Marine Policy, 34, 522-533. VISWANATHAN, K. K., MUSTAPHA, N., SUSILOWATI, I., SIASON, I. M. & TICAO, C. 1997. Enforcement and compliance with fisheries regulations in Malaysia, Indonesia and the Philippines. Fisheries Co-Management Project Research Report. WALTERS, C. 2003. Folly and fantasy in the analysis of spatial catch rate data. Canadian Journal of Fisheries and Aquatic Sciences, 60, 1433-1436. WALTERS, C., CHRISTENSEN, V. & PAULY, D. 1997. Structuring dynamic models of exploited ecosystems from trophic mass-balance assessments. Reviews in Fish Biology and Fisheries, 7, 139-172. WALTERS, C., PAULY, D. & CHRISTENSEN, V. 1999. Ecospace: Prediction of mesoscale spatial patterns in trophic relationships of exploited ecosystems, with emphasis on the impacts of marine protected areas. Ecosystems, 2, 539-554. WATSON, R. & PAULY, D. 2001. Systematic distortions in world fisheries catch trends. Nature, 414, 534-536. WHO 2012. World Health Organization World Health Statistics 2012. World Health Organization. WORM, B., BARBIER, E. B., BEAUMONT, N., DUFFY, J. E., FOLKE, C., HALPERN, B. S., JACKSON, J. B. C., LOTZE, H. K., MICHELI, F., PALUMBI, S. R., SALA, E., SELKOE, K. A., STACHOWICZ, J. J. & WATSON, R. 2006. Impacts of Biodiversity Loss on Ocean Ecosystem Services. Science, 314, 787-790. XIAO, Y. S. 2004. Use of individual types of fishing effort in analyzing catch and effort data by use of a generalized linear model. Fisheries Research, 70, 311-318. YAMADA, H., TAKAGI, N. & NISHIMURA, D. 2006. Recruitment abundance index of Pacific bluefin tuna using fisheries data on juveniles. Fisheries Science, 72, 333-341. YE, Y. & DENNIS, D. 2009. How reliable are the abundance indices derived from commercial catch-effort standardization? Canadian Journal of Fisheries and Aquatic Sciences, 66, 1169-1178. ZADEH, L. A. 1965. Fuzzy Sets. Information Control, 8, 338 - 353.

201

APPENDICES Appendix A-3.1 (Trophic spectra trends) Summary statistics of trends in catch proportions (logit-transformed). Species catches were distributed by trophic class of 0.1 increments according to their trophic level and were smoothed with a three-increment weighted moving average, except for TL ≤ 2.5.

TL

Slope b

SE

t-stats

2.0 2.5 2.9 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5

0.014 -0.030 0.169 0.059 0.054 0.009 0.031 0.032 -0.053 -0.035 -0.054 -0.020 -0.015 -0.001 -0.030 -0.047 -0.004 0.002 0.047

0.057 0.014 0.050 0.016 0.013 0.009 0.015 0.019 0.024 0.021 0.029 0.016 0.012 0.006 0.016 0.017 0.018 0.016 0.015

0.243 -2.201 3.366 3.653 4.122 1.012 2.037 1.666 -2.195 -1.640 -1.868 -1.280 -1.286 -0.093 -1.899 -2.755 -0.233 0.119 3.135

pvalue 0.814 0.059 0.010 0.006 0.003 0.341 0.076 0.134 0.059 0.140 0.099 0.236 0.234 0.928 0.094 0.025 0.822 0.908 0.014

202

Appendix B-4.1 (GLM models of stocks’ CPUE) Summary statistics of most parsimonious model from GLM analyses of nominal CPUE for overall demersal resources and for each family categories. Dependent Variable: Overall demersal resources Type III Sum Source of Squares df Mean Square a Corrected Model 667.497 73 9.144 Intercept 2489.284 1 2489.284 YEAR 5.586 9 .621 MONTH 31.884 11 2.899 GEAR 221.754 2 110.877 ZONE .095 1 .095 MONTH * GEAR 34.741 22 1.579 YEAR * GEAR 49.882 17 2.934 GEAR * ZONE 8.073 2 4.036 YEAR * ZONE 13.152 9 1.461 Error 254.193 411 .618 Total 6243.511 485 Corrected Total 921.691 484 a. R Squared = .724 (Adjusted R Squared = .675) Dependent Variable: Ariidae Type III Sum Source of Squares df Mean Square a Corrected Model 170.449 21 8.117 Intercept 826.629 1 826.629 YEAR 17.303 9 1.923 MONTH 43.754 11 3.978 ZONE 97.867 1 97.867 Error 46.308 68 .681 Total 3540.327 90 Corrected Total 216.757 89 a. R Squared = .786 (Adjusted R Squared = .720) Dependent Variable: Balistidae Type III Sum Source of Squares df Mean Square a Corrected Model 87.260 15 5.817 Intercept 1320.322 1 1320.322 YEAR 14.297 9 1.589 GEAR 9.417 2 4.709 ZONE 1.625 1 1.625 YEAR * ZONE 11.294 3 3.765 Error 44.182 51 .866 Total 3162.944 67 Corrected Total 131.441 66 a. R Squared = .664 (Adjusted R Squared = .565)

