UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

In search of viable policy options for responsible use of sardine resources in the Bali Strait, Indonesia Buchary, Eny Anggraini 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2010_spring_buchary_eny.pdf [ 24.57MB ]
Metadata
JSON: 24-1.0069259.json
JSON-LD: 24-1.0069259-ld.json
RDF/XML (Pretty): 24-1.0069259-rdf.xml
RDF/JSON: 24-1.0069259-rdf.json
Turtle: 24-1.0069259-turtle.txt
N-Triples: 24-1.0069259-rdf-ntriples.txt
Original Record: 24-1.0069259-source.json
Full Text
24-1.0069259-fulltext.txt
Citation
24-1.0069259.ris

Full Text

IN SEARCH OF VIABLE POLICY OPTIONS FOR RESPONSIBLE USE OF SARDINE RESOURCES IN THE BALI STRAIT, INDONESIA  by ENY ANGGRAINI BUCHARY Ir. (Fisheries), Bogor Agricultural University, Indonesia, 1991 M.Sc., University of British Columbia, Canada, 1999  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Resource Management and Environmental Studies)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) March 2010  © Eny Anggraini Buchary, 2010  Abstract Traditional fisheries in developing countries are often marginalized from mainstream policymaking. This is crucial as many people depend on these fisheries for their livelihood. Using a case study of a traditional, medium-scale sardine (Sardinella lemuru) fishery employing paired purse seiners (slerek) in the upwelling ecosystem of the Bali Strait, Indonesia, the overall objective of this dissertation is to search for viable policy options for responsible use of the sardine resources. This is achieved by exploring issues in multiple domains: biological, ecological, social, economic and human dimensions. A synthesis of the official catch statistics shows that administrative inefficiencies and lack of good governance have created different versions of production statistics for the fishery. Not all sardine caught and landed are entirely reported; on average fishing-day, only about 45% of the catch is landed in government landing sites. Analyses show that the slerek fishery, contributes to the practise of illegal, unreported and unregulated (IUU) fishing. From 1950 to 2001, only half to one-third of what were actually caught were reported. A socio-economic analysis indicates that financial uncertainty and poverty are the main reasons for this IUU fishing practise. Poverty index of average slerek fishing households plunged from 25.7 in 2004 to -8.6 in 2008, as the loan interest rate was set up to 38% per annum and sardine are dwindling. Single-species and ecosystem-based assessments concluded that the slerek fishery has overexploited the sardine resources. Ecosystem analyses (Ecopath with Ecosim: EwE) using 20-year simulations (2001-20) suggest that climatic variability would increase the fishery; with a caveat of increased landing volatility. Insights derived from five harvest strategies tested using stochastic El Niño effects show that only 50% fishing effort reduction from the 2001 level could provide a sustainable option in the long term.  ii  Finally, an evaluation of the sustainability status of these harvest strategies was implemented using RAPFISH with a newly-proposed evaluation field, the human dimensions of traditional fisheries. Results show that trade-offs between economic and human dimension options are crucial in our case, as forgone values from human dimension option is more than what we can derive from choosing an economic option.  iii  Table of Contents Abstract ............................................................................................................................... ii Table of Contents ............................................................................................................... iv List of Tables ................................................................................................................... viii List of Figures .................................................................................................................... ix Acknowledgements .......................................................................................................... xiii Co-Authorship Statement................................................................................................ xvii 1 General Introduction and Review of the Bali Strait Fisheries ...................................... 1 1.1 General Introduction and Research Goals ........................................................... 1 1.2 Research Design in Data Collection and Analyses .............................................. 4 1.3 Final Disposition and Storage of the Raw Data ................................................. 14 1.4 Review of the Bali Strait Fisheries: Synthesis from the Fieldwork ................... 18 1.4.1 Area Description .............................................................................. 18 1.4.2 The Respondents and their Particulars ............................................. 20 1.4.3 The Fisheries System: Fishing Gear and the Stakeholders .............. 23 1.4.4 The Evolution of the Lemuru Fishery in the Strait .......................... 31 1.4.5 The Culture of Conflict Sown by Resource Appropriation Problems .......................................................................................... 34 References ................................................................................................................... 38 2 Beyond Fisheries Catch Statistical Data for the Bali Strait: the Story Behind the Numbers ...................................................................................................................... 42 2.1 Introduction ........................................................................................................ 42 2.2 The Slerek Fishery: Background on its Development and Statistics System .... 43 2.3 Some Challenges to Fisheries Catch Statistics .................................................. 47 2.4 Materials and Methods ....................................................................................... 51 2.4.1 Data Collection Methods ................................................................. 51 2.4.2 Analytical Methods .......................................................................... 52 2.5 Results and Discussion ...................................................................................... 53 2.5.1 The Fate of the Caught Fish Once Landed ...................................... 53 2.5.2 How the Landed Fish are Enumerated and Reported Throughout the System ........................................................................................ 58 2.5.3 Inconsistencies in Data Reporting ................................................... 62 2.6 Conclusions ........................................................................................................ 76 References ................................................................................................................... 78 3 Unaccounted Values: Under-reporting Sardine Catches as Part of a Strategy against Poverty in the Bali Strait, Indonesia ........................................................................... 89 3.1 Introduction ........................................................................................................ 89 3.2 Highlight of the Fishery ..................................................................................... 90 3.3 Materials and Methods ....................................................................................... 93 3.3.1 Data Collection Methods ................................................................. 93 iv  3.3.2 Analytical Methods .......................................................................... 93 3.4 Results and Discussion .................................................................................... 101 3.4.1 Official Records of Landed Lemuru .............................................. 101 3.4.2 Estimated True Catch ..................................................................... 102 3.4.4 Measuring Relative Poverty in Fisheries ....................................... 110 3.5 Conclusions ...................................................................................................... 111 References ................................................................................................................. 113 4 Ecosystem Analysis of the Bali Strait, Indonesia: Exploring Options under Uncertain Catch Reporting and Environmental Variability ..................................... 116 4.1 Introduction ...................................................................................................... 116 4.2 On Understanding the Ecosystem of the Bali Strait and its Fishery ................ 120 4.2.1 Previous Research .......................................................................... 120 4.2.2 On Fisheries Management Target: Realities and Challenges ........ 124 4.3 Materials and Methods ..................................................................................... 129 4.3.1 Data Collection Methods ............................................................... 130 4.3.2 Analytical Methods ........................................................................ 131 4.4 Results and Discussion .................................................................................... 155 4.4.1 The Ecopath Models of the Bali Strait: Past and Present-Day ...... 155 4.4.2 Fitting the Models using Times Series Drivers as Forcing Function ......................................................................................... 158 4.4.4 Exploring MSY Estimates: Single-Species in Contrast with Ecosystem-Based ........................................................................... 169 4.4.5 Exploring Options for Harvest Strategy ........................................ 182 4.5 Conclusions ...................................................................................................... 200 References ................................................................................................................. 203 5 Incorporating Human Dimensions of Fisheries in Appraising the Sustainability of Simulated Harvest Strategies for a Traditional Fishery: the Bali Strait Sardine Case.............................................................................................................. 215 5.1 Introduction ...................................................................................................... 215 5.2 Indonesia’s Fisheries Management: Some Highlights ..................................... 217 5.3 The Study Area – Bali Strait Ecosystem.......................................................... 220 5.4 Materials and Methods ..................................................................................... 222 5.4.1 Data Collection Methods ............................................................... 223 5.4.2 Analytical Methods ........................................................................ 224 5.5 Results and Discussion .................................................................................... 245 5.5.1 RAPFISH Analysis ........................................................................ 245 5.5.2 Comparison across Harvest Strategy and Climatic Effect Scenarios ........................................................................................ 253 5.6 Conclusions ...................................................................................................... 256  v  References ................................................................................................................. 258 6 Conclusions, Policy Implications and Recommendations ........................................ 266 6.1 Summary .......................................................................................................... 266 6.2 Key Findings, Recommendations, and Policy Implications ............................ 267 6.2.1 General Introduction and Review of the Bali Strait Fisheries ....... 267 6.2.2 Beyond Fisheries Catch Statistical Data ........................................ 269 6.2.3 Under-Reporting Sardine Catches as a Strategy against Poverty .. 272 6.2.4 Exploring Options under Uncertain Catch Reporting and Environmental Variability ............................................................. 275 6.2.5 Human Dimensions of Fisheries in Appraising the Sustainability Status of Fishery ............................................................................ 281 6.3 Concluding Remarks: Lessons from the Case of Bali Strait Sardine Fishery . 283 References ................................................................................................................. 289 Appendix A. Historical Overview of the Multi-Ethnic Communities of Muncar and Negara Districts that Shapes the Fisheries System in the Bali Strait. ....... 293 Appendix B. List of Various Research Permits from Both Local and Provincial Authorities in East Java and Bali; and Copy of Certificate of Approval from the Behavioural Research Ethics Board of the University of British Columbia (BREB-UBC). .................................................................................................. 296 Appendix C. Supplementary Material for Chapter 1. .............................................. 301 C1. Photographs of the Rented Living Space during the Fieldwork and of Some Typical Daily Activities. ........................................... 301 C2. Kaleidoscope of the Lemuru Fishery in the Bali Strait.................. 303 Appendix D. Stakeholders of the Bali Strait Lemuru Fisheries and their InterRelationships. ................................................................................................... 306 D.1 Various Stakeholders in the Bali Strait’s Lemuru Fisheries System as Observed During the Field Observations...................... 306 Appendix E. The Use of Triangular Distribution in Monte Carlo Resampling Routine ............................................................................................................. 312 Appendix F. Supplementary Material for Chapter 4. .............................................. 314 F1. Vulnerability Estimates from the Pre-Fitted Past Model, Computed as a Function of the Trophic Levels of the Groups. ..... 314 F2. Vulnerability Estimates from the Fitted Past Model, Estimated from the Optimal Search Routine. ................................................. 316 Vulnerability Estimates from the Present-Day Model, Computed F3. as an Adjustment to the Fitted Past Model, as such that for Each Predator-Prey Interaction, the Vulnerability * Predation Mortality should be Consistent across Time Period. ..................................... 319 F4. Species Composition of all Functional Groups of the Bali Strait Model in 1980. ............................................................................... 322 vi  F5. F6. F7. F8.  F9. Appendix G. G1. G2.  G3. G4.  Diet Composition Matrix for all Functional Groups of the Bali Strait Model in 1980. ..................................................................... 325 Landing (t⋅km-2) of all Functional Groups of the Bali Strait Model in 1980 Caught by 16 Different Fishing Gear. ................... 329 Time Series Forcing Function Used to Drive the 1980’s Ecopath Model of the Bali Strait During the Model Fitting Procedure. ...... 332 Primary Production Anomalies (PPA), 1980 – 2001, as Searched and Generated by Ecosim Using the 1980’s Ecopath Model of the Bali Strait during the Model Fitting. ........................................ 333 Time Series of Southern Oscillation Index (SOI), 1950 – 2008 (Bureau of Meteorology of Australia, 2009) ................................. 334 Supplementary Material for Chapter 5. .............................................. 336 Social, Economic and Cultural Views of the Bali Strait Slerek Fishery System ............................................................................... 336 Scores Used for the Ten Simulated Harvest Strategies for the Bali Strait Slerek Fishery in the Six RAPFISH Evaluation Fields .............................................................................................. 342 Details on How the Scores were allocated for Each Attributes and Scenarios ................................................................................. 348 Disaggregated Analyses on Each Attribute Scores in Different Scenarios for the Bali Strait Slerek Fishery in the Six RAPFISH Evaluation Fields ........................................................................... 372  vii  List of Tables Table 1.1  Types of fishing gear in the Bali Strait as noted during the field observations ................................................................................................. 23  Table 2.1  The effects of Indonesian geopolitical map changes to the Indonesian fisheries statistical areas: 1973 to present .................................................... 72  Table 4.1  Various studies on the biology and ecology of Bali sardinella (Sardinella lemuru) and its fishery ................................................................................ 120  Table 4.2  General stock assessment methods implemented for various marine fisheries resources in Indonesia .................................................................. 128  Table 4.3  Relative adjustment factors applied to the reported catch of non-lemuru species in the Bali Strait ............................................................................. 134  Table 4.4  Various single species stock assessment studies on the lemuru fisheries in the Bali Strait .......................................................................................... 144  Table 4.5  Input and output (in bold) parameters of the revised Ecopath model of the Bali Strait in 1980 (i.e., the ‘past model’) .................................................. 156  Table 4.6  Input and output (in bold) parameters of the Ecopath model of the Bali Strait in 2001 (i.e., the ‘present-day’ model) ............................................. 157  Table 4.7  Distribution of MSY estimates from this study using Fox model (nonequilibrium-fitting method, least-squares fit error model), and under different initial proportion and time-lag values ......................................... 172  Table 4.8  Relative biomass change attained under close-all fishing scenario compared to those attained under the other harvest strategy scenarios ..... 195  Table 5.1  The Bali Strait slerek fisheries analysed using RAPFISH, their codes in the MDS ordination graphs and types of climate scenario .................... 228  Table 5.2  RAPFISH attributes in their respective fields and notes on their scoring .. 230  Table 5.3  Proposed RAPFISH attributes for the new ‘Human Dimensions of Traditional Fisheries’ evaluation field ....................................................... 244  viii  List of Figures Figure 1.1  Research design model implemented in this thesis: a mixed methods approach that uses a concurrent triangulation strategy .............................. 5  Figure 1.2  The Bali Strait ............................................................................................ 6  Figure 1.3  The schematic diagram of water mass movement in the southern coast of Java and the Bali Strait ........................................................................ 20  Figure 1.4  Distribution of respondents based on age groups, recruited from both Jembrana and Banyuwangi regencies, on both sides of the Bali Strait ... 21  Figure 1.5  Distribution of occupations amongst respondents, recruited from both sides of the Bali Strait .............................................................................. 22  Figure 1.6  Types of occupation characteristics within the ‘fisher’ group. ................ 22  Figure 1.7  Bali Strait-style purse seines, locally known as Slerek............................ 27  Figure 1.8  Schematic diagram of a simplified representation of the network of stakeholder interdependence of the slerek fishery system in the Bali Strait. ........................................................................................................ 30  Figure 2.1  Schematic diagram of the fate of caught lemuru once landed ................. 58  Figure 2.2  Schematic diagram of the slerek fishery system in the Bali Strait that shows fate of landed lemuru, how landed lemuru are officially enumerated and how the statistics are reported and used throughout the official channels ...................................................................................... 63  Figure 2.3  Profile of reported/nominal annual catch (tonnes) of Lemuru (Sardinella lemuru) from the Bali Strait, 1950 - 2005, as recorded by various sources and agencies across jurisdictions ............................................................. 65  Figure 2.4  Distribution of reported/nominal annual catch (tonnes) of the Sardinella lemuru from the Bali Strait, 1950 – 2001 ................................................ 66  Figure 2.5  Fisheries management zones of Indonesia ............................................... 75  Figure 3.1  Distribution of reported/nominal annual catch (tonnes) of the Sardinella lemuru from the Bali Strait, 1950 – 2001 .............................................. 103  Figure 3.2  Estimated true catch (tonnes) of the lemuru from the Bali Strait, 1950 – 2001 ........................................................................................... 106  Figure 3.3  Some examples of lending model schemes for small-scale and mediumscale enterprises developed by the Bank Indonesia ............................... 108  Figure 3.4  Comparison of estimated range of debt-to-asset ratios of slerek fishing households (this study) after the increase of fuel price with the median of debt-to-asset ratios of all publicly listed companies in Indonesia, 2001 – 2004............................................................................................ 109  Figure 3.5  National poverty line of rural Indonesia, 1976 – 2008 ......................... 110  ix  Figure 3.6  Relative changes of poverty index (P-Index) amongst slerek fishing fleet owners in the Bali Strait, 2004 – 2008, as compared to Indonesia’s rural population .............................................................................................. 111  Figure 4.1  The influence of environmental variability in the Bali Strait on the lemuru fishery ........................................................................................ 117  Figure 4.2  Statistically significant correlations between the detrended lemuru catch anomaly (x 103 txkm-2) and the Southern Oscillation Index (SOI) of the El Niño episodes, 1950 – 2001 .............................................................. 118  Figure 4.3  Other environmental variability factors that affect the Bali Strait lemuru fishery ........................................................................................ 119  Figure 4.4  Time series of slerek fishing effort in the Bali Strait, 1974 to 2004 ...... 139  Figure 4.5  Standardized mean catch per unit effort (CPUE) of the sardine groups, 1974 – 2001 ............................................................................... 140  Figure 4.6  The number of slerek fleet operating in the Bali Strait (solid black line) and the slerek licenses capped by the governments of East Java and Bali province (dashed black line)............................................ 152  Figure 4.7  Adaptive fishing effort scenario: slerek harvest strategies were designed in proportion to the forecast SOI runs .................................... 154  Figure 4.8  Food web diagram of the Ecopath models ............................................. 157  Figure 4.9  Time series of predicted biomasses (line) and observed relative biomasses (blue dots) of the 3 functional groups of sardine in the 1980 model of the Bali Strait, with each group representing the growth stanza of the fish ........................................................................ 160  Figure 4.10  Time series of predicted yield (line) and observed yield (blue dots) of the 3 functional groups of sardine in the 1980 model of the Bali Strait, with each group representing the growth stanza of the fish ........ 161  Figure 4.11  Time-series of predicted: (A) biomass, and (B) yield of jellyfish functional group, as a result of fitting the 1980 Ecopath model of Bali Strait ........................................................................................... 163  Figure 4.12  Drying rack facility in Muncar, East Java, which was built specifically for jellyfish landing in 1997-98 .......................................... 164  Figure 4.13  Mixed trophic impacts on the jellyfish group as impacted group .......... 165  Figure 4.14  Twenty-year forecasting of SOI using bootstrap sampling (drawing with replacement) from the original 708 SOI data points, 1950 – 2008............................................................................................ 168  Figure 4.15  Relationship between annual CPUE of slerek and annual slerek fishing effort since the beginning of the fishery in 1974 ....................... 170  Figure 4.16  Relationship between initial proportion and: (A) sum of squares of residuals (SSQ), and (B) MSY, under two different time-lags ................................................................................................. 172 x  Figure 4.17  Box-and-whisker plots of comparative of statistics between 17 singlespecies surplus production MSY estimates (equilibrium fitting) from previous studies (Table 4.4) and single-species surplus production MSY estimates (non-equilibrium fitting) generated by 1,000 bootstrap replicates in this study ............................................................................ 173  Figure 4.18  One thousand separate bootstrap estimates of the MSY from the surplus production (Fox) model (non-equilibrium fitting method) of the Bali Strait’s lemuru stock, 1974 – 2001 ...................................... 174  Figure 4.19  Important diagnostic outputs from non-equilibrium fitting of the surplus production (Fox) model using least squared fit error model for the Bali Strait lemuru, 1974 - 2001 .................................................. 175  Figure 4.20  Ecosystem-based yield (red line) and biomass equilibriums (blue line) for the Bali Strait sardinella, Sardinella lemuru ............................ 179  Figure 4.21  Mixed trophic impacts in the past model of the Bali Strait ................... 180  Figure 4.22  Mixed trophic impacts in the present-day model of the Bali Strait ....... 181  Figure 4.23  Predicted biomass (tykm-2) from over-optimistic scenario .................... 183  Figure 4.24  Predicted yield (tykm-2) from over-optimistic scenario ......................... 184  Figure 4.25  Predicted biomass (tykm-2) from stakeholder scenario .......................... 185  Figure 4.26  Predicted yield (tykm-2) from stakeholder scenario ............................... 186  Figure 4.27  Predicted biomass (tykm-2) from government scenario ......................... 188  Figure 4.28  Predicted yield (tykm-2) from government scenario .............................. 189  Figure 4.29  Predicted biomass (tykm-2) from conservation scenario ........................ 190  Figure 4.30  Predicted yield (tykm-2) from conservation scenario ............................. 191  Figure 4.31  Predicted biomass (tykm-2) from adaptive fishing effort scenario ......... 193  Figure 4.32  Predicted yield (tykm-2) from adaptive scenario .................................... 194  Figure 4.33  Predicted biomass (tykm-2) from close-all fishing scenario................... 196  Figure 4.34  Annual predicted yields (tykm-2) of the sardine groups after 20 years of simulation across harvest strategy scenarios and SOI runs tested ...................................................................................................... 197  Figure 4.35  Total predicted yields (tykm-2) of the sardine groups in 20 years of simulation across harvest strategy scenarios and SOI runs tested .............................................................................................. 197  Figure 4.36  Predicted end-state biomass (tykm-2) of the sardine groups after 20 years of simulation, across harvest strategy scenarios and SOI runs tested ...................................................................................................... 198  Figure 5.1  Two-dimensional RAPFISH ordinations, in six labelled evaluation fields, of the ten simulated harvest strategies for the Bali Strait slerek fisheries .................................................................................................. 246 xi  Figure 5.2  Leverage (percentage influence on median) of individual attributes for six evaluation fields calculated from sensitivity analysis of the ten appraised simulated harvest strategies for the Bali Strait slerek fisheries ....................................................................................... 251  Figure 5.3  Uncertainty analysis, in six labelled evaluation fields, calculated from the RAPFISH ordinations of the ten simulated harvest strategies for the Bali Strait slerek fisheries .......................................... 252  Figure 5.4  Multidisciplinary kite representation of the evaluation of the Bali Strait slerek fisheries grouped by harvest strategy scenarios................. 254  Figure 5.5  Multidisciplinary kite representation of the evaluation of the Bali Strait slerek fisheries grouped by climatic effects scenario ........... 255  Figure C1.1  Rented living space during the fieldwork .............................................. 300  Figure C1.2  Typical daily activities in our ‘field office’ ........................................... 301  Figure C2.1  The horizon of the Bali Strait in: (A) March – April 1975, (B) August 2004..................................................................................... 302  Figure C2.2  The way the lemuru was handled and traded once the boat drops anchor is still the same, either in: (A) March – April 1975 with payang, or in (B) October 2004 with slerek ........................................... 303  Figure C2.3  As lemuru fishery evolved from using payang to slerek, the size of the boats also gets bigger, and so is the size of the baskets (godong) used to carry the fish .............................................................. 304  Figure E1  Example distribution of the error assumption used in the Monte Carlo 5,000 re-sampling routine ................................................. 312  xii  Acknowledgements  A journey completing a PhD degree feels like a long adventure at sea -- full of many surprises, pleasant and unpleasant ones, wonderful and dull moments, and even scary storms. I have been blessed to be accompanied and supported by many people along the way to make this dissertation possible, and now is my opportunity to thank them. First of all, I would like to express my deepest and most sincere gratitude to my thesis supervisor, Dr Tony J. Pitcher, who has been consistent in giving his academic guidance and moral support in a supportive and constructive manner. His spawning ideas, enthusiasm and healthy skepticism have continously challenged me left, right and centre. When he came to visit me in Bali when I did my fieldwork, he enlightened me with a neat technique in dealing with the uncertainty of respondents’ perspectives, which challenged not just me, but also my two field assistants. He remained motivated and optimistic, even during the most difficult part of this journey. He patiently read my chunk chapters to improve them into something more readable and presentable, and promptly and efficiently returned them to me. Last but not least, he had been providing me with various funding and material support throughout my PhD work, and opportunities to attend conferences and build my network. Terima kasih, Pak Tony! I must also thank my committee members for their contribution to and their interest in this work. I would like to thank Dr Ussif Rashid Sumaila for his cheerful enthusiasm and ever-friendly nature, his open-door policy, and most importantly, his encouragement and motto ‘Keep Pushing’ that have inspired me to keep a good fighting spirit especially at a time when I am at a low ebb. Dr Les Lavkulich has been my saviour, once again; he went beyond his obligation to help me in this journey. As usual, he always provided sound chairing and direction, and positive outlook in every committee meeting. For all of these, I especially want to express my most sincere gratitude to him. I also thank Dr I Gede Sedana Merta of the Research Centre for Capture Fisheries of Indonesia’s Ministry of Marine Affairs and Fisheries, my committee member who is based in Indonesia, for sharing his extensive knowledge and experience in the Bali Strait, and for his review and constructive feedbacks on my chunk chapters. A fateful event in 2000 during the FAOxiii  sponsored workshop on ‘Ecopath with Ecosim (EwE)’ at UBC Fisheries Centre and before I enrolled into this PhD program, has brought me to meet Dr Subhat Nurhakim, also from the Research Centre for Capture Fisheries of Indonesia’s Ministry of Marine Affairs and Fisheries. He has since become my ‘informal research advisor’ and my Indonesian mentor, and has been helping me all along since then in many ways. He was always available to be consulted even through text messages, and most importantly, the ‘surat sakti’ (powerful letter) that he provided me with and that accompany my every applications for local reseach permits in the study area has proven to be really powerful. For these, I express my hearfelt thanks. I would like to take this opportunity to thank all faculty and research associates at the UBC Fisheries Centre, for sharing information, ideas, thoughtful discussions, and interesting debates, during those Friday coffee breaks and seminars, and during our inhouse conferences, workshops, and symposiums where I was lucky enough to be one of those student rapporteurs, volunteers, or participants. I have been inspired by all of them. My dissertation has been greatly inspired by the fishing communities of the Bali Strait. My dissertation would not have been possible without their warm hospitality and generosity in sharing their life experience, their valuable time and their perspectives. Specifically, I would like to mention Bapak Haji Zainulllah Baijuri who was always more than happy to answer my questions, Bapak Haji Mursyid for inviting me and my two assistants to join his slerek fleet in a participatory fishing activity, Bapak Sudibyo SW who was very enthusiastic in providing data for this research, and three respondents (SB70, SB56 and SB73, all slerek owners) that kindly donated their own personal logbooks (buku bakul) to me during the fieldwork. I owe them all a big and endless ‘THANK-YOU’. Unwavering help and support from Abdul Muis and Uma Khumairoh, the two dedicated research assistants, are also greatly acknowledged, without which this research would not have been possible, and I am also indebted to them for giving me a learning experience to be ‘The Boss’. Ibu Sufat Meiwati of Muncar and Mbah Durahman of Pengambengan, the two wonderful land ladies that opened their homes to us, are also greatly acknowledged. The timely help of Bapak J.B. Susanto of Pengambengan is also  xiv  acknowledged for saving us from almost being ‘homeless’ in Muncar by introducing us to Ibu Sufat. In