F 14.784 4024.871 1.004 4.687 179.275 .153 2.553 4.744 6.526 2.363

Sig. .000 .000 .436 .000 .000 .696 .000 .000 .002 .013

F 11.919 1213.858 2.823 5.841 143.712

Sig. .000 .000 .007 .000 .000

F 6.715 1524.075 1.834 5.435 1.876 4.346

Sig. .000 .000 .085 .007 .177 .008

Partial Eta Squared .724 .907 .022 .111 .466 .000 .120 .164 .031 .049

Partial Eta Squared .786 .947 .272 .486 .679

Partial Eta Squared .664 .968 .244 .176 .035 .204

203

APPENDIX B-4.1 (Continued) Dependent Variable: Carangidae Type III Sum Source of Squares df Mean Square a Corrected Model 356.334 81 4.399 Intercept 1568.327 1 1568.327 YEAR 53.214 9 5.913 MONTH 22.295 11 2.027 GEAR 3.708 2 1.854 ZONE 8.632 1 8.632 MONTH * GEAR 40.705 20 2.035 YEAR * GEAR 67.823 16 4.239 GEAR * ZONE 8.121 2 4.060 MONTH * ZONE 19.134 11 1.739 YEAR * ZONE 38.698 9 4.300 Error 140.735 190 .741 Total 10251.734 272 Corrected Total 497.069 271 a. R Squared = .717 (Adjusted R Squared = .596) Dependent Variable: Drepaneidae Type III Sum Source of Squares df Mean Square a Corrected Model 34.105 26 1.312 Intercept 2.444 1 2.444 YEAR 20.744 9 2.305 MONTH 13.435 11 1.221 GEAR 5.166 1 5.166 ZONE .766 1 .766 SST 1.607 1 1.607 LN(chl-a) .440 1 .440 PPT .907 1 .907 eqSOI 2.243 1 2.243 MONTH * GEAR .000 0 . YEAR * GEAR .000 0 . GEAR * ZONE .000 0 . MONTH * ZONE .000 0 . YEAR * ZONE .000 0 . Error 27.354 23 1.189 Total 2949.334 50 Corrected Total 61.459 49 a. R Squared = .555 (Adjusted R Squared = .052) Dependent Variable: Ephippidae Type III Sum Source of Squares df Mean Square a Corrected Model 79.878 21 3.804 Intercept 257.047 1 257.047 YEAR 24.734 9 2.748 MONTH 15.660 11 1.424 ZONE 6.860 1 6.860 Error 5.395 16 .337 Total 2101.468 38 Corrected Total 85.273 37 a. R Squared = .937 (Adjusted R Squared = .854)

F 5.939 2117.331 7.982 2.736 2.503 11.654 2.748 5.723 5.482 2.348 5.805

Sig. .000 .000 .000 .003 .085 .001 .000 .000 .005 .010 .000

Partial Eta Squared .717 .918 .274 .137 .026 .058 .224 .325 .055 .120 .216

F 1.103 2.055 1.938 1.027 4.344 .644 1.351 .370 .763 1.886 . . . . .

Sig. .409 .165 .097 .455 .048 .430 .257 .549 .391 .183 . . . . .

Partial Eta Squared .555 .082 .431 .329 .159 .027 .055 .016 .032 .076 .000 .000 .000 .000 .000

Sig. .000 .000 .000 .005 .000

Partial Eta Squared .937 .979 .821 .744 .560

F 11.281 762.313 8.150 4.222 20.344

204

APPENDIX B-4.1 (Continued) Dependent Variable: Gerreidae Type III Sum of Source Squares df Mean Square a Corrected Model 95.683 13 7.360 Intercept 5.864 1 5.864 MONTH 27.309 11 2.483 ZONE 54.953 1 54.953 SST 7.772 1 7.772 Error 43.237 47 .920 Total 3353.658 61 Corrected Total 138.920 60 a. R Squared = .689 (Adjusted R Squared = .603) Dependent Variable: Haemulidae Type III Sum of Source Squares df Mean Square a Corrected Model 74.078 33 2.245 Intercept 441.582 1 441.582 YEAR 22.841 9 2.538 MONTH 13.743 11 1.249 GEAR 2.925 2 1.462 ZONE 7.560 1 7.560 YEAR * GEAR 10.791 6 1.799 MONTH * ZONE 6.902 4 1.726 Error 46.025 84 .548 Total 6523.702 118 Corrected Total 120.103 117 a. R Squared = .617 (Adjusted R Squared = .466) Dependent Variable: Lactariidae Type III Sum of Source Squares df Mean Square a Corrected Model 154.179 22 7.008 Intercept 431.646 1 431.646 YEAR 10.072 9 1.119 MONTH 66.381 11 6.035 GEAR 40.528 1 40.528 ZONE 48.922 1 48.922 Error 31.041 106 .293 Total 2908.846 129 Corrected Total 185.220 128 a. R Squared = .832 (Adjusted R Squared = .798)

F 8.001 6.374 2.699 59.736 8.448

Sig. .000 .015 .009 .000 .006

Partial Eta Squared .689 .119 .387 .560 .152

F 4.097 805.936 4.632 2.280 2.669 13.798 3.283 3.149

Sig. .000 .000 .000 .017 .075 .000 .006 .018

Partial Eta Squared .617 .906 .332 .230 .060 .141 .190 .130

F 23.931 1473.983 3.822 20.607 138.395 167.058

Sig. .000 .000 .000 .000 .000 .000

Partial Eta Squared .832 .933 .245 .681 .566 .612

205

APPENDIX B-4.1 (Continued) Dependent Variable: Leiognathidae Type III Sum Source of Squares df Mean Square a Corrected Model 278.251 46 6.049 Intercept 1097.617 1 1097.617 YEAR 68.918 9 7.658 MONTH 50.799 11 4.618 GEAR 14.507 1 14.507 ZONE 17.282 1 17.282 MONTH * GEAR 41.104 11 3.737 YEAR * GEAR 51.068 7 7.295 GEAR * ZONE 3.382 1 3.382 YEAR * ZONE 41.974 5 8.395 Error 120.904 146 .828 Total 5743.879 193 Corrected Total 399.155 192 a. R Squared = .697 (Adjusted R Squared = .602)

Dependent Variable: Lethrinidae Type III Sum Source of Squares df Mean Square a Corrected Model 42.564 9 4.729 Intercept 2106.822 1 2106.822 YEAR 2.778 5 .556 GEAR 14.683 1 14.683 YEAR * GEAR 10.735 3 3.578 Error 23.229 36 .645 Total 2586.262 46 Corrected Total 65.792 45 a. R Squared = .647 (Adjusted R Squared = .559)

Dependent Variable: Lutjanidae Type III Sum Source of Squares df Mean Square a Corrected Model 673.307 58 11.609 Intercept 1090.567 1 1090.567 YEAR 29.681 9 3.298 MONTH 13.654 11 1.241 GEAR 99.347 2 49.674 ZONE 205.376 1 205.376 YEAR * GEAR 32.488 10 3.249 YEAR * ZONE 22.068 9 2.452 MONTH * GEAR 28.601 14 2.043 GEAR * ZONE 86.468 1 86.468 Error 154.305 315 .490 Total 8390.952 374 Corrected Total 827.611 373 a. R Squared = .814 (Adjusted R Squared = .779)

F 7.304 1325.447 9.247 5.577 17.518 20.869 4.512 8.810 4.084 10.137

Sig. .000 .000 .000 .000 .000 .000 .000 .000 .045 .000

Partial Eta Squared .697 .901 .363 .296 .107 .125 .254 .297 .027 .258

F 7.329 3265.165 .861 22.756 5.546

Sig. .000 .000 .517 .000 .003

Partial Eta Squared .647 .989 .107 .387 .316

F 23.698 2226.304 6.732 2.534 101.405 419.258 6.632 5.006 4.171 176.517

Sig. .000 .000 .000 .004 .000 .000 .000 .000 .000 .000

Partial Eta Squared .814 .876 .161 .081 .392 .571 .174 .125 .156 .359

206

APPENDIX B-4.1 (Continued) Dependent Variable: Mullidae Type III Sum of Source Squares df Mean Square a Corrected Model 115.029 8 14.379 Intercept 513.815 1 513.815 YEAR 13.583 7 1.940 ZONE 22.438 1 22.438 Error 22.484 31 .725 Total 1041.390 40 Corrected Total 137.513 39 a. R Squared = .836 (Adjusted R Squared = .794)

Dependent Variable: Nemipteridae Type III Sum of Source Squares df Mean Square a Corrected Model 339.648 18 18.869 Intercept 349.385 1 349.385 YEAR 29.869 8 3.734 GEAR .002 1 .002 ZONE 2.733 1 2.733 YEAR * GEAR 4.640 1 4.640 YEAR * ZONE 24.968 6 4.161 GEAR * ZONE 25.383 1 25.383 Error 93.600 133 .704 Total 4241.006 152 Corrected Total 433.248 151 a. R Squared = .784 (Adjusted R Squared = .755)