this Ph.D. journey, I have been helped by many people from academia, government, industry, communities and NGOs in different parts of the world, and my sincere gratitude to all of them. My special thanks go to: Dr Cameron Ainsworth of Northwest Fisheries Science Center-NOAA, Seattle, USA for his expertise and tacit knowledge on EwE modelling, Dr Dedi S. Adhuri of World Fish Centre, Penang, Malaysia, for giving me the lead to the meaning of rituals in fishing communities, Yusmansyah and Dewa Gede Raka Wiadnya of the University of Brawijaya, Malang, Indonesia for giving me access to the digital database of Bali Strait fisheries statistics, Dr Jackie Alder, formerly of the UBC Fisheries Centre, for giving me access to the entire copies of Indonesia’s fisheries statistics reports, Dr Jon Wittwer, the founder of Vertex42.com, for his online tutorial on Monte Carlo simulations using Excel software, Dawit Tesfamichael of UBC Fisheries Centre for showing me the nuts and bolts of RAPFISH, and Jeff Ferrara of GeoEye, USA, for his technical help on understanding IKONOS satellite imageries. My special thanks also go to John Nixon, Margaret North, Dr Robert North, and Kerrie O’Donnell for partially proof-reading parts of my thesis, and Nigel Haggan for proof-reading my defense slides. I am also very grateful to Janice Doyle and Ann Tautz at the UBC Fisheries Centre for their effective administrative support during my residency at the Centre. Former IT gurus at the UBC Fisheries Centre, Gerry O’Doherty and Rosalie Casison, also earned my deepest gratitude for making sure that my computers working all the time. Dr Syaiful Biran, DTM&H, MSc.CTM, SpPD-KPTI of Merdeka Medical Center of Denpasar, Bali, Indonesia, and Robert P. Reynolds, MD, FRCPC of Vancouver General Hospital are greatly acknowledged for their attentive and effective care during the most difficult time of this journey. Barbara Paulsen, RMT is also acknowledged for her care that had kept me strong and flexible to withstand the grueling long hours in front of the computer. I also would like to thank all my fellow graduates and colleagues at the UBC Fisheries Centre, from whom I learned so much, both professionally and personally. It has been a wonderful and life-enhancing experience to get to know you all and to work  xv  with some of you. I have made some good friends through the years, unfortunately, I can’t name them all. However, I am especially indebted to my past and present colleagues at Dr Pitcher’s PERF Research Group for the stimulating discussions, and to my ‘All-You-Can-Eat Sushi Dinner & Karaoke Gang’, with whom I often share fun times together to replenish our ‘mind and soul’: Dr Wiliam Cheung, Vicky Lam, Dr Jonathan Anticamara, Dr Marivic Pajaro, Louise Teh, Lydia Teh, Dr Deng Palomares, Dale Marsden, Dr Yajie Liu, Rhea Ravanera, and Grace Ong. I have also been blessed to have met Margaret North and Dr Robert North who provided me with a home away from home; you will always be my Canadian family forever. I am indebted to many previous and present house-mates at the Norths’ family home and to the Norths themselves, with whom I often first tested my research ideas on and got constructive criticism from. I must also thank Chad Osborn, who first helped me get settled in Canada and helped me to feel at home in this beautiful, cold country. Major funding for the 2004-05 fieldwork was provided by the Doctoral Research Award Grant (No. 101924-99906075-019) from the International Development Research Centre, Ottawa, Canada (www.idrc.ca). Additional funding supporting the fieldwork, the research-at-large and the PhD study are also acknowledged from: Graduate Student Research Assistantship from the Sea Around Us Project (2001-03), UBC Partial University Graduate Fellowship (2002-04), UBC Ph.D. Tuition Fee Award (2003-05), NSERC Graduate Research Fellowship (through Prof. T.J. Pitcher: 2003-04, 2005-06, and 2007-08), Canada Study Grant for Female Doctoral Students (2003-04, 2005-07), M. Dorothy Mawdsley Bursary (2006-07), Glen Cropsey Memorial Bursary (2006-07), UBC Bursary (2007-09), UBC Graduate Student Travel Fund (2002), UBC Fisheries Centre Trust Fund (2002), GLOBEC Travel Award (2008) and SSHRC Travel Award (2008). Last but not least, I would like to express my deepest gratitude to my dearest parents, Buchary Abdulwahid and Rini Kustrini Buchary, who have supported all my endeavours and never failed to offer help. To my sisters, Ina and Yanti, my brothers-in-law, Irfan and Dendy, and my nephews, Anif and Nadhif, thank you for everything. To my dearest Oca, I would like to give my sincere gratitude and love for his unwavering support, patience, sagacity, friendship and encouragements. You are all my rock.  xvi  Co-Authorship Statement  A version of Chapter 2, “Beyond Fisheries Catch Statistical Data for the Bali Strait: the Story Behind the Numbers”, will be submitted for publication with T.J. Pitcher as coauthor. A version of Chapter 3, “Unaccounted Values: Under-reporting Sardine Catches as a Part of a Strategy against Poverty in the Bali Strait, Indonesia”, is in press for publication as an invited chapter of a refereed Wiley-Blackwell book. T.J. Pitcher and U.R. Sumaila are co-authors on this paper. The detailed methodology and raw data spreadsheets of IUU analyses from this chapter will be archived as a contribution in the Fisheries Centre Research Report, and will not be co-authored. A version of Chapter 4, “Ecosystem Analysis of the Bali Strait, Indonesia: Exploring Options under Uncertain Catch Reporting and Environmental Variability”, will be submitted for publication with T.J. Pitcher as co-author. The 1980 and 2001 massbalanced Ecopath models developed in this chapter will be archived as a contribution in the Fisheries Centre Research Report, and will not be co-authored. A version of Chapter 5, “Incorporating Human Dimensions of Fisheries in Appraising the Sustainability of Simulated Harvest Strategies for a Traditional Fishery: the Bali Strait Sardine Case”, will be submitted for publication with T.J. Pitcher as coauthor. As the first author of each chapter that is in press or will be submitted for publication, I have been the core contributor in terms of identification and design of the research program, performing the research, conducting data analyses and synthesis, and preparing the manuscripts. The co-authors have all been helping in preparing the manuscripts.  xvii  1 General Introduction and Review of the Bali Strait Fisheries  1.1 General Introduction and Research Goals The majority of small- and medium-scale fisheries, in particular in developing countries, are traditional and artisanal, multispecies and multi-gear fisheries. The fisheries are known to employ many people and are crucial in providing food and livelihoods to a large number of people living in coastal areas. For example, in the year of 2000 alone, it was estimated that about 21 million tonnes of catches per year was landed by nearly 12 million small-scale fishers worldwide; a contribution between 25 to 33% of global marine fisheries landings (Chuenpagdee, Liguori, Palomares, & Pauly, 2006). Despite this significant contribution, small- and medium-scale fisheries are still quite marginalized from mainstream policymaking (Pauly, 1997); and many policies formulated are inadequate, unsustainable and biased towards band-aid solutions (Bailey & Jentoft, 1990). Lack of holistic understanding of the trade has precluded implementation of effective policies and management. Known for strong patron-client relationships and socio-cultural norms, these fisheries are also plagued by poverty. In fact, it has long been known that fishers and their families in this fisheries sector are generally among the poorest strata within Third World societies and often regard their livelihood as the option of the last resort (e.g., Smith, 1979; Bailey, 1988; Pauly, 1997). The marginality and disempowerment of the small- and medium-scale fisheries are such that they are vulnerable to uncertainties, volatility and other external dynamics that adversely affect their catches. Upwelling ecosystems, which are dominated by small pelagic species, are especially susceptible to uncertainties, volatility and external dynamics. One can therefore assume that small- and medium-scale fisheries that exploit upwelling ecosystems in the developing world must have developed a strategy to maintain their survival.  1  Indonesia’s fisheries are dominated by small- and medium-scale fisheries 1. For example, according to the data of Indonesia’s Ministry of Marine Affairs and Fisheries, by 2000, the numbers of registered small- and medium-scale fishing boats accounted for 99% of the total fishing boats, while large-scale fishing boats only account for less than 1%. In Indonesia, the Bali Sardinella (Sardinella lemuru, locally known as the 'lemuru') fishery in the Bali Strait is a medium-scale fishery that uses purse seine nets, hauled by a pair of boats as a tandem unit called a slerek. The fishery is known as an upwelling fishery and it serves as the major contributor to the overall lemuru landing in the country (Merta, 1995). Like many other upwelling fisheries, it has also experienced a 'boom and bust' nature affected by environmental variability such as El Niño (e.g., Ghofar & Mathews, 1996), resulting in disruptions to the economic and social well-being of the fishers that exploit the stock, and other people involved in the processing and marketing industry. Given their vulnerability to uncertainties, volatility and other external dynamics, the small- and medium-scale fisheries in the strait have developed strategies to ride out rough times. One particular adaptation strategy observed, for example, is to under-report their catches so as to pay less landing tax, which in turns will affect feed back to management policies and the sustainability of the resources, and eventually the livelihood of the fishing community in question. A multi-species stock assessment of the fisheries in the Bali Strait is lacking. There is also limited ecosystem research on the Bali Strait. Meanwhile, explicit recognition of environmental variability and uncertainties in the catch of the Bali Strait fisheries in management is lacking. Thorough understanding about the human-oriented realities of the fisheries is also lacking. All of these issues pose difficult tasks for scientists in 1  Priyono & Sumiono (1997) provided descriptions of the fisheries sectors in Indonesia. Small scale fisheries are defined as consisting of all fishing units which use sail-powered boats (i.e., dug-out boats and plank-built boats), and outboard engine-powered boats. As well, fishers who operate without any gear (i.e., gleaners) and fishers who operate fixed or mobile gears without any boat are assigned to small scale fisheries, regardless of the size of their gears. The second category is medium scale fisheries, which includes privately-owned (as opposed to company-owned) boats, of generally less than 5 gross tonnes (GT) and up to 30 GT, that use inboard engines. At the upper end of the category are the large scale fisheries, which include boats of >30 GT and beyond (i.e., to greater than 200 GT). The large scale fisheries are operated in isolation from the small and medium scale fisheries, and owned by a combination of private domestic (Indonesian) companies, joint-venture corporations, and stateowned enterprises.  2  assessing the impact of lemuru fishing in order to provide sound management advice and for the managers to formulate viable policy options for responsible use of the sardine resources. Therefore, this dissertation aims to fill this void by examining the following: 1. To explore and understand the dynamics of the fishery using a synthesis of historical accounts and socio-culture perspectives (Chapter 1 and 2); 2. To trace the time series of reported sardine catch data throughout the multiple jurisdictional levels of the complex Indonesian fisheries statistics system (Chapter 2); 3. To estimate the time series of the true sardine catch and calculate the unreported catch; and to examine the factors underlying under-reported and unreported catch (Chapter 3); 4. To explore of the extent of maximum sustainable yield that can be sustained by the ecosystem using single-species and ecosystem-based assessments (Chapter 4); 5. To analyze the ecosystem effect of the fishery given the estimated true catch and environmental variability, i.e., El Niño (Chapter 4); 6. To explore various trade-offs in policy options using ‘Ecopath with Ecosim’ ecosystem-based modelling; policy options are devised using multiple sources of information, including local knowledge and stakeholders’ perspectives (Chapter 4); and 7. To evaluate the sustainability status of these policy options using multidimensional scaling ordination, i.e., RAPFISH rapid appraisal system, that takes into account not only ecological factors, but also social, technological, economics and ethics, and a new field called ‘human dimension of traditional fisheries’ is introduced to help understand the dynamics of traditional and artisanal fisheries (Chapter 5).  3  1.2 Research Design in Data Collection and Analyses In general, developing country fisheries often lack a standardized and systematic way of doing things. The traditional and cultural context of the fishery make the study of these fisheries even more difficult than it already is due to the informal, kin-ship and patron-client nature of the enterprise. Therefore, in this thesis to attempt a holistic understanding and thorough knowledge of the system and issues, I implemented a mixed methods approach (sensu Creswell, 2003) for the research design. This approach combines both quantitative and qualitative approaches, in both the data collection and data analysis components. The strategy that I used to collect the data follows a concurrent triangulation strategy (sensu Creswell, 2003, Fig. 11.3a, p. 214; and see Figure 1.1), as a means to offset the weaknesses inherent within a single approach.  4  +  Quantitative (QUAN)  Qualitative (QUAL)  QUAN Data Collection:  QUAL Data Collection:  • Landing site censuses • Opportunistic surveys: fishing ports, fishing villages and fish markets • Desk studies on records, reports, archives, databases, interview & FGDs transcripts, taxonomy, toponymy & etymology records, and satellite imageries • Meta-analysis of models and other empirical studies  • Direct observation (living in the fishing community) • In-depth semi-structured interviews • Focus group discussions (FGDs) • Taxonomy, toponymy and etymology interviews • Showing video footage of underwater world • Participatory fishing • Post-fieldwork personal communication • Taking photographs & video footage of relevant object of observations • Keep field notes/journal during the fieldwork  QUAN Data Analysis: • Reported catch analysis (ch.2) • Estimating true catch (ch.3) • Monte Carlo simulations (ch.3, 4) • Poverty index analysis (ch.3) • Debt-to-assets ratio analysis (ch.3) • Time series analysis (ch.4) • Ecopath modelling (ch.4) • Bootstrap sampling (ch.4) • Ecosim simulations (ch.4) • Single-species & ecosystem-based stock assessment (ch.4) • Sensitivity analysis (ch.4) • RAPFISH (multi-dimensional scaling ordination) analysis (ch.5)  QUAL Data Analysis: • Historical analysis (ch.1, 2, 3, 5) • Documents/transcripts/archival analyses (ch.1, 2, 3, 4, 5) • Issue-focused analysis (ch.1, 2, 3, 4, 5) • Photography and video footage analysis (ch.1, 2, 3, 5)  Data results compared: to confirm, cross-validate and corroborate findings, and to find convergence Sequential form of data collection  +  ‘ch.’  Data triangulation & cross-validation Simultaneous or concurrent form of data collection = Chapter  Figure 1.1  Research design model implemented in this thesis: a mixed methods approach that uses a concurrent triangulation strategy (sensu Creswell, 2003).  5  Fieldwork in the study area was undertaken in order to understand how the fishery system works and to collect data, both quantitatively and qualitatively. The study area is administratively bounded by two provinces (Bali and East Java), and within these provinces, it is located in two regencies (Jembrana in Bali; and Banyuwangi in East Java). In Jembrana the centre of the fishery is in Negara district, and in Banyuwangi, the Muncar district (Figure 1.2)  Java  Sea  Madura Island Sampang  Banyuwangi Regency  East Java Province Muncar District Kedungrejo  Jembrana Regency Negara District Pengambengan  Bali Sea Bali Island  Lombok Island  Blambangan Peninsula  Indian Figure 1.2  Ocean  The Bali Strait. Muncar district on the east coast of East Java is the most important fishing port (with Kedungrejo village as its hub) for the Bali sardinella (lemuru) purse seine (slerek) fishery, while Negara district on the west coast of Bali is the second most important (with Pengambengan village as its hub). Both Banyuwangi and Jembrana regencies are delineated with dashed lines. Sampang village on the nearby Madura Island is where the ancestors of most Muncar-based slerek fishers originally came from (see Appendix A for details). Inset shows the archipelago of Indonesia with the location of the study area pointed by an arrow.  