Dependent Variable: Priacanthidae Type III Sum of Source Squares df Mean Square a Corrected Model 130.372 33 3.951 Intercept 69.339 1 69.339 YEAR 13.962 8 1.745 MONTH 5.155 11 .469 GEAR 2.298 1 2.298 ZONE 26.094 1 26.094 Rainfall 1.457 1 1.457 YEAR * ZONE 6.210 3 2.070 MONTH * ZONE 5.435 6 .906 Error 2.988 18 .166 Total 1233.510 52 Corrected Total 133.360 51 a. R Squared = .978 (Adjusted R Squared = .937)

F 19.825 708.430 2.675 30.936

Sig. .000 .000 .027 .000

Partial Eta Squared .836 .958 .377 .499

F 26.812 496.454 5.305 .003 3.883 6.593 5.913 36.068

Sig. .000 .000 .000 .954 .051 .011 .000 .000

Partial Eta Squared .784 .789 .242 .000 .028 .047 .211 .213

F 23.797 417.657 10.512 2.823 13.843 157.176 8.778 12.468 5.456

Sig. .000 .000 .000 .025 .002 .000 .008 .000 .002

Partial Eta Squared .978 .959 .824 .633 .435 .897 .328 .675 .645

207

APPENDIX B-4.1 (Continued) Dependent Variable: Psettodidae Type III Sum of Source Squares df Mean Square a Corrected Model 176.990 26 6.807 Intercept 1734.270 1 1734.270 YEAR 102.308 9 11.368 MONTH 13.633 11 1.239 ZONE 2.994 1 2.994 YEAR * ZONE 11.441 5 2.288 Error 29.068 85 .342 Total 4614.708 112 Corrected Total 206.057 111 a. R Squared = .859 (Adjusted R Squared = .816)

Dependent Variable: Sciaenidae Type III Sum of Source Squares df Mean Square a Corrected Model 183.360 27 6.791 Intercept 1210.188 1 1210.188 YEAR 48.664 9 5.407 MONTH 61.329 11 5.575 ZONE 6.415 1 6.415 SOI 8.623 1 8.623 YEAR * ZONE 14.012 5 2.802 Error 111.812 107 1.045 Total 4275.821 135 Corrected Total 295.171 134 a. R Squared = .621 (Adjusted R Squared = .526)

Dependent Variable: Serranidae Type III Sum Source of Squares df Mean Square a Corrected Model 439.198 32 13.725 Intercept 8342.168 1 8342.168 YEAR 26.348 9 2.928 MONTH 17.031 11 1.548 GEAR 167.631 1 167.631 ZONE 8.261 1 8.261 YEAR * GEAR 22.381 9 2.487 GEAR * ZONE 54.700 1 54.700 Error 173.308 298 .582 Total 11233.661 331 Corrected Total 612.506 330 a. R Squared = .717 (Adjusted R Squared = .687)

F 19.906 5071.324 33.241 3.624 8.755 6.691

Sig. .000 .000 .000 .000 .004 .000

Partial Eta Squared .859 .984 .779 .319 .093 .282

F 6.499 1158.108 5.174 5.335 6.139 8.252 2.682

Sig. .000 .000 .000 .000 .015 .005 .025

Partial Eta Squared .621 .915 .303 .354 .054 .072 .111

F 23.600 14344.196 5.034 2.662 288.238 14.204 4.276 94.056

Sig. .000 .000 .000 .003 .000 .000 .000 .000

Partial Eta Squared .717 .980 .132 .089 .492 .045 .114 .240

208

APPENDIX B-4.1 (Continued) Dependent Variable: Siganidae Type III Sum of Source Squares df Mean Square a Corrected Model 10.097 3 3.366 Intercept 217.969 1 217.969 GEAR 2.861 2 1.430 ZONE 7.628 1 7.628 Error 21.636 53 .408 Total 2428.942 57 Corrected Total 31.733 56 a. R Squared = .318 (Adjusted R Squared = .280)

Dependent Variable: Stromateidae Type III Sum Source of Squares df Mean Square a Corrected Model 63.054 24 2.627 Intercept 208.680 1 208.680 YEAR 22.528 9 2.503 MONTH 22.401 11 2.036 GEAR 5.231 1 5.231 ZONE 6.778 1 6.778 MONTH * GEAR 7.427 2 3.713 Error 44.280 39 1.135 Total 5027.584 64 Corrected Total 107.334 63 a. R Squared = .587 (Adjusted R Squared = .334)

Dependent Variable: Trichiuridae Type III Sum Source of Squares df Mean Square a Corrected Model 56.300 16 3.519 Intercept 1529.011 1 1529.011 YEAR 26.736 9 2.971 GEAR 3.484 1 3.484 ZONE 11.734 1 11.734 YEAR * GEAR 27.782 5 5.556 Error 115.187 99 1.164 Total 4643.861 116 Corrected Total 171.488 115 a. R Squared = .328 (Adjusted R Squared = .220)