6  In total, I spent 147 days (~ 5 months) in the study area, comprising visits in 2002, 2004 – 2005 and 2006. In 2002, as dictated by local custom, courtesy visits to the local communities in the area to meet with community leaders, some community members, government officials, village councils, and local experts were conducted from June 12 – 15, 2002; June 29 – July 10, 2002, and July 15, 2002 (total 17 days). In addition, accompanied by local fishery officers, I toured the main fishing ports and fishing villages of Jembrana and Banyuwangi regencies on opposite sides of the Bali Strait. The main aims were self-introduction, explanation about the purpose of the study, getting guidance on how participants should be selected, and the identification of issues important to the local fishing communities: no formal interviews were conducted. Subsequently from 2004 through 2006, with two local research assistants 2, I spent an accumulated total of 130 days (4.3 months) in the study area, living within the fishing communities to get local data, first-hand accounts, some immersion experience and to better comprehend the issues 3. Pilot fieldwork 4 took place from March 18 – 21, 2004 and March 29 – April 13, 2004, while the actual fieldwork took place from July 14 to October 21, 2004 and February 13 – 19, 2005. A follow-up visit was carried out from December 14 – 16, 2006. Additionally, occasional communication (viz., telephone, email and text messages) with some respondents occurred after the field season was completed. In a triangulation of data results, results from thirteen data collection methods were compared and corroborated (Figure 1.1). The first data collection method was direct observation by living within the fishing communities throughout the entire fieldwork period. In the Negara district, we rented a living space in a house in Pengambengan 2  Mr. Abdul Muis (a Masters graduate in Coastal and Ocean Resources Management from Bogor Agricultural University, West Java, Indonesia, currently working as a management strategy consultant with MDConsults, in Jakarta, Indonesia) and Ms. Uma Khumairoh (a Bachelor graduate in Botanical Sciences from Brawijaya University, East Java, Indonesia, and is currently a Masters student at Wageningen University, the Netherlands).  3  Major funding for the fieldwork was provided by the Doctoral Research Award Grant (No. 10192499906075-019) from the International Development Research Centre, Ottawa, Canada (www.idrc.ca).  4  During the pilot fieldwork period, in addition to testing the data collection methods as will be explained later, my assistants and I also spent time in acquiring various research permits from both local and provincial authorities in East Java and Bali (Appendix B).  7  village (Appendix C1, Figure C.1.1a). We shared the house with the owner, an elderly widowed lady (60+ years old in 2004) with no children. Her family and extended relatives work in the lemuru fishery as boat crew and traditional fish flakes & fish oil makers. Meanwhile, in Muncar district we rented a living space in a house in Tembokrejo village (Appendix C1, Figure C.1.1b), whose owner is also a widow (40+ years old in 2004) who works in a fish meal plant/cannery as an accountant. She has two sons, aged 15 and 10 years old in 2004. These living spaces also served as our ‘field offices’ where we discussed and consolidated our logistic plans and conducted some of our desk-top works (Appendix C1, Figure C.1.2). Despite our busy schedules with our interviews and other research activities, we were involved in the daily lives of the village; for example, we regularly had our meals in local eateries and mingled with the locals, we were invited to lunches, dinners and Ramadhan fast-breaking meals. We were also invited to a local community meeting, I was once asked to become a designated photographer in a ceremony ritual for baby naming and first hair-cut, we organized a birthday party for one of our landlady’s sons, we joined communal praying gatherings in local mosques, we cast our votes 5 and observed the 2004 Presidential election held locally in Muncar district, and we also occasionally hung around with boat crew when they were mending the nets during the full moon period. The second method was individual in-depth semi-structured interviews using a set of questionnaires. Questionnaires using structures from Neis et al. (1999), Cheung & Sadovy (2004), and Ainsworth (2004) were developed and revised iteratively during pilot tests to reflect the Indonesian context. Fifty-one respondents were interviewed for an average of 2 to 3 hours, but some lasted for 4 to 5 hours, split into two days. The third method was 1 to 2 hour focus group discussions (FGD, Morgan & Krueger, 1998) with several (3-6) people of similar experience, age groups and gender. Pre-selected topics guided the discussions and techniques were revised iteratively during the pilot tests. A total of 41 discussants were reported from 9 discrete FGDs. In total, 5  Both of my assistants and I are Indonesian citizens and we registered ourselves in Muncar district to cast our votes for the 2004 Presidential election.  8  there were 92 respondents recruited for both interviews and FGDs, with four respondents participating in both sessions. To effectively access fishers’ knowledge, shared understanding of local terms for fish, fishing grounds and fishing gear is required (Johannes, 1993, cited in Neis et al., 1999). Therefore, in some in-depth interviews and FGDs, I also conducted taxonomy/toponymy 6/etymology 7 interviews, as a fourth method. These interviews aimed to collect and record, in the respondents’ mother-tongue language, the local common names of fish, marine mammals, seabirds and invertebrates, fishing gear, local fishing grounds, oceanographic, ecological and other fishery-related phenomena. Respondents freely explored a set of colourful posters of fish, marine mammals, seabirds, invertebrates, and fishing gear along with local nautical maps. Additionally, depending on the availability of electricity for our laptop, as a fifth method, I also presented some video footage 8 of underwater world of the tropics that show predator-prey relationship and ecosystem effects of fishing after we finished with the interview or FGD. This is aimed to explore further their opinions and to tease out certain issues that we just discussed in the interviews and FGDs. Separate courtesy visits explaining the project, anonymity and data rules were paid to all potential respondents.  6  According to Encyclopedia Britannica, Inc. (2008b), toponymy is defined as: “taxonomic study of place-names, based on etymological, historical, and geographical information. A place-name is a word or words used to indicate, denote, or identify a geographic locality such as a town, river, or mountain. Toponymy divides place-names into two broad categories: habitation names and feature names. A habitation name denotes a locality that is peopled or inhabited, such as a homestead, village, or town, and usually dates from the locality's inception. Feature names refer to natural or physical features of the landscape and are subdivided into hydronyms (water features), oronyms (relief features), and places of natural vegetation growth (meadows, glades, groves)”.  7  Etymology is defined by Encyclopedia Britannica, Inc. (2008a) as “the history of a word or word element, including its origins and derivation”. The inquiry of the etymology and translation of species common names from local languages to Indonesian and English followed the techniques developed by Buchary, Wahyuningsih, & Pauly (2003).  8  The length of the video shown was about 2 to 10 minutes per footage. Original source of video footage is from the “Blue Planet” series of the BBC (BBC Worldwide Ltd., 2002).  9  Prior verbal 9 consent to voluntary participation, audio-taping, and photography 10 were obtained on each day the interview and FGD took place. As information is unevenly distributed among members of fishing communities, being concentrated among skippers and retired fishers, particularly those with long careers and who are especially observant (Neis et al., 1999), I employed the ‘snowball sampling’ method (sensu Babbie, 1989, cited in Neis et al., 1999) to identify local experts for the interviews and FGDs. To adequately represent all community members, willing participants over the age of 18 11 were selected based on: (1) geographic characteristics such as location of villages where respondents reside, and location of fishing grounds or fishing related activities; (2) demographic characteristics such as age groups, occupation types (viz., fishers, non-fishers and local experts), gender and ethnic groups; and (3) fishing sector characteristics such as fishing fleet types, or whether the fishers are migratory or permanent. Non-local experts who do not reside in the area were also canvassed and recruited. The sixth method was participatory fishing, which arose as an invitation from fishers themselves. Many fishers in Muncar district believe that having student observers or researchers on their boat brings luck in one of the busiest fishing grounds in Indonesia. We received invitations from owners of a slerek purse seine (overnight, September 22 – 23, 2004), a banjang sero or tidal weir (low tide, October 2, 2004), and a bagan tancap or stationary bamboo platform lift-net (night of October 8, 2004).  9  Verbal consent is deemed more culturally appropriate in many communities in Indonesia, as many people have low levels of literacy and they would experience discomfort with a requirement to read and/or sign a document.  10  It is interesting to note that in several cases, even before we asked for their verbal consent, we were asked first by the would-be respondents if we would bring cameras; and when we said yes, we were asked to take family photos of them and their family members. We sent copies of these to them, as our courtesy.  11  In accordance with UBC Behavioural Research Ethics Board’s requirements, the age of majority in British Columbia (19) is used in this study. Meanwhile, the age of majority in Indonesia is 18 (male) and 15 (female) according to Indonesia’s Civil Code (Kitab Undang-Undang Hukum Perdata).  10  The seventh method comprised a census at fish landing sites to tally the overall number of catch weighing scales used in the area. Weighing scales are either official government scales located at government landing sites, or unofficial scales owned by private entrepreneurs and located outside government landing sites. Systematic censuses (1 in Negara district and 5 in Muncar district) were carried out from the back of a motorbike. The motorbike was driven relatively slowly to cruise along all streets once only for each census. An estimated area with a radius of 1-km (in Negara) and 3-km (in Muncar) from the fishing port was covered. The eighth method comprised opportunistic surveys of the fishing ports and fishing villages performed from the back of a motorbike. For example, in the back alley of a hamlet in Muncar district, I came across a small enterprise processing elasmobranch carcasses and fins, leading to agreement for interviews and data acquisition (E. Buchary, personal observation, 16 December 2006). Visits to local fish markets improved our knowledge of the local common names of fish and invertebrates and of the most common fish and invertebrates caught and landed. The ninth method was a desk study of various written records, reports, archives, databases, interview and FGDs transcripts, taxonomy/toponymy/etymology records, and satellite imageries. I focused on official reports and archives from government agencies, from village level (viz., fishing port/fish auction site) up to national level in Jakarta, and to international level at the headquarters of FAO-UN in Rome, Italy. I also reviewed fishers’ personal log-books 12, Indonesian newspaper archives, transcripts from interviews  12  Donated by some respondents during the fieldwork. Locally, these personal log-books are called “buku bakul” (buku = book, and bakul = basket, or vendor in Javanese and Madurese colloquial). As proof to financial accounting’s nuts and bolts of the fishing business, many fishers consider that these logbooks can bring local social disorder as much as bringing financial administration order for the person who owns the log-books. As such, many respondents admitted that they would usually destroy these log-books after certain fishing seasons once outstanding debts and financial obligations are reconciled. However, when interviewed in 2004, one particular successful fisher (who owns fleets of purse seine and long line) still keep his purse seine log-books series from decades ago for unknown reasons. When informed about the importance of the fishing log-books to this study and permission was requested to photocopy them, he answered that he would donate these log-books instead. Some other fishers also donated their log-books dated from non-continuous random years.  11  and FGDs, species records acquired during the taxonomy/toponymy/etymology interviews, research publications, various databases, and satellite imageries of the study area. The databases reviewed include: the Bali Strait fisheries catch database maintained by the Faculty of Fisheries of Brawijaya University, in Malang, East Java, Indonesia (Fakultas Perikanan UNIBRAW, Dinas Kelautan dan Perikanan Propinsi Jawa Timur, & Dinas Kelautan dan Perikanan Propinsi Bali, 2007), The Cephalopod Page (http://www.thecephalopodpage.org/),  SeaLifeBase  (http://www.sealifebase.org/),  FishBase (http://www.fishbase.org/), the archives database of the El Niño Southern Oscillation  Index  (SOI)  (http://www.bom.gov.au/climate/current/soihtm1.shtml),  CephBase (http://www.cephbase.utmb.edu/),  the Asian Development Bank database  (www.adb.org), the WorldBank database (www.worldbank.org), and the Bank Indonesia database (http://www.bi.go.id/sipuk/en/index.asp?id=4). Satellite imagery archives were also used to independently cross-check the number of the slerek boats moored at the landing sites and along the beach. Two sets of satellite imageries were acquired: (1) for the area of Muncar coast and neighbouring districts, the imagery was purchased from GeoEye (www.geoeye.com) 13: IKONOS Geo-1m resolution and  in  color,  metadata  source  ID:  2001012702383820000011623317  and  2001012702383820000011623317, dated January 27, 2001; and (2) for the area of Negara coast and neighbouring districts, the imagery was a GeoEye image (IKONOS Geo-1m resolution and in color, metadata source ID: 2003071902503300000011600883, dated July 19, 2003) that was distributed by i-cubed and uploaded at Yahoo Map and can be freely downloaded from Yahoo Map (http://ca.maps.yahoo.com/broadband#mvt= h&trf=0&lon=114.606001&lat=-8.4008&mag=1). The tenth method was meta-analysis of models and other empirical studies. These include  ecosystem  models  maintained  at  the  database  of  Ecopath  models  (www.ecopath.org) and those published at Christensen & Pauly (1993). 13  Helpful technical support from Mr. Jeff Ferrara of GeoEye, USA is gratefully acknowledged.  12  The eleventh method used was occasional personal communication, via cellular phone 14, email and text messages, with some respondents during and after the field season was completed to clarify, confirm and verify some information. Taking photography and video footage of relevant objects of observation were techniques employed as the twelfth method. These were implemented during participatory fishing activities (i.e., the sixth method), and various opportunistic surveys of the fishing ports, fishing villages and fish markets (i.e., the eighth method). The use of photography and video footage in the fieldwork is intended to provide more thorough documentation of the issues and as a visual tool to review, reflect on the issues and reappraisal after the fieldwork is completed (sensu Collier & Collier, 1986, and Marshall & Rossman, 1999, p. 124 - 126). The use of video footage was only implemented in the public area setting. Photography was used in both public area and private area (i.e., interview and FGD, the second and third method) settings. Prior consent to photography implemented during the interview and FGD was obtained on each day the interview and FGD took place. For the thirteenth method, I kept my own personal field notes/journal during the entire fieldwork to record various related incidents and information, and latter are used to review and reflect on the issues after the fieldwork is completed. Data triangulation between results obtained through different sets of data collection and data analytical methods was used to confirm, cross-validate and corroborate findings, and to find convergence (Figure 1.1). The triangulation process helps: (1) reduce data uncertainty in data collection and analysis, (2) reduce biases in respondents’ responses and memory recollection, and (3) to see pattern and propensity.  14  The use of cellular phone in the fieldwork is paramount considering that most public phones did not work and that land lines were not available in remote areas in Indonesia.  