Sig. .000 .000 .037 .000

Partial Eta Squared .318 .910 .117 .261

F 2.314 183.798 2.205 1.794 4.607 5.970 3.271

Sig. .010 .000 .043 .089 .038 .019 .049

Partial Eta Squared .587 .825 .337 .336 .106 .133 .144

F 3.024 1314.136 2.553 2.994 10.085 4.776

Sig. .000 .000 .011 .087 .002 .001

Partial Eta Squared .328 .930 .188 .029 .092 .194

F 8.244 533.933 3.504 18.684

209

APPENDIX B-4.1 (Continued) Dependent Variable: Dasyatidae Type III Sum Source of Squares df Mean Square a Corrected Model 343.462 44 7.806 Intercept 1339.025 1 1339.025 YEAR 13.325 9 1.481 MONTH 19.955 11 1.814 GEAR 127.799 1 127.799 ZONE 3.438 1 3.438 YEAR * GEAR 4.189 4 1.047 YEAR * ZONE 32.364 9 3.596 MONTH * GEAR 17.659 9 1.962 Error 64.703 190 .341 Total 4503.826 235 Corrected Total 408.165 234 a. R Squared = .841 (Adjusted R Squared = .805)

F 22.922 3932.039 4.348 5.327 375.281 10.095 3.075 10.560 5.762

Sig. .000 .000 .000 .000 .000 .002 .017 .000 .000

Partial Eta Squared .841 .954 .171 .236 .664 .050 .061 .333 .214

210

Appendix C-4.2 (Missing value patterns of data) Overall summary of missing values in 21 datasets used in this study is given below: Category EPHIPPIDAE MULLIDAE LETHRINIDAE PRIACANTHIDAE SIGANIDAE GERREIDAE STROMATEIDAE BALISTIDAE ARIIDAE PSETTODIDAE TRICHIURIDAE LACTARIIDAE HAEMULIDAE SCIAENIDAE NEMIPTERIDAE LEIOGNATHIDAE DASYATIDAE CARANGIDAE SERRANIDAE LUTJANIDAE OVERALL DEMERSAL

Missing N

Percent

682 680 674 668 663 659 656 653 630 608 604 602 602 585 568 527 485 448 389 346 235

94.7% 94.4% 93.6% 92.8% 92.1% 91.5% 91.1% 90.7% 87.5% 84.4% 83.9% 83.6% 83.6% 81.3% 78.9% 73.2% 67.4% 62.2% 54.0% 48.1% 32.6%

Valid N 38 40 46 52 57 61 64 67 90 112 116 118 118 135 152 193 235 272 331 374 485

EPHIPPIDAE MULLIDAE LETHRINIDAE PRIACANTHIDAE SIGANIDAE GERREIDAE STROMATEIDAE BALISTIDAE ARIIDAE PSETTODIDAE TRICHIURIDAE LACTARIIDAE HAEMULIDAE SCIAENIDAE NEMIPTERIDAE LEIOGNATHIDAE DASYATIDAE CARANGIDAE SERRANIDAE LUTJANIDAE OVERALL DEMERSAL

211

Appendix D-4.3 (Plots of CCF analyses) Cross-correlational functions of the three environmental variables (SST, chl-a and rainfall) as leading indicators and residuals of stocks’ CPUE as lagging indicator. Dashed lines show 95% confidence intervals. ZONE 2 CCF with chl-a

CCF with SST 0.5

CCF with rainfall

Overall demersals

0.3 0.1 -0.1 -0.3 -0.5 0.5

Carangidae, jacks

0.3 0.1 -0.1 -0.3 -0.5 0.5

Sciaenidae, croakers

0.3 0.1 -0.1 -0.3 -0.5 0.5

Lutjanidae, snappers

0.3 0.1 -0.1 -0.3 -0.5 0.5

Serranidae, groupers

0.3 0.1 -0.1 -0.3 -0.5 0

12

24

36

48

0

12

24

36

48

0

12

24

36

212

48

APPENDIX D-4.3 (Continued)

ZONE 2 (Continued) CCF with chl-a

CCF with SST

CCF with rainfall

Haemulidae, grunts

0.5 0.3 0.1 -0.1 -0.3 -0.5 0.5

Lactariidae, false trevallies

0.3 0.1 -0.1 -0.3 -0.5 0.5

Leiognathidae, ponyfishes

0.3 0.1 -0.1 -0.3 -0.5 0.5

Nemipteridae, threadfin breams

0.3 0.1 -0.1 -0.3 -0.5 0.5

Dasyatidae, rays

0.3 0.1 -0.1 -0.3 -0.5 0

12

24

36

48

0

12

24

36

48

0

12

24

36

48

213

APPENDIX D-4.3 (Continued)

ZONE 3

CCF with chl-a

CCF with SST Overall demersals 0.5

CCF with rainfall

0.3 0.1 -0.1 -0.3 -0.5 0.5

Lutjanidae, snappers

0.3 0.1 -0.1 -0.3 -0.5 1.0

Serranidae, groupers

0.5 0.0 -0.5 -1.0 1.0

Nemipteridae, threadfin breams

0.5 0.0 -0.5 -1.0 1.0

Leiognathidae, ponyfishes

0.5 0.0 -0.5 -1.0 0.5

Dasyatidae, rays

0.3 0.1 -0.1 -0.3 -0.5 0

12

24

36

48 0

12

24

36

48 0

12

24

36

48

NOTE: CCF plots of Carangidae, Sciaenidae, Haemulidae and Lactariidae are not given due to large number of missing values. Also note the different y-axis. 214

Appendix E-5.1 (Interview questions) Interview questions (translated) and map used during the survey. Questionnaire on Capture Fishery of Brunei Darussalam 2011

REF. CODE:

Department of Fisheries, MIPR, would like to utilise the knowledge of commercial fishermen in Brunei as part of an evaluation of the marine capture fishery. This should not take more than 30 minutes, as the answers need only be estimates. All answers will be treated with confidentiality and anonymity. Thank you for your assistance. DATE form is completed: ................................................. Section A: Fishers background

Interview LOCATION:...................................................... ..........

Respondent name:.......................................................... Vessel / Company name: ................................................................ Nationality:...................... Age: ................. Years fishing in Brunei waters? ............years; Full-time / Part-time Type of gear used in fishing operations: [ ] Trawl [ ] Purse seine [ ] Long line [ ] Outboard engine boat:...................................................... On the map provided, (1) please circle (O) your most frequently visited site, and (2) please cross (X) the fishing ground you believed has the highest fish density. Section B: Perception of change 1. Name all the marine species which you believed to have been depleted by fishing. ....................................................................................................................................................................................... .. Please choose whether the following has increased, stayed the same (no change) or decreased, since when you first started fishing: Increase No change Decrease 2. Amount of catch for the demersal species in general [ ] [ ] [ ] 3. Abundance of Brunei marine resources in general

[

]

[

]

[

]

4. Fish size of Brunei marine resources in general

[

]

[

]

[

]

5. Average fish price in Brunei market in general

[

]

[

]

[

]

6. The following question is refering to the average amount of catch obtained for each species/groups based on the time period given: (1 – Low, 2 – Medium, 3 – High) Species Red snapper

Grouper

Japanese scad (pelagics)

Period Pre-1970s 1970 – 1984 1984 – 2000 2000 – now Pre-1970s 1970 – 1984 1984 – 2000 2000 – now Pre-1970s 1970 – 1984 1984 – 2000 2000 – now

1 [ [ [ [ [ [ [ [ [ [ [ [

Average catch 2 3 ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [

Species ] ] ] ] ] ] ] ] ] ] ] ]

Ponyfish

Mackarel (pelagics)

Stingray

Period Pre-1970s 1970 – 1984 1984 – 2000 2000 – now Pre-1970s 1970 – 1984 1984 – 2000 2000 – now Pre-1970s 1970 – 1984 1984 – 2000 2000 – now

1 [ [ [ [ [ [ [ [ [ [ [ [

Average catch 2 3 ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [

] ] ] ] ] ] ] ] ] ] ] ]

7. Have you ever caught a red snapper (membangan)? YES [ ] NO [ ] If no, please proceed to question 14. 8. If yes, what was your best ever catch of this species in one day? .............. kg 9. How old were you when you caught your best catch of this species? ............... years [OR Which year?] 10. Please mark with a star () in the map next page, where was your best catch caught? 11. What was the size of the largest fish of this species you ever caught? ................... kg 12. How old were you when you caught this largest fish? ............. years [OR Which year?] 13. What type of fishing gear did you use? [ ] Trawl [ ] Purse seine [ ] Line and hook [ ] Traps [ ] Others: ................................. 14. How do you feel about the way Brunei marine resources are being managed? .............................................................................................................. .......................................................................... Any questions/comments should be addressed to: Miss Syazana Ebil, Tel: 8848158 (m) / 2260669 (r), Email: [email protected]

215

Small version of the map used in the survey (A4 size).

216

Appendix F-5.2 (Example of ‘logical consistency’) Example of a fisher’s response for the map-based portion of the interview survey is given below, where notes were taken on field to check for logical consistency of points/lines drawn. In this example, the fisher frequents the reef patches near Kuala Tutong, stating in general that it would take him around an hour to reach the fishing ground which is about 4-5 miles from shore. As for the location of the best day’s catch of red snapper, he recalled the catch to be made about 3 miles away from Magpie oilfield, in the direction of Champion oilfield. The fisher gave an approximate depth of the area in “depa”, which is a local unit of depth measurement often used by (older) fishers. The values were then converted to meters.

217

Appendix G-5.3 (Interview survey input data) Illustration of fishers’ raw input data for the map-based portion of the interview survey.