13  1.3 Final Disposition and Storage of the Raw Data This section is taken in whole from the final report (Ref#: H04-80176-003) that I submitted to the Behavioural Research Ethics Board of the University of British Columbia (BREB-UBC) on March 11, 2009. As explained in the Application for Behavioural Ethical Review form 15 (Box#29 in the form, Ref#: B04-0176, dated Feb 12, 2004) submitted to BREB-UBC, the raw data (i.e., interview audiotapes, including the transcripts, in original language), in accordance with UBC Policy #85 16, will be kept for at least 5 years after the work is published. These data will be kept by myself and by my supervisor (Dr Tony Pitcher) for future reference and will be stored under secured conditions, for example in computer hard disk and/or other electronic storage media with password protection, and in a locked cabinet within a premise with restricted access. If no follow up study takes place (see Box#30 in the form, Ref#: B04-0176, dated Feb 12, 2004), relevant hard copy documents of raw data will be shredded/destroyed/burned and relevant audiotapes will be demagnetized to ensure that confidentiality and anonymity will not be breached. Any other electronic data deriving and/or synthesized from the raw data will be retained by me under secured conditions as noted above. As described in the ethics application, some audiotapes are invaluable such as those containing traditional and local ecological knowledge and fishing history embedded in the oral history interviews with the community elders. To preserve this invaluable information for future research use, consent from all respondents if they would be willing to have these audiotapes archived at the UBC library system following the end of the study had been sought during the fieldwork and approval was granted. I had also explained to them about how and when it would be appropriate for others to have access to this archived information. Physical archiving in a library, while providing preservation for future use, provides limited access in a sense that future users need to specifically  15  Copy of the Certificate of Approval from BREB-UBC can be seen in Appendix B.  16  www.universitycounsel.ubc.ca/policies/policy85.pdf.  14  contact the library with their intention and come to the library (or order through interlibrary loan) to gain access to the audiotapes. This way would help filter the process that only future users with genuine research intention that would have access to these audiotapes. However, upon the completion of the fieldwork and after consulting with Ms. Sally Taylor (a reference librarian at UBC library specializing in fisheries sciences), we were advised that unlike some other university libraries, UBC has no mechanism/provision to archive interview audiotapes. We were told that the only available and relevant way of archiving raw data (like the kinds that we have) at UBC library is by way of electronic print-out archiving (i.e., pdf format or word document); in this case it is for the printed version of the interviews (i.e., the transcripts) submitted as an appendix of the thesis. We were not aware of the inability of UBC library system to archive audiotapes when we first submitted the ethics approval application in February 2004. Currently, the only way that a thesis is submitted at UBC is through electronic submission into a digital repository system (at cIRcle, www.circle.ubc.ca). The cIRcle system is openly accessible to anyone on the web, and allows direct and immediate global access to the thesis the minute right after the electronic submission. Appropriate measures to protect the anonymity and confidentiality of the respondents have been applied in the transcripts. In the transcripts, the interviews were all transcribed 17 in Bahasa Indonesia, the lingua franca of the study area and the language of the interviews. We believe that preserving the original language of the interviews in the transcripts would preserve the nuance of the context and perspectives of the issues. Selected specific sentences deemed important for the analysis were noted (in balloon style) with brief English translation, located on the right edge of the transcripts. To ensure anonymity, all names of respondents in the transcripts have all been changed into an 17  Transcription was conducted by research assistant, Mr. Abdul Muis, and was effectively completed in about six months time. With each interview lasting from 2 – 3 hours to 4 – 5 hours (in a few cases), and each FGD lasting from 1 – 2 hours, I ended up having 600+ pages of transcriptions (single space, Word document).  15  alpha-numeric coding system. Likewise, references to people’s names in the transcripts have also been changed into pseudonyms 18 and references to special locations, places, periods or conditions deemed to have sensitive nature (e.g., location of fish spawning grounds, location of fishing grounds, whether someone has been accused by the locals of committing a socially unacceptable act, etc.) and may potentially give adverse impact in the future for the local community in question, have also been concealed appropriately (i.e., blackened-out). However, given the immediate global access to cIRcle, we consider that this kind of digital repository system would give some vulnerability to the subjects mentioned in the transcripts. There is also the potential misuse of the transcripts by irresponsible researchers, i.e., plagiarism. We therefore plan the following as an alternative way to archive this information: (1)  Instead of submitting the transcripts as an appendix of the thesis, it will be archived as a supplement of the thesis, a separate document.  (2)  While the thesis will be submitted electronically to UBC cIRcle system once the degree program is completed, the supplement (containing selected parts of the interview transcripts 19, with full anonymity and confidentiality measures as noted above) will be embargoed until 5 years after the degree program is completed. In other words, there will be a 5-year delay in the disclosure of these selected parts of interview transcripts.  18  To avoid any similarity to any person living or dead in the study area, pseudonyms will be chosen from an unrelated cultural background (e.g., Anglo-Saxon). This is because some information provided may be locally sensitive and may potentially have unexpected consequences in the future.  19  These selected transcripts would contain invaluable information, such as traditional and local ecological knowledge and fishing history embedded in the oral history interviews with the community elders, as noted in the UBC ethics application and approved by the BREB-UBC (see copy of certificate in Appendix B).  16  The embargo period is intended to: (1)  Provide an extra degree of protection (in addition to anonymity and confidentiality measures noted above) to the subjects in question, in lieu of limited access provided by physical archiving in a library, and  (2)  Provide a protection from potential would-be plagiarists that may benefit from these transcripts had the transcripts is accessible globally right after the study is completed. It is hoped that a 5-year embargo period would give enough time for publications of the work.  Conclusively, we have informed BREB-UBC about this alternative plan for the final disposition and storage of raw data in the final report of the completion of behavioural study (Ref#: H04-80176-003, dated March 11, 2009), and it was approved on March 12, 2009. Additional note not included in the BREB-UBC final report (Ref#: H04-80176-003): Technical details on the submission of embargoed materials were discussed with UBC Library (Ms. Sally Taylor), UBC Faculty of Graduate Studies (Ms. Max Read) and my supervisor (Dr Tony Pitcher) following the thesis defense on February 10, 2010, and prior to digital submission of the thesis to cIRcle on March 4, 2010. Ms. Taylor offered me two options for the submission of the embargoed document to cIRcle: (1)  Submit the embargoed supplementary document together with the thesis on March 4, 2010 to cIRcle, but an automatic 5-year embargo protection would be applied to the supplementary document;  (2)  Submit the embargoed supplementary document in five (5) years time to UBC cIRcle, and in the mean time, I shall be responsible for looking after the data set during the five (5) years embargo time by keeping the data set under secured conditions as noted above.  17  Notwithstanding the secured measures that have been applied to UBC’s cIRcle digital information repository system, I still feel not comfortable to take a risk in submitting the embargoed document together with the thesis in March 2010. Therefore, I have decided to choose option #2. As agreed in the discussion, in five (5) years time, I shall contact the cIRcle Coordinator 20 at the UBC Library to deposit it into cIRcle then.  1.4 Review of the Bali Strait Fisheries: Synthesis from the Fieldwork 1.4.1 Area Description The Bali Strait is a 3,126 sq.km funnel-shaped marine environment that is located between the island of Java and Bali, in central Indonesia (Figure 1.2) adjacent to the Indian Ocean in the south and the Bali Sea in the north. It is shallow (ave. depth = 50m) in the northern part while deeper (~ 400m to 1,400m) in the southern part; the northern opening is about 2.5 km wide, while the southern is wider at 55 km (Merta, Widana, Yunizal, & Basuki, 2000). The oceanographic conditions of the strait are influenced by bi-annual monsoons (Wyrtki, 1961, cited in Ritterbush, 1974; and various accounts from respondents) and El Niño and La Niña events that occur interchangeably every 2 to 7 years (Climate Assessment for the Southwest (CLIMAS), 2003; and various accounts from respondents). The northwest monsoon that lasts from November to March brings winds that generate a coastal current that flows along the south coast of Java towards the east. Meanwhile, the southeast monsoon lasts from June to October where the coastal current goes in the opposite direction, from east to west as it is pushed by the south equatorial current of the Indian Ocean, and reaches maximum around July (Burhanuddin & Praseno, 1982, and Saliyo, 1973; cited in Merta et al., 2000) when it generates an upwelling in the area south of Sumbawa, Lombok, Bali and the eastern tip of East Java (Figure 1.3). However, recent surveys (in 1997: Merta et al., 2000; and in 2004: Hendiarti et al., 2005) indicate that the timing of the upwelling process (denoted by lowest average sea surface 20  Currently held by Hilde Colenbrander (hilde.colenbrander@ubc.ca).  18  temperature) may be delayed to August. The months of April and May represent an intermonsoonal period where currents become weak. Although most respondents were not aware of the term ‘El Niño’ per se, they fondly remember the year of 1997-98, noted for its major El Niño (Ghofar, Mathews, Merta, & Salim, 2000), when the amount of sardines caught was beyond their imagination to the point that some boats sank with overloaded fish. The sardine fishery therefore has a boom and bust nature as expected for small pelagic planktivorous fish driven by El Niño and La Niña events, respectively. The two monsoons described above govern the seasonality of the sardines as the upwelling brings nutrients enhancing phytoplankton and consequently zooplanktons, and eventually fish. The SE monsoon is considered as ‘the fish season’ by the fishers in the strait, with the glut peaks around September to October, and occasionally the glut would continue to the early part of November. Off season occurs during the NW monsoon and the inter-monsoonal period. It is quite common that many purse seines would get as little as 0.5 ton of sardines/fleet/fishing trip during the off season. On the contrary, a maximum landing of 15 – 20 tons/fleet/fishing trip during the regular glut period and 30 – 40 tons/fleet/fishing trip during El Niño years (like that in 1997-98) are quite normal; meanwhile, discarding and high-grading are also common during the glut and El Niño. Slerek fishing trips are conducted for 11 to 20 days each month during new moon 21 phases only, with an average of 15 fishing-days/month. A fishing-day is normally comprised of one trip, although 2-3 trips/fishing-day is not uncommon during the glut and El Niño seasons. During full moons, the crew members work on the land to mend nets and do necessary repairs.  21  The fishing cycle for sardines is lunar. Skippers and Captains acknowledged that the sardines cannot be fished when the moon is full, because it is hard to visually find them under the moonlight. Their iridescent bodies would glow in the water, making it difficult to see them in the moonlight. Thus, all lemuru fishing is done during the new moon phases, when the moon is dark.  19  EAST JAVA  EAST JAVA  B A L I  B A L I  SEC  CCSJ  UPWELLING  A Figure 1.3  B  The schematic diagram of water mass movement in the southern coast of Java and the Bali Strait (Nontji & Ilahude, 1975, cited in Wudianto, 2001): (A) during NW monsoon, and (B) during SE monsoon. CCSJ = wind-driven coastal current of southern Java, SEC = south equatorial current of the Indian Ocean. Reprinted and modified from Wudianto (2001) with permission.  1.4.2 The Respondents and their Particulars Ninety-two respondents (18 females and 74 males, excluding non-local experts) were locally recruited during the fieldwork using snow-ball sampling. The main reason for the unbalanced gender ratio recruited is because males dominate the activities related to fishing and fisheries in the strait. The youngest age recruited was 25 years old, while the oldest was 80+ years old. Generally, respondents recruited were biased (n = 26) toward 40 – 49 age group (Figure 1.4). The 92 respondents were living in 15 coastal villages and 2 towns, spread in two regencies and within two provinces. Meanwhile, information was also sought from nonlocal experts who have intimate knowledge of the Bali Strait fisheries and its ecosystem. Unlike the names of the locally-recruited respondents which shall remain anonymous (see section 1.3)18 for privacy and confidentiality reasons, names of these non-local experts will be disclosed and annotated appropriately.  20  Figure 1.4  Distribution of respondents based on age groups, recruited from both Jembrana and Banyuwangi regencies, on both sides of the Bali Strait.  One parameter for respondents’ recruitment was the type of main occupation: in total there are 15 types canvassed for the interviews and FGDs. ‘Fisher’ (or nelayan in Indonesian) is the occupation that has the highest number (n = 38) of respondents (Figure 1.5). In the Indonesian language, the context of nelayan is quite broad and does not exclusively include only those people that catch fish for their livelihood, but also those people that own the fleets. This is because most fleet owners started their career as either ordinary crew members or individual operators. Being a fleet owner can be considered as the top achievement of a nelayan. Most crew members, individual operators and skippers interviewed all aspire to be fleet owners one day. When this ‘fisher’ group is further broken down into specific occupation characteristics (Figure 1.6), ‘individual operator’ is the most frequent (n = 16).  21  Figure 1.5  Distribution of occupations amongst respondents, recruited from both sides of the Bali Strait.  Figure 1.6  Types of occupation characteristics within the ‘fisher’ group. In the Bali Strait, a skipper is also the captain of the boat that he operates. Some fleet owners also become skippers of their own boats.  22  1.4.3 The Fisheries System: Fishing Gear and the Stakeholders There are sixteen types of fishing gear operating in the Bali Strait and two types of fishing gear mooring in the strait but operating elsewhere (Table 1.1). Table 1.1  No.  Types of fishing gear in the Bali Strait as noted during the field observations: June – July 2002, March – April 2004, July – October 2004, February 2005 and December 2006.  Fishing gear (English, local name)  Target species (Latin, English, Indonesian or local name)  Fishing gear that moor and operate in the Bali Strait: 1.  Pair-boat system purse seine (slerek, Figure 1.7)  During SE monsoon: ƒ Mainly Sardinella lemuru (lemuru sardines; lemuru), all agegroups During NW monsoon, a random mix of: ƒ Decapterus spp. (scads; layang), all age-groups ƒ Euthynus affinis (eastern little tuna; tongkol) ƒ Scomberomorus spp. (mackerel, tenggiri)  2.  Fixed lift-net attached to a bamboo platform and operated with kerosene lamps (bagan tancap)  A random assortment of: ƒ Loligo spp.(squids; cumi) ƒ Juveniles (sempenit) and sub-adults (protolan) of Sardinella lemuru, especially after spawning season (in June) ƒ Some shrimps ƒ Various small demersal fish, such as Leiognathus spp. (splendid ponyfish, peperek)  3.  Scoop or dip-net (serok)  Mainly used in the 1940s – 1970s, it is rarely used now. However, some small-scale fishers still use this gear. Target species are a random assortment of: ƒ Loligo spp.(squids; cumi) ƒ Juveniles (sempenit) and sub-adults (protolan) of Sardinella lemuru, especially after spawning season (in June); also a bit of the adults (lemuru) ƒ Other small pelagics ƒ Various small demersal fish, such as Leiognathus spp. (splendid ponyfish, peperek)  4.  Gleaning (cari kerang dan bulung)  These are mainly done by women, sometimes accompanied by their children. They glean by using simple tool (e.g., screw driver) or with no tool at all. Their target species are seaweed (bulung) for those gleaners on the coast of Negara district, and benthic invertebrates, especially molluscs (kerang) for those gleaners on the coast of Muncar district.  23  Table 1.1  Types of fishing gear in the Bali Strait as noted during the field observations: June – July 2002, March – April 2004, July – October 2004, February 2005 and December 2006.  No.  Fishing gear (English, local name)  Target species (Latin, English, Indonesian or local name)  5.  Push net operated using a small boat equipped with an outboard engine (slodo mesin, or simply slodo)  A random assortment of: ƒ Juveniles (sempenit) and sub-adults (protolan) of Sardinella lemuru, especially after spawning season (June) ƒ Sergestid shrimps ƒ Various small demersal fish  6.  Small seine net (payang oras)  During SE monsoon: ƒ Mainly Sardinella lemuru (lemuru sardines; lemuru), all agegroups During NW monsoon, a random mix of: ƒ Decapterus spp. (scads; layang), all age-groups ƒ Euthynus affinis (eastern little tuna; tongkol) ƒ Scomberomorus spp. (mackerel, tenggiri)  7.  Large seine net (payang Panarukan)  During SE monsoon: ƒ Mainly Sardinella lemuru (lemuru sardines; lemuru), all agegroups During NW monsoon, a random mix of: ƒ Decapterus spp. (scads; layang), all age-groups ƒ Euthynus affinis (eastern little tuna; tongkol) ƒ Scomberomorus spp. (mackerel, tenggiri)  8.  Handline (pancing)  Mainly demersal fish, such as: ƒ Trichiurus spp. (hairtails; layur) ƒ Epinephelus spp.(groupers; kerapu) ƒ Various species of elasmobranch juveniles (sharks; cucut, and rays; pari) ƒ Lutjanus spp. (snappers, kakap) ƒ Nemipterus spp. (sweetlips, kurisi)  9.  Tidal weir with nylon net (banjang sero or banjang waring)  Anything that got caught up in the weir during the ebb and flow cycles, particularly: ƒ Various kinds of shrimps: Metapenaeus spp. (udang kecrek, udang manis putih); mysids shrimps (gerago) ƒ Portunus pelagicus (swimming crabs; rajungan) ƒ Leiognathus spp. (splendid ponyfish, peperek) ƒ Juveniles (sempenit) and sub-adults (protolan) of Sardinella lemuru, especially after spawning season (June)  24  Table 1.1  No.  Types of fishing gear in the Bali Strait as noted during the field observations: June – July 2002, March – April 2004, July – October 2004, February 2005 and December 2006.  Fishing gear (English, local name)  Target species (Latin, English, Indonesian or local name)  10.  Traditional lobster diving, using goggles, net and spear (selam udang barong dengan jaring dan tombak)  Panulirus spp. (reef lobsters; udang barong)  11.  Multifilament nylon drift gill net (jaring insang, jaring gilnet)  Mainly target Katsuwonus pelamis (skipjack tuna; cakalang, benglong), but also Euthynus affinis (eastern little tuna; tongkol) and the juvenile and sub-adult (size 2 – 5 kg/fish) of Thunnus albacares (baby tuna). Also catch elasmobranch (sharks; cucut, and rays; pari), small and medium pelagics (including S. lemuru), and a bit of demersal species (small, medium and large).  12.  Monofilament nylon bottom gill net (jaring senar)  Mainly target medium and small pelagics (including S. lemuru), large, mediun and small demersals species, crabs, shrimps and lobster, and a bit of elasmobranch (sharks; cucut, and rays; pari).  13.  Muncar-based one-boat system purse seine (perahu gardan) – smaller in size than the migrant one-boat system (see below #16)  Target species is similar to slerek fleets. During SE monsoon: ƒ Mainly Sardinella lemuru (lemuru sardines; lemuru), all agegroups  14.  Beach seine (pukat pantai)  During NW monsoon, a random mix of: ƒ Decapterus spp. (scads; layang), all age-groups ƒ Euthynus affinis (eastern little tuna; tongkol) ƒ Scomberomorus spp. (mackerel, tenggiri) More or less similar to slerek, but with different proportion. During SE monsoon: ƒ Mainly Sardinella lemuru (lemuru sardines; lemuru), mainly sub-adult and juvenile groups During NW monsoon, a random mix of: ƒ Decapterus spp. (scads; layang), all age-groups ƒ Euthynus affinis (eastern little tuna; tongkol) ƒ Scomberomorus spp. (mackerel, tenggiri) ƒ Also some small and medium demersal fish ƒ Various kinds of shrimps: Metapenaeus spp. (udang kecrek, udang manis putih) ƒ Portunus pelagicus (swimming crabs; rajungan) ƒ Various species of elasmobranch juveniles (sharks; cucut, and rays; pari)  25  Table 1.1  Types of fishing gear in the Bali Strait as noted during the field observations: June – July 2002, March – April 2004, July – October 2004, February 2005 and December 2006.  No.  Fishing gear (English, local name)  Target species (Latin, English, Indonesian or local name)  15.  Turtle hunter (pemburu penyu)  Local Balinese turtle hunters in the strait used to be rampant up to 1980s. However, by late 1980s, their numbers declined. Conservation awareness is credited to their demise. In 2004, turtle hunters still exist (only on the coastal areas in Negara district) in very small numbers and only target few turtles, about 10 turtles/year (mainly green turle, Chelonia mydas) for Balinese religious ceremonial purposes only (Respondent SB38, Respondent SB39, and Respondent SB41, focus group discussion, 21 August 2004).  Fishing gear that come as migrant fishing fleets, temporarily moor and operate in the Bali Strait: 16.  Migrant one-boat system purse seine (kapal Tubanan)  During SE monsoon: ƒ Mainly Sardinella lemuru (lemuru sardines; lemuru), all agegroups During NW monsoon, a random mix of: ƒ Decapterus spp. (scads; layang), all age-groups ƒ Euthynus affinis (eastern little tuna; tongkol) ƒ Scomberomorus spp. (mackerel, tenggiri)  Fishing gear that moor in the Bali Strait but fish elsewhere: 17.  Traditional longliner, operated using wooden boats (rawe mungsing)  Various species of adult sharks from Carcharhinidae, Sphyrnidae, and Orectolobidae families. Fishing grounds are mainly in Indonesian waters 22 not far from Australian waters. Also in waters in the vicinity22 of Kalimantan and Sulawesi islands (Respondent SB70, personal communication, 16 December 2006).  18.  Modern longliner, operated using large fiberglass boats equipped with modern equipment (rawe tuna)  Various species of tuna, such as Thunnus alalunga (albacore tuna), Thunnus obesus (bigeye tuna), and Thunnus maccoyii (southern bluefin tuna). Fishing grounds are mainly in eastern Indonesia, such as the Banda Sea, Arafura Sea and Seram Sea.  Note: See Appendix A for descriptions on ethnic-based specialization with regards to the ownership and operating of fishing gear.  22  As per the UBC-BREB ethics protocol of my research, the exact fishing ground location is herewith concealed although they were disclosed during the interview.  26  Despite this array of distinct fishing gear targeting different species, in terms of bulk catch landed, Sardinella lemuru (lemuru sardines) is the dominant species with more than 90% of total catch (Hendiarti et al., 2005; based on official records). Meanwhile, narrative evidence from interviews suggests that it is about 75 – 80% of the total catch on average. These are mainly exploited by the pair-boat system purse seine (Slerek, Figure 1.7; Hendiarti et al., 2005; and various accounts from respondents).  Float  Float line  Net, ¾ - 1"  Weight  Weight line  Ring Purse line Boats  Figure 1.7  Bali Strait-style purse seines, locally known as Slerek, are usually composed of a pair of boats, the net boat (left, see photo) and the fish boat (right, see photo) that work in tandem (see diagram). Net hauling is operated manually by 25 to 50+ crew (in 2004), and fish schools are eye-spotted by the skipper/Captain sitting at the high bench, also called ‘the stage’ (see red arrow) of the net boat. The engines are mounted and ‘clipped’ on the edge of the boat (blue circle) and the long shaft of the propeller would go sideways; each fleet would have 5 – 7 engines in 2004 at 30 horsepower (HP)/engine, totalling to 150 – 210 HP/fleet. Needing more speed to chase the fish schools, the net boat would have 3 – 4 engines, while the fish boat would have 2 – 3 engines. Schematic diagram is from Wudianto (2001), modified and reprinted here with permission. Photo credit: Eny Buchary, in Pangpang Bay, Muncar, 2004.  Administratively speaking, the Bali Strait fisheries are managed by both Banyuwangi Regency, in the province of East Java, and Jembrana Regency, in the province of Bali and there has been a joint licensing agreement between the two provinces since 1977 (Wudianto, 2001). This joint agreement was formalized into a fishery management body  27  in 2000 comprising stakeholders from both provinces, overseen by both provincial and regency governments (FAO-Norway Government Cooperative Programme, 2001). However, regular and meaningful working collaboration is minimal mainly due to lack of funding (S. Nurhakim 23, personal communication). In addition, awareness and understanding of how a fishery management body should and can function amongst government officials involved is lacking (Respondent SB65, personal communication, 28 September 2004). The fisheries system in the Bali Strait can be regarded as anything but a homogeneous and streamlined system. The fishery system relies on a crowded and labour-intensive network of stakeholder interdependence (Figure 1.8), which receives and shares the value of the catch (and even the by-products and wastes) in an intensely distributive ramified fashion, from the moment the fish are caught, landed, bought, sold, re-bought and re-sold, and so on (see Appendix D for descriptions on the stakeholders and their inter-relationships). The network is very efficient in making use of the fish resources caught and landed. Almost everything is used and processed into something marketable; even the fish oil waste flushed by the various fish meal plants and canneries into effluent pipes are strained and processed into industrial-grade grease for distribution and sale into regional market. This network system has always been there; it is more crowded now, but perhaps as labour-intensive as it was when Emmerson (1980, Table 4, p.50) observed the fishery in 1975. Therefore, a simple perturbation in one link of the network would have a serious consequence of disrupting the whole web of interdependence. The fact that the fisheries in the strait are very much shaped by the history of the multi-ethnic community that build Muncar and Negara districts (Appendix A), when resource appropriation is deemed unfair, inter-gear and thus, inter-ethnic violence could erupt and turned ugly (see section 1.5.5).  23  Senior fisheries scientist at the Research Centre for Capture Fisheries of Indonesia’s Ministry of Marine Affairs and Fisheries  28  The network of stakeholder interdependence (Figure 1.8) develops over time and has formed into a community safety net. Its intricacies are nurtured by the prevalent patron-client relationships and strong kinship, which made up the foundation of a social safety net. At individual level, fishers and their families would specifically invest in gold jewelries 24 for their wives during the southeast monsoons and El Niño events as a store of wealth and as a personal safety net 25, to help buffer themselves against uncertainties of income during the northwest monsoons and La Niña events when fish are scarce. The lack of formal government social safety net, inaccessible financial resources, lack of other livelihood options and the prevalent cyclical boom and bust nature of the fishery have enabled the fishing community to evolve in such a way so that social safety net, as embodied by the patronage institution, provides a place where both personal safety net and community safety net rest upon and thrive so that they help buffer the community against any shocks and uncertainties.  24  Other types of store of wealth were also observed during the fieldwork in the study area, for example, land for agriculture (e.g., rice fields), motorcyles, cars, luxury home furnishing, etc. However, interview accounts suggest that since gold jewelries have always been the easiest type of investment to sell and/or pawn in short notice, fishers’ wives prefer to invest in gold jewelries rather than in any other types of store of wealth.  25  During the interview session with Respondent SB70 (personal communication, 5 October 2004), who is an owner of two pairs of slerek boats and a patron to all crew members and the two skippers, his wife admitted that, with her husband’s approval, she would buy gold jewelries during the glut seasons, notably during southeast monsoons and El Niño episodes, for the wives of all crew members and the wives of the two skippers. She asserted that she did this as a means to encourage them, as their clients, to buffer themselves against uncertainties by having some gold as personal safety net.  29  Figure 1.8 Schematic diagram of a simplified representation of the network of stakeholder interdependence of the slerek fishery system in the Bali Strait. See Appendix D for descriptions on the stakeholders.  30  1.4.4 The Evolution of the Lemuru Fishery in the Strait Respondent SB23 who was 80+ years old during the interview explained that the lemuru fishery had started in the strait even before Japanese troops came to the area in 1942. At the time, scoop or dip nets (serok) and small dug-out boats (sampan or jukung) with sail were normally used. The dug-out boat had a maximum capacity of 0.2 – 0.3 tonnes when fully loaded. Most lemuru caught then were usually the largest size (representing the oldest age-group or senescent stage), known locally as bei kocing or lemuru kucing, about 20 – 25 cm long (Respondent SB23, personal communication, 10 August 2004). During the National Revolution period (1945 – 49) 26, fishing in the strait was temporarily halted by Dutch troops by means of force. The temporary halt was enough to make all sorts of fish so abundant thereafter that fishers could catch them using cast nets (jala buang) and handlines (pancing) from the beach (Respondent SB23, personal communication, 10 August 2004). After the revolution, during the 1950s to 1960, the lemuru fishery in the strait grew slowly. Soemarto (1960) described various gear that were used then: (1) jala oras or payang oras (seine nets made of cotton and leaves of Gebang palm, Corypha utan; operated by wooden sailing boats ranging in size from (Length x Breadth x Depth): 5.5m x 0.8m x 0.5m to 7.5m x 1.5m x 0.75m, with 3 – 6 crew members, including the skipper), (2) jaring lemuru or jaring eder (cotton gill nets, also operated using the same wooden sailing boats), (3) jala buang (cast nets made of cotton, operated using dug-out boats that were 4m x 0.6m x 0.5m in size, with 1 – 2 crew members), and (4) serok (scoop or dip nets made of cotton thread, operated using the same dug-out boats).  26  Following Japan’s surrender in World War II on August 15, 1945 and Indonesia’s proclaimed independence on August 17, 1945; the Dutch troops came back to Indonesia’s archipelago to reestablish their rule. It was not until December 1949 that the Dutch finally surrendered and left the archipelago. This period, 1945 – 1949, was known in Indonesia’s history as a National Revolution period, and Respondent SB23 remembered this period as a chaotic and bloody time.  31  Slerek purse seines were introduced in the Bali Strait in 1974. Their introduction was not without violence (see section 1.5.5). Following the violent incidents, ownership of slerek fleets resumed in Muncar: first using the once despised cooperative system sponsored by the government that lasted until the early 1980s (Respondent SB35, personal communication, 19 August 2004), then replaced by the usual patron-client system which resulted in an exponential rate of increase in fleet numbers. Merta (1995) reported that slerek fleets started with 10 units in 1974, and increased to 44 units in 1975, to 119 units in 1977 and to 295 units in 1990 (of which only 273 units had licenses). By 2002, the number decreased with about 150 units of active pairs of slerek from both sides of the strait (Dewa Gede Raka Wiadnya 27, personal communication, 14 June 2002). Meanwhile, in 2004 the number continued to decline. It was noted that the number of active pairs had declined to only 113 units in 2004 (Respondent SB35, personal communication, 19 August 2004; and Respondent SB53, personal communication, 12 September 2004), despite the fact that there were 273 licenses issued. The main perceived reason for this decline is because fishing was no longer profitable for some fishers. In terms of boat size, boats used for lemuru fishery have also undergone an increase in size over the past decades. During the 1930s through the 1940s, the dug-out boats had a tonnage of 0.2 to 0.3 GT (Respondent SB23, personal communication, 10 August 2004). In the 1950s through the 1960s, boat size increased to 0.4 to 1.7 GT (reestimated 28 using data in Soemarto, 1960). In the 1970s, they were between 3.5 to 4 GT (re-estimated28 using data provided by Respondent SB73, personal interview, 11 October 2004). By the end of the 1980s, boat size increased to 15 GT (re-estimated28 using data in  27  Professor at the Faculty of Fisheries, Brawijaya University, Malang, East Java.  28  Gross tonnage formula used is based on the Decree of the Minister of Marine Affairs and Fisheries KEPMEN No.10/2003: (Length x Breadth x Depth x Coefficient of Block)/2.83 Coefficient of block (Cb) for slerek boat is 0.56 (Iskandar & Pujiati, 1995). In naval architecture, Cb indicates a ratio between boat displacement volume and the multiplication product of its dimension (Length x Breadth x Depth).  32  Merta, 1995), and jumped to 28.7 – 36.8 GT in the 1990s (re-estimated28 using data in Wudianto, 2001). During the fieldwork in 2004, on average slerek boats were in the order of 25 – 30 GT (Respondent SB56, personal communication, 20 September 2004), although in some cases, I also found slerek boats that were larger, as much as 40 GT (Respondent SB2, personal communication, 3 August 2004) and 52 GT (re-estimated28 using data provided by Respondent SB44 (personal communication, 24 August 2004) 29. Interestingly, a few fishers that realized the decline of lemuru stocks, like Respondent SB35 (personal communication, 19 August 2004), reduced their boat size from 30 GT to 15 GT. The mesh size of the slerek purse seine nets has always been ¾”, which experts considered as detrimental 30 to the lemuru stocks as it catches the juvenile sardines (sempenit, approx. 3 to 10 cm TL) and the sub-adult sardines (protolan, approx. 10 to 13 cm TL) as well (Wudianto, 2001). Local governments of both East Java and Bali provinces had implemented regulations to limit the use of purse seine net mesh size to minimum 1” to allow both sempenit and protolan to escape, but this regulation has proven to be ineffective 31 and its enforcement is lacking (Dewa Gede Raka Wiadnya27,  29  Unfortunately, progression of fishing boat size is not monitored in Indonesia. Often the case, a fishing boat with a size greater than 30GT would still carry a medium-scale license, which is the original license of the boat when it was first issued many years ago when the boat was still under 30GT. This is a practise that can be considered a form of IUU fishing amongst domestic fishing fleet and is discussed elsewhere (Buchary, Pitcher, & Willoughby, 2008).  