218

Appendix H-5.4 (List of depleted species from interview survey) List of depleted species cited by fishers which they perceived had been depleted in their lifetime, with 18 species being cited by all groups. Local name

Family name

Scientific name

Common name

Ikan merah Rumahan

Lutjanidae Scombridae

Tenggiri

Scombridae

Temanung Ikan putih Duai putih Beberahan Kerapu Udang Tamban Membangan Jarang gigi Kelapa-kelapa Ketumbang Kerisi Tongkol Kerisi Bali

Carangidae Carangidae Stromateidae Lutjanidae Serranidae Penaeidae Clupeidae Lutjanidae Sciaenidae Lactariidae Lutjanidae Nemipteridae Scombridae Lutjanidae

red snapper small mackerel narrow-barred Spanish mackerel yellowtail scad jacks silver pomfret John's snapper grouper prawn goldstripe sardinella Malabar red snapper tiger-toothed croaker false trevally blubberlip snapper threadfin breams skipjack tuna goldband snapper

Bekalang

Carangidae

Kembura

Mugilidae

Kuasi

Clupeidae

Kurau

Polynemidae

Kanai Parang Menangin Umpak Gagok Terubuk Ayam laut Menukuk Pusu Puput Belanak Yu Kerapu sambuisambui Luluk

Centrarchidae Chirocentridae Polynemidae Haemulidae Ariidae Clupeidae Balistidae Carangidae Engraulidae Clupeidae Mugilidae Carcharhinidae

Lutjanus erythropterus Rastrelliger spp. Scomberomorus commerson Atule mate Caranx spp. Pampus argenteus Lutjanus johnii Epinephelus spp. Penaeus spp. Sardinella gibbosa Lutjanus malabaricus Otolithes ruber Lactarius lactarius Lutjanus rivulatus Nemipterus spp. Katsuwonus pelamis Pristipomoides multidens Scomberoides commersonnianus Chelon spp. Anodontostoma chacunda Leptomelanosoma indicum Micropterus spp. Chirocentrus spp. Polydactylus sextarius Pomadasys hasta Arius thalassinus Tenualosa macrura Abalistes stellaris Seriola spp. Setipinna spp. Ilisha elongata Moolgarda seheli Carcharhinus dussumieri

Serranidae Clariidae

Malabua

Cyprinidae

Pat-pat kuning Sulit gadong Sulit hijau Sumpit-sumpit Duai hitam Langguran

Osphronemidae Caesionidae Caesionidae Toxotidae Carangidae Carangidae

Number of respondents (by fishing experience years) <10 11-39 >40 Total years years years 11 17 11 39 18 11 9 38 4

11

13

28

9 8 6 9 7 1 1 2 2 2 1 2 3 1

10 13 10 11 7 9 7 4 5 5 4 1 1 2

8 5 9 5 1 4 5 2 1 1 1 2 1 1

27 26 25 25 15 14 13 8 8 8 6 5 5 4

Talang queenfish

1

1

1

3

mullet

0

3

6

9

gizzard shad

0

2

5

7

Indian threadfin

0

1

5

6

Borneo black bass wolf herring blackspot threadfin silver grunt giant catfish longtail shad starry triggerfish amberjack hairpin anchovy slender shad bluespot mullet whitecheek shark

0 0 0 0 0 0 0 0 0 0 0 1

3 1 0 0 3 2 1 0 0 2 1 0

3 3 3 3 2 2 2 2 2 1 1 1

6 4 3 3 5 4 3 2 2 3 2 2

Chromileptes altivelis

humpback grouper

0

0

1

1

Clariidae spp. Osteochilus melanopleurus Osphronemus goramy Caesio cuning Caesio caerulaurea Toxotes jaculatrix Parastromateus niger Caranx tille

airbreathing catfishes

0

0

1

1

carp

0

0

1

1

golden guorami deep-bodied fussilier scissor-tail fussilier banded archerfish black pomfret Tille trevally

0 0 0 0 0 0

0 0 0 0 3 3

1 1 1 1 0 0

1 1 1 1 3 3

219

Rumahan bini Bakara Salman Selunsong Bilis Basong-basong Layaran Sulit Pulut-pulut Bakulan Barracuda Bubuk Jukut Keratang Kitang Lapih Rumahan laki Satak Sembilang Sotong

Scombridae Palinuridae Carangidae Latidae Leiognathidae Carangidae Istiophoridae Caesionidae Leiognathidae Scombridae Sphyraenidae Penaeidae Channidae Serranidae Scatophagidae Haemulidae Scombridae Scyllaridae Plotosidae Loliginidae

Udang galah

Palaemonidae

Udang siar

Penaeidae

Ungah

Lutjanidae

Usus Geronggong Anunan Bedukang Pisang-pisang Taweh Utik Tungap Kukut Rantau Pagong-pagong Besar Camau Mengkarai Sulai