30  At ¾” mesh size, protolan would be caught but sempenit would be gilled and stuck in the net; therefore, fishers usually would avoid sempenit schools during the net setting, as catching sempenit would pose risks to the nets and to the crew who would manually haul them (Respondent SB53, personal communication, 12 September 2004; Respondent SB56, personal communication, 20 September 2004). Nevertheless, sempenit cannot be totally avoided as their existence often overlaps with protolan’s. Once caught, about 40% of sempenit would be damaged as they are gilled in the net and became mushy, therefore, they are discarded (Respondent SB55, personal communication, 17 September 2004).  31  This is because fishers found that by using ¾” mesh size net, fishing can be carried out throughout the year by avoiding sempenit, but catching protolan, adult sardines (lemuru, approx. 13 to 18 cm TL), and senescent sardines (lemuru kucing, approx. 18 to 25 cm TL) and therefore, guaranteeing income continuity (Respondent SB4, personal communication, 4 August 2004). Nobody is willing to forego this income continuity opportunity (Respondent SB4, personal communication, 4 August 2004), as most of them is poor and have short term view in financial planning (Respondent SB35, personal communication, 19 August 2004).  33  personal communication, 14 June 2002; Respondent SB35, personal communication, 19 August 2004). Except for the use of engines, the slerek fishery has always been a fishery whose nets are operated entirely in manual mode since its introduction in 1974. The slerek fishery in the strait is the only fleet that resembles a commercial enterprise in terms of the sheer abundance of the fish caught and of how the government licensing system operates. However, the slerek fleets are managed like a small-scale fishery entity where patronclient relationships and kinship are very strong and these serve as the social safety net for the fishing community in question. A simple kaleidoscope on the evolution of lemuru fishery in the Bali Strait is presented in Appendix C2. 1.4.5 The Culture of Conflict Sown by Resource Appropriation Problems Despite the existence of a lemuru fishery management body that is supposed to manage and control the appropriation of lemuru stocks in the strait (FAO-Norway Government Cooperative Programme, 2001), the exploitation of lemuru continues in an open access, mare liberum fashion. Anybody that could afford to buy the means to fish can access the resources any time they want. Since the beginning, when slerek was introduced in 1974, the exploitation of lemuru in the Bali Strait follows a style typical of common-pool resource, resulting in a random outbreak of inter-gear and inter-ethnic violent incidents (e.g., Emmerson, 1987). In the Bali Strait, purse seine nylon nets were first acquired from Japan by some Muncarbased fish plants and canneries at the beginning of 1974 and were made into pair-boat style slereks (Respondent SB77, personal communication, 17 October 2004). At the time, there were only 14 units of privately-owned slerek fleets (Respondent SB77, personal communication, 17 October 2004). Most of them were owned by Chinese-descent Indonesians who also owned the fish plants and canneries (Emmerson, 1987). The first half of 1974 showed a marked increase of lemuru catches landed by these slerek boats 34  which created social tension with traditional fishers (mostly payang) who were still using nets made of cotton. By mid-1974, following a petition by the traditional fishers declaring that the new gear destroyed lemuru stocks, the operation of these privately-owned slereks was banned by the local government in Muncar (Emmerson, 1987; Respondent SB77, personal communication, 17 October 2004). Ironically, responding to the pleas of the same traditional fishers to catch more fish with better techniques, the local government assisted by the Marine Fisheries Research Institute in Jakarta (Wudianto, 2001) - made a plan to develop the lemuru fisheries by empowering traditional fishers with an aid package that comprised similar gear (Emmerson, 1987). The new gear would also be managed through a new government-sponsored fisheries cooperative system called Koperasi Unit Desa “Mino Blambangan” 32 (Emmerson, 1987; Respondent SB77, personal communication, 17 October 2004; Respondent SB73, personal interview, 11 October 2004). Unfortunately, the government only had money to provide 7 units of this new slerek gear (each with 12 crew members); while thousands of other expectant fishers were not selected to receive the aid package (Emmerson, 1987). The manner in which the aid packages were distributed was considered unfair by many fishers (Respondent SB73, personal interview, 11 October 2004). The new cooperative system did not work. One thing very obvious was the fact that from the beginning, the cooperative system tried to undermine the strong kinship and patron-client relationship in the area, which the government considered as exploitative to the crew members and skippers (Kasijanto, 1976; Emmerson, 1987). Perhaps it was a combination of these two factors: (1) perceived injustice of fishing gear aid package distribution, and (2) government attempt to undermine the wellestablished patron-client relationship; that triggered a violent incident in the history of the 32  See details on the history of fisheries cooperatives development in Muncar in Chapter 2, section 2.2.  35  Bali Strait known as Malamun - an acronym of Malapetaka Muncar, or Muncar’s Calamity - on September 30, 1974 (Respondent SB75, personal communication, 13 October 2004). During Malamun, the new government slerek fleets were set afire by traditional fishers in Muncar (Kompas, 1974; Emmerson, 1987; Respondent SB77, personal communication, 17 October 2004; Respondent SB73, personal interview, 11 October 2004). Thousands of disappointed fishing folk (mostly Madurese payang fishers) ran amok in the fishing port and surrounding areas, created chaos and destruction for the whole day against Chinese-owned shops and properties 33, and even targeting bagan (fixed lift-net) bamboo structures owned by the Buginese fishers; which prompted the police and military personnel to be deployed to take control of the situation (Kompas, 1974; Emmerson, 1987; Respondent SB77, personal communication, 17 October 2004; Respondent SB73, personal interview, 11 October 2004; Respondent SB55, personal communication, 17 September 2004). Apart from a major loss due to the destructions (about Rp.99.5 million, or about USD 240,000 in 1974 value), luckily nobody was reported killed (Kompas, 1974). As the new government cooperative system did not work (although it is still exist until now but assumes very minimal role), the patron-client relationship resumed, and as a result the slerek fleet increased exponentially. Fishing activities in the strait carried on without any major incident until the 1990s, when migrant Tubanan boats started to come in the strait (mainly to Muncar district), targeting the same lemuru species. At the beginning when there were only 13 units in 1992, nobody really complained and everybody was quite happy fishing together (Respondent SB73, personal interview, 11 October 2004). However, resource appropriation conflicts started to emerge with the local slerek fishers when a total of 130 33  Latent dislike against the Indonesians of Chinese descent is a racial prejudice that has deep historical roots, which stemmed from Dutch colonial policies and practices centuries ago (Indonesian Legal Studies Foundation, 2006; Utomo, 2000). The discrimination continued throughout the Soekarno’s (1945–1967) administration and was institutionalized further during Soeharto’s (1967–1998) regime (Utomo, 2000). Fortunately, following the fall of Soeharto in 1998, effort to turn the tide of the prejudice was pioneered during the presidency of B.J. Habibie (1998–1999) and K.H. Abdurrahman ‘Gus Dur’ Wahid (1999–2001) and maintained by the subsequent presidents by acknowledging the cultural heritage of the Indonesian Chinese minorities and repealing some of the most notorious laws that undermine them (Indonesian Legal Studies Foundation, 2006). Nevertheless, the prejudice still has not completely disappeared and somehow persists in the sub-consciousness of the nation. It continues to haunt the nation each time there is unrest.  36  units were noted in 2004 (Respondent SB53, personal communication, 12 September 2004). Moreover, in later years, they have tended to stay much longer and with bigger boats, i.e., 35 - 40 GT in 2004 (Respondent SB67, personal communication, 30 September 2004). The main complaint brought about by the slerek fishers, interestingly was not about the conflict over the fish per se. It was rather, the difference of fishing methods used that created tension, and therefore, conflicts (Respondent SB67, personal communication, 30 September 2004). For example, slerek must operate in pitch darkness, during the new moon phase, as the skipper relies on his own vision (while perching on the high bench) to visually spot the lemuru schools. On the other hand, the Tubanan boats use high intensity halogen light to attract lemuru schools. It is the bright light used by the Tubanan boats that is blamed by the slerek skippers to have caused the glare that affect their vision (Respondent SB67, personal communication, 30 September 2004; Respondent SB70, personal communication, 5 October 2004). The tension between slerek and Tubanan fleets culminated in late 2006 when some crew members of a slerek fleet set a Tubanan boat at sea afire while fishing (Respondent SB76, personal communication, 16 December 2006). The incident was quickly resolved when all juragan darat (boat owners) of slerek fleet in Muncar, in a spirit of camaraderie, pitched in money to pay for the compensation to the owner of the victimized Tubanan boat (Respondent SB53, personal communication, 16 December 2006). It seems that by 2008 the flow of migrant Tubanan boats has been stemmed and those boats that were in Muncar were deported back to their original hometowns on the northern coast of East Java (Media Partisipasi, 2008); only time will tell whether the fishing community in the Bali Strait could turn the tide of the culture of conflict that have haunted them for decades.  37  References Acciaioli, G. (2005). Legacy of conflict: Bugis IDPs have travelled as warriors, farmers, fishers and traders. Inside Indonesia 82 (April - June 2005). Retrieved 10 April, 2008, from http://insideindonesia.org/content/view/166/29/ Ainsworth, C. H. (2004). How we carried out 'Back-to-the-Future' Community Interviews. In T. J. Pitcher (Ed.), Back to the Future: Advances in Methodology for Modelling and Evaluating Past Ecosystems as Future Policy Goals (Vol. 12 (1), pp. 117 - 125). Vancouver, Canada: Fisheries Centre, University of British Columbia. Babbie, E. (1989). The practice of social research (5th ed.). Belmonth, CA: Wadsworth. Bailey, C. (1988). The Political Economy of Marine Fisheries Development in Indonesia. Indonesia, 46, 25-38. Bailey, C., & Jentoft, S. (1990). Hard choices in fisheries development. Marine Policy, 14(4), 333 - 344. Buchary, E. A., Pitcher, T. J., & Willoughby, N. (2008). Indonesia's non-recorded fish problem revisited. In D. Kalikoski & T. J. Pitcher (Eds.), Assessing Illegal, Unreported and Unregulated Fishery Catches (IUU): Some case studies (Vol. 16 (5)). Vancouver, Canada: Fisheries Centre, University of British Columbia (in press). Buchary, E. A., Wahyuningsih, D. S., & Pauly, D. (2003). Documenting the common names of Indonesian fishes. FishBytes, The Newsletter of the Fisheries Centre at UBC, 9(5), 6-7. Burhanuddin, & Praseno, D. P. (1982). Lingkungan perairan Selat Bali [The marine environment of the Bali Strait], Proceedings of Lemuru Fisheries Seminar (pp. 27 - 36): Puslitbangkan, Badan Balitbangtan, Deptan (Jakarta). Cheung, W. L., & Sadovy, Y. (2004). Retrospective evaluation of data-limited fisheries: a case from Hong Kong. Reviews in Fish Biology and Fisheries, 14, 181 - 206. Christensen, V., & Pauly, D. (1993). Trophic models of aquatic ecosystems, ICLARM Conference Proceedings 26 (Vol. 26, pp. 390). Manila, Philippines, and Copenhagen, Denmark: International Center for Living Aquatic Resources Management (ICLARM), Inetarnational Council for the Exploration of the Sea (ICES) and Danish International Development Agency (DANIDA). Chuenpagdee, R., Liguori, L., Palomares, M. L. D., & Pauly, D. (2006). Bottom-Up, Global Estimates of Small-Scale Marine Fisheries Catches, Fisheries Centre Research Reports (Vol. 14, pp. 105). Vancouver, Canada: Fisheries Centre, University of British Columbia. Climate Assessment for the Southwest (CLIMAS). (2003). Glossary of Climate Terms. Retrieved 19 July, 2009, from http://www.climas.arizona.edu/forecasts/glossary.html Collier, J., Jr., & Collier, M. (1986). Visual Anthropology: Photography as a Research Method (Revised & enlarged ed.): University of New Mexico Press.  38  Creswell, J. W. (2003). Research Design: Qualitative, Quantitative, and Mixed Methods Approaches (2nd ed.). Thousand Oaks, CA, USA: Sage Publications, Inc. Encyclopedia Britannica Inc. (2008a). Etymology. Retrieved November 25, 2009, from Dictionary.com website: http://dictionary.reference.com/browse/etymology Encyclopedia Britannica Inc. (2008b). Toponymy. Retrieved November 25, 2009, from Dictionary.com website: http://dictionary.reference.com/browse/toponymy Emmerson, D. K. (1980). Rethinking artisanal fisheries development: western concepts, Asian experiences. Staff Working Paper No. 423 (October 1980). Washington, D.C., USA: The World Bank Group. Emmerson, D. K. (1987). Orders of meaning: understanding political change in a fishing community in Indonesia. In B. Anderson & A. Kahin (Eds.), Interpreting Indonesian Politics: Thirteen Contributions to the Debate. Interim Report Series (Publication No. 62) (pp. 149 - 164). Ithaca, New York, USA: Cornell Modern Indonesia Project, Southeast Asia Program, Cornell University. Fakultas Perikanan UNIBRAW, Dinas Kelautan dan Perikanan Propinsi Jawa Timur, & Dinas Kelautan dan Perikanan Propinsi Bali. (2007). DataBase Perikanan Version 1.0. [Fisheries Database Version 1.0] (Version 1.0). Malang, East Java, Indonesia. FAO-Norway Government Cooperative Programme. (2001). FISHCODE Management: Report on a Workshop to Refine the Draft Management Plan for the Bali Strait Sardine (Lemuru) Fishery. GCP/INT/648/NOR. Field Report F-18 (En) (pp. 33). Rome, Italy: Food and Agricultural Organisation of the United Nations (FAO). Ghofar, A., & Mathews, C. P. (1996). Box. 2 The Bali Strait lemuru fishery. In D. Pauly & P. Martosubroto (Eds.), Baseline studies of biodiversity: the fish resources of Western Indonesia. ICLARM Stud. Rev. 23 (pp. 146). Manila, Philippines: ICLARM. Ghofar, A., Mathews, C. P., Merta, I. G. S., & Salim, S. (2000). Incorporating the Southern Oscillation Indices to the Management Model of the Bali Strait Sardinella Fishery. In FAO/NGCP (Ed.), FISHCODE Management: Papers Presented at the Workshop on the Fishery and Management of Bali Sardinella (Sardinella lemuru) in Bali Strait. GCP/INT/648/NOR Field Report F-3-Suppl. (En) (pp. 43 - 52). Rome, Italy: Food and Agriculture Organization of the United Nations (FAO). Hendiarti, N., Suwarso, Aldrian, E., Amri, K., Andiastuti, R., Sachoemar, S. I., et al. (2005). Seasonal variation of pelagic fish catch around Java. Oceanography, 18(4), 113-123. Indonesian Legal Studies Foundation. (2006). Discrimination against Ethnic Chinese in Indonesia. Asia-Pacific News, 43, 2 - 3. Iskandar, B. H., & Pujiati, S. (1995). Keragaan Teknis Kapal Perikanan di Beberapa Wilayah Indonesia [Technical performance of fishing boats in several areas in Indonesia] (pp. 51). Bogor, Indonesia: Jurusan Pemanfaatan Sumberdaya Perikanan. Fakultas Perikanan dan Ilmu Kelautan, Institut Pertanian Bogor, Bogor, Indonesia [Department of Fisheries Resource Utilization. Faculty of Fisheries and Marine Science, Bogor Agricultural University].  