Sillaginidae Carangidae Rhinobatidae Ariidae Lutjanidae Carangidae Ariidae

Rastrelliger brachysoma Panulirus ornatus Elagatis bipinnulata Lates calcarifer Leiognathus spp. Decapterus maruadsi Istiophorus platypterus Caesio spp. Leiognathus equulus Euthynnus affinis Sphyraena spp. Acetes spp./Lucifer spp. Channa spp. Epinephelus lanceolatus Scatophagus argus Diagramma picta Rastrelliger kanagurta Thenus orientalis Plotosus spp. Loligo spp. Macrobrachium rosenbergii Metapenaeus spp. Lutjanus argentimaculatus Sillago sihama Megalaspis cordyla Rhynchobatus djiddensis Hexanematichthys sagor Lutjanus lutjanus Alectis indica Arius oetik

short mackarel spiny lobster rainbow runner barramundi ponyfish Japanese scad Indo-Pacific sailfish fusiliers common ponyfish bonito barracuda shrimp common snakehead giant grouper spotted scat painted sweetlips Indian mackerel flathead lobster eeltail catfish squid

0 0 0 0 4 3 3 2 1 0 0 0 0 0 0 0 0 0 0 0

3 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

3 2 2 2 5 4 4 3 2 1 1 1 1 1 1 1 1 1 1 1

giant river prawn

0

1

0

1

prawn mangrove red snapper silver sillago torpedo scad giant guitarfish Sagor catfish bigeye snapper Indian threadfish catfish

0

1

0

1

0

1

0

1

0 2 1 1 1 1 1 0 0 0 0 0 0 1 1

1 0 0 0 0 0 0 2 0 0 0 3 1 0 0

0 0 0 0 0 0 0 1 1 1 1 0 0 0 0

1 2 1 1 1 1 1 3 1 1 1 3 1 1 1

* Scientific name for last 8 species could not be identified as the descriptions given by fishers were vague.

220

Appendix I-5.5 (Selected pelagic stocks’ trend from official data) Abundance of narrow-barred Spanish mackerel and Japanese scads between 2000 and 2009 based on standardised CPUE of official LS fishery data.

Japanese scads

Log-transformed CPUE (MT/day)

narrow-barred Spanish mackerel -4.00

0.00

-5.00

-2.00

-6.00

-4.00

-7.00

-6.00

-8.00 2000

2003

2006

2009

-8.00 2000

2003

2006

2009

YEAR

Test for linear trend in one-way ANOVA Narrow-barred Spanish mackerel: F9,289 = 26.264, p-value = <0.001 Japanese scad: F9,164 = 12.230, p-value = 0.001

221

Appendix J (Manuscript published)

Syazana Ebil, Charles R.C. Sheppard, Ranimah Wahab, Andrew R.G. Price, James C. Bull (2013) Changes in community structure of finfish catches in Brunei Darussalam between 2000 and 2009, Ocean & Coastal Management, Volume 76, Pages 45-51 (http://www.sciencedirect.com/science/article/pii/S0964569113000537)

222

Loading...

Download (6Mb) - Warwick WRAP - University of Warwick

University of Warwick institutional repository: http://go.warwick.ac.uk/wrap A Thesis Submitted for the Degree of PhD at the University of Warwick htt...

7MB Sizes 0 Downloads 0 Views

Recommend Documents

Download (34Mb) - Warwick WRAP - University of Warwick
ideas in the story or about spelling, grammar, pronunciation, etc. (Kurikulum. Semakan Bahasa Inggeris: Kursus Orientasi

Download (4Mb) - Warwick WRAP - University of Warwick
2.1 Basic operating principles of a steam power plant . ..... 2.10 Steam cycle for a typical supercritical power plant .

Download - Warwick WRAP - University of Warwick
of matter are prerequisites for any plausible account of coexistence and ..... Latour and the vitalist materialism of Gi

Download - Warwick WRAP - University of Warwick
methodology followed by three case studies, each of which recounts and analyses a series of lessons with three ..... to

Download (20Mb) - Warwick WRAP - University of Warwick
President Maria Estela Martinez de Perön (Isabel) to left-wing trade unions but. Balbin opposed him. Under the 1976 ...

Download (24Mb) - Warwick WRAP - University of Warwick
argentinos'. See Borges, El idioms de las augentinos (Buenos Aires: Gleizer, 1928). 7 For mote details on the discussion

Download (14Mb) - Warwick WRAP - University of Warwick
por um imperativo categorico da historia, pela sua acgao ultramarin em descobertas e conquistas, e ... "assimilados" and

Download (5Mb) - Warwick WRAP - University of Warwick
University in BOTH hard copy and in digital format. The digital version should normally be saved as a single pdf file. 2

Download (607Kb) - Warwick WRAP - University of Warwick
To see the final version of this paper please visit the publisher's website. Access to the published ... therefore, need

Download - Warwick WRAP - University of Warwick
2002-10-19-pdf >. 4 The slogan "Trentamila nel Duemila" was launched in 1999 through a series of meetings in all Italian