39  Johannes, R. E. (1993). Integrating traditional ecological knowledge and management with environmental assessment. In J. T. Inglis (Ed.), Traditional knowledge: concepts and cases (pp. 33 - 40). Ottawa, Canada: International Development Research Centre. Kasijanto, M. J. (1976, 16 July). Ke Muncar sekarang [To Muncar now]. Kompas Daily News. Kompas. (1974, 3 October). Kerusuhan di Muncar Rugikan Rp.99,5 Juta [Violence in Muncar costs Rp.99.5 million]. Kompas Daily News. Marshall, C., & Rossman, G. B. (1999). Designing Qualitative Research. Thousand Oaks, CA, USA: Sage Publications, Inc. Media Partisipasi. (2008, 23 June). Bila Selat Bali Menunggu Kepunahan [When the Bali Strait waits for its demise]. Retrieved 26 July, 2009, from http://icdhre.or.id/index2.php?option=com_content&do_pdf=1&id=45. Merta, I. G. S. (1995). Review of the Lemuru fishery in the Bali Strait. In M. Potier & S. Nurhakim (Eds.), BIODYNEX: Biology, Dynamics, Exploitation of the Small Pelagic Fishes in the Java Sea (pp. 97 - 106). Jakarta, Indonesia. Merta, I. G. S., Widana, K., Yunizal, & Basuki, R. (2000). Status of the Lemuru Fishery in Bali Strait: its Development and Prospects. In FAO/NGCP (Ed.), FISHCODE Management: Papers Presented at the Workshop on the Fishery and Management of Bali Sardinella (Sardinella lemuru) in Bali Strait. GCP/INT/648/NOR Field Report F-3-Suppl. (En) (pp. 1 - 42). Rome, Italy: Food and Agriculture Organization of the United Nations (FAO). Morgan, D. L., & Krueger, R. A. (1998). The Focus Group Kit. Thousand Oaks, USA: Sage Publications, Inc. Neis, B., Schneider, D. C., Felt, L., Haedrich, R. L., Fischer, J., & Hutchings, J. A. (1999). Fisheries assessment: what can be learned from interviewing resource users? Canadian Journal of Fisheries and Aquatic Sciences, 56, 1949 - 1963. Nontji, A., & Ilahude, A. G. (1975). Ekologi fitoplankton Selat Bali [The ecology of phytoplankton in the Bali Strait]. Oseanologi di Indonesia, 5, 25 - 41. Pauly, D. (1997). Small·Scale Fisheries in the Tropics: Marginality, Marginalization and Some Implications for Fisheries Management. In E. K. Pikitch, D. O. Huppert & M. P. Sissenwine (Eds.), Global Trends: Fisheries Management, American Fisheries Society Symposium (Vol. 20, pp. 40-49). Bethesda, Maryland, USA: American Fisheries Society. Priyono, B. E., & Sumiono, B. (1997). The marine fisheries of Indonesia, with emphasis on the coastal demersal stocks of the Sunda shelf. In G. Silvestre & D. Pauly (Eds.), Status and management of tropical coastal fisheries in Asia. ICLARM Conf. Proc. 53 (pp. 38-46). Mandaluyong City, Philippines, and Makati City, Philippines: Asian Development Bank, and International Center for Living Aquatic Resources Management. Ritterbush, S. W. (1974). Certain aspects of the population biology of the Bali Strait Lemuru fishery. Unpublished M.Sc. Thesis, University of Hawaii, Honolulu, USA.  40  Saliyo, B. (1973). Keadaan oseanografi daerah-daerah penangkapan ikan lemuru di Selat Bali [Oceanographic condition of lemuru fishing grounds in the Bali Strait] (pp. 1 - 16). Jakarta, Indonesia: Lembaga Penelitian Perikanan Laut [Marine Fisheries Research Council]. Smith, I. R. (1979). A Research Framework For Traditional Fisheries, ICLARM Studies and Reviews, No. 2. Manila, Philippines: International Center for Living Aquatic Resources Management (ICLARM). Soemarto. (1960). Craft and gear utilized in the sardine fishery at Muntjar, Indonesia. In H. Rosa, Jr. & G. Murphy (Eds.), Proceedings of the World Scientific Meeting on the Biology of Sardines and Related Species, Vol. II: Species Synopses and Subject Synopses (pp. 1247 1264). Rome, Italy: Food and Agriculture Organization of the United Nations (FAO). Stacey, N. (2007). Boats to burn: Bajo fishing activity in the Australian Fishing Zone (Vol. 2). Canberra, Australia: Australian National University (ANU) E Press. Utomo, D. (2000, 24 May). INDONESIA: Discrimination against ethnic Chinese. International News, Green Left Weekly #406 Retrieved 26 July 2009, from http://www.greenleft.org.au/2000/406/23611 Wudianto. (2001). Analisis sebaran dan kelimpahan ikan lemuru (Sardinella lemuru, Bleeker 1853) di perairan Selat Bali: kaitannya dengan optimasi penangkapan. [The analysis of distribution and abundance of lemuru (Sardinella lemuru, Bleeker 1853) in the Bali Strait: its relation with catch optimization]. Unpublished Ph.D. Dissertation, Bogor Agricultural University, Bogor, Indonesia. Wyrtki, K. (1961). Physical oceanography of Southeast Asian waters, Naga Report (Vol. 2, pp. 195). San Diego, CA: Scripps Institute of Oceanography.  41  2  Beyond Fisheries Catch Statistical Data for the Bali Strait: the Story Behind the Numbers 34  2.1 Introduction An accurate evaluation of the impact of fisheries is impossible to derive if one were to work on inaccurate catch data. This not only leads to poor fisheries management policies, but also to unwise investment decisions by companies in the fisheries sector and risky credit provisions to these companies by banks. This is particularly true for many developing world fisheries, including those in Indonesia, where there are still many challenges in the capacity, efficiency and effectiveness of national statistical systems. On top of technical barriers, the issue of statistical data quality is made worse in Indonesia because working in the statistics division of a government office or organization is still considered as a bad career choice; being transferred to a statistics division is considered as a punishment or demotion, and few government employees at the non-decision making level know the benefit and purpose of statistical data collection and  maintenance  (Nurhakim23,  personal  communication;  Soecipto 35,  personal  communication, 28 September 2004). To begin the evaluation of the impact of fisheries in the Bali Strait, one needs to look at the state of the data used in the management framework. Therefore, in this Chapter 2, I explore what lies behind the numbers within fisheries statistical data sets. The case study focuses on medium-scale 36 domestic purse seine fishing fleets in the Bali Strait, locally known as slerek (see Figure 1.7 in Chapter 1). Slerek mainly targets sardine (Sardinella 34  A version of this chapter will be submitted for publication. Buchary, E.A. and Pitcher, T.J. Beyond Fisheries Catch Statistical Data for the Bali Strait: the Story Behind the Numbers.  35  Former Chief at the Regency Fisheries Office of Banyuwangi Regency, East Java, Indonesia.  36  In Indonesia’s fisheries system (Priyono & Sumiono, 1997), medium-scale fishing fleets include privately-owned (as opposed to company-owned) boats, of generally less than 5 gross tonnes (GT) and up to 30 GT, that mostly use inboard engines. Medium-scale fishing boats tend to have access to few or no shore-based amenities such as ice plants, cold storage facilities or workshops. A fisher usually owns one or, at the most, a few fishing units.  42  lemuru), commonly known as lemuru. Using information acquired from comparative analyses of reported official catch records, I compared nominal catch data of S. lemuru from the Bali Strait from 1972 to 2001 across all administrative jurisdictions to assess its reporting consistency, and from 1950 to 1971 to reconstruct the historical catches of lemuru prior to formal government data collection. Direct observations in the field and information gained from individual semi-structured interviews and focus group discussions (FGDs) were used to systematically trace the flow of how the data were collected, enumerated and reported in the system. Synthesis from these three approaches was used to: (1) understand how the fisheries data collection and reporting systems work, (2) point out where and why data gaps occurred, and (3) provide some suggestions for improvements.  2.2 The Slerek Fishery: Background on its Development and Statistics System The lemuru fishery had existed in the Bali Strait even before Japanese troops came to the area in 1942 (Respondent SB23, personal communication, 10 August 2004). Respondent SB23 recalled that during 1930s – 1940s in Negara district, he and his fellow fishers used scoop (serok) and cast nets (jala buang) to catch lemuru, with sail powered dug-out boats (jukung) (personal communication, 10 August 2004). Interviews with two very old (80+ years old) respondents (Respondent SB19, personal communication, 8 August 2004; Respondent SB23, personal communication, 10 August 2004) indicated that there were no other significant fisheries in those decades except the lemuru fishery and that the fishery was mainly for subsistence as market access was very hard. During the National Revolution period (1945 – 1949), fishing in the strait was temporarily halted by Dutch troops by means of force and was resumed in 1950 (Respondent SB23, personal communication, 10 August 2004). After the revolution, in the 1950s, the lemuru fishery in the strait grew slowly. Two types of gear emerged (Soemarto, 1960): (1) jala oras or payang oras (seine nets made of cotton and leaves of Gebang palm, Corypha utan), and (2) jaring lemuru or jaring eder  43  (cotton gill nets); both of these gear were operated using wooden sail powered boats (perahu or sampan), and both targetted lemuru. During 1950s and 1960s, fixed lift nets using a bamboo platform (bagan tancap), targetting squid and small fish (including juvenile and sub-adult lemuru), came to Muncar district brought in en masse (see details in Appendix A, in Chapter 1) by Bugis fishers and their families from South Sulawesi (Respondent SB52, personal communication, 12 September 2004). In 1968, another gear, a bamboo tidal weir (banjang kerei), emerged in Muncar district and targeted small fish, mainly juvenile and sub-adult lemuru (Respondent SB69, personal communication, 2 October 2004). Both of these gear have been operated on a small scale in Pangpang Bay, calm and enclosed waters in Muncar district, ever since their arrival, and never made it to Negara district. A new gear to catch lemuru using a purse seine net and a pair of boats was first introduced in Muncar in 1974 (Respondent SB73, personal interview, 11 October 2004), and was later introduced in Negara district in 1975 (Respondent SB13, personal communication, 6 August 2004). At that time, the size of the boats was merely 3.5 to 4 GT (re-estimated using data provided by Respondent SB73, personal interview, 11 October 2004). The new gear was called ‘slerek’ and its introduction was not without intense violence (see details in Chapter 1). Slerek fleets continued to grow in number and in size. Except for the use of engines, the slerek fishery has always been operated entirely in manual mode, i.e., the purse nets are set and pulled by hand. In 2004, on average slerek boats were in the order of 25 – 30 GT (Respondent SB56, personal communication, 20 September 2004). However, by 2004 some fishers, realizing the decline of lemuru stocks, reduced their boat size from 30 GT to 15 GT (Respondent SB35, personal communication, 19 August 2004). Propelled by the significant fuel price increase beginning in 2005 (PT. Pertamina (Persero), 2008), many fishers reduced their fishing frequency and some even sold their boats altogether (Respondent SB56, personal communication, 20 June 2008).  44  By 2004, along with slerek there are sixteen other types of fishing gear operating in the Bali strait and two types of fishing gear mooring in the strait but operating elsewhere (see Table 1.1, Chapter 1). Despite this array of distinct fishing gear targeting different species, in terms of bulk catch landed, S. lemuru is the dominant species with more than 90% of total catch and mainly caught by slerek (Hendiarti et al., 2005). According to Merta, Widana, Yunizal, & Basuki (2000), from 1976 – 1994, slerek landing comprised 80 - 99% of total catch landed in Muncar. Other species caught by slerek include round scads (Decapterus spp., layang), eastern little tuna (Euthynnus affinis, tongkol), and mackerel (Scomberomorus spp., tenggiri) that are caught either as by-catch or targeted in the off-lemuru season in much lower numbers than lemuru. Bi-annual monsoons (Wyrtki, 1961, cited in Ritterbush, 1974) and El Niño events that occur every 2 to 10 years (Ghofar, Mathews, Merta, & Salim, 2000) influence the oceanographic conditions of the strait, and therefore, seasonality governs the lemuru fishery. The southeast monsoon (June – October) generates an upwelling in the strait and is considered as ‘the fish season’ glut by the fishers. The glut peaks around September to October, and occasionally it would continue to the early part of November. Off season occurs during the Northwest monsoon (November – March) and the inter-monsoonal period (April and May). The fishery also has a boom-and-bust nature as expected for small pelagic planktivorous fish whose recruitment is driven by El Niño events. Commercial lemuru fishing in the Bali Strait is restricted to fishers based in Muncar district, specifically in Kedungrejo village (Banyuwangi regency, East Java province) and in Negara district, specifically in Pengambengan village (Jembrana regency, Bali province) (Figure 1.2 in Chapter 1). Therefore, processing facilities such as fish meal plants and canneries are also concentrated in these two districts. Despite its resemblance to a commercial enterprise in terms of the sheer abundance of the fish caught and of how the government licensing system operates, the slerek fleets are managed like a smallscale fishery entity where patron-client relationships and kinship are very strong (Respondent SB4, personal communication, 3 August 2004; Emmerson, 1976, 1987; Budhisantoso, Setiawati, Subagyo, & Suhardi, 1991).  45  The first account of national fisheries statistics in Indonesia was from 1973, when Indonesia had its first national agricultural census (Yamamoto, 1980). Using this census data as a reference, a standardized national marine fisheries statistical system was established in 1976 with the help of Dr Tadashi Yamamoto, a fishery statistician assigned by FAO and UNDP from 1973 to 1978 to overhaul and implement a proper national fisheries statistics system for Indonesia (Yamamoto, 1980). Results of this major project were codified and documented for the nation-wide use (Directorate General of Fisheries, 1975a, 1975b, 1980b, 1990b). Although the official national standardized system was officially launched in 1976, the project managed to work on the data collected from prior years (notably going back to 1973 37) to be included as part of the new system (Yamamoto, 1980). Prior to the establishment of this national fisheries statistical system, each region had their own way of enumerating and recording fisheries statistics, and therefore the earlier data sets cannot easily be used to make nation-wide comparison (Wahyudi, 1996). In Muncar district, for example, the amount of lemuru landed was first tallied and recorded in late 1920s, after a local marine fisheries cooperative, Koperasi Perikanan Laut “Minak Djinggo”, was established in 1927 (Respondent SB75, 1997a). In 1949, Minak Djinggo cooperative evolved into another local marine fisheries cooperative, Koperasi Perikanan Laut “Pantai Bahagia” (Respondent SB75, 1997a). Nevertheless, these early records (1927 – 1949) of lemuru catch and other fisheries information could not be found anywhere. It is possible that these data were destroyed during the Japanese occupation in Indonesia during the Second World War (1942 – 1945) and during the National Revolution period (1945 – 1949) (Respondent SB75, personal communication, 13 October 2004). Once fishing activities in the strait resumed in 1950 (after the end of National Revolution), the catch data continued to be tallied and recorded by Pantai Bahagia cooperative. In 1958, this cooperative evolved into another local cooperative, Koperasi Simpan Pinjam “Nelayan Muncar”; and in 1965, it evolved again and was split into two 37  For the Bali Strait, it went back to 1972 (Directorate General of Fisheries, 1974).  46  local cooperatives: Koperasi Perikanan Laut “Sumber Mulyo”, and Koperasi Perikanan Laut “Sumber Rezeki” (Respondent SB75, 1997a). While the standardized statistical system was in the process of being established, a government-sponsored marine fisheries cooperative, Koperasi Unit Desa “Mino Blambangan” was founded in 1975 in Muncar, as an amalgamation of the two local cooperatives previously founded: Sumber Mulyo and Sumber Rezeki cooperatives. As noted by Wahyudi (1996), the catch data collected and tallied by these local cooperatives were considered as non-standard, and as such they were not included in any official government reports and documents. For the Bali Strait lemuru catch records, official catch data started in 1972 (Directorate General of Fisheries, 1974). Fortunately, Ritterbush (1974) was able to get hold of the 1950 – 1971 data set 38 for his masters thesis when he visited Muncar from October 1973 – March 1974; and I used it in my analysis depicted in Figure 2.3 (as data source #8). No information on the account of lemuru statistics enumeration from Negara district prior to 1973 was found. Nevertheless, historical accounts from various interviews suggest that lemuru fishing in Negara was much smaller in scale compared to that in Muncar and it has always been the case.  2.3 Some Challenges to Fisheries Catch Statistics In managing the fishery resources using stock assessment tools, fishery managers rely heavily on the catch data as a basis for input to design management options. Like any stock inventory system, fishery stock assessment needs an informative and transparent ‘fish stock balance’. This balance takes into account how much fish is extracted from the ocean in total in relation to how much is actually produced through growth, immigration and recruitment. At equilibrium, as derived from Russell’s equation (E. S. Russell, 1931):  38  Ritterbush’s (1974) masters thesis, in fact, is the first and only account that published this 1950 – 1971 data set of lemuru landed at the official landing site at the village, and tallied by the now-defunct local marine fisheries cooperatives in Muncar. In his thesis, he also presented the lemuru catch of 1972-73.  47  Total fish stock extraction from the ocean ≈ Total fish stock production in the ocean Where: Total fish stock extraction  = total fishery catch = nominal landings + discards (+ take-home + other catch diversion)  And: Total fish stock production  = growth + immigration + reproductive output – mortality – predation  In the past, often accepted at face value, the accuracy of ‘total fishery catch’ statistics used as data input needed for the ‘total stock extraction’ side of the balance was rarely challenged. But in the past ten years this has changed with the increased awareness of Illegal, Unreported and Unregulated (IUU) fishing. Pitcher, Watson, Forrest, Valtýsson, & Guénette published the first paper that provides an in-depth review on IUU fishing and systematic quantification of the missing catch (2002). In their paper, in light of IUU fishing, they described that fisheries catches may be separated into three components: (1) nominal catch reported to a monitoring agency, generally to a national body that itself reports to the Food and Agriculture Organization of the United Nations (FAO), (2) reported discards, usually as a result of an observer program, and (3) unreported catch that comprised: (a) unreported discards, (b) unmandated or unregulated catch, and (c) illegal catch, which also includes misreported and under-reported catch.  48  IUU fishing has been a subject of global concern since the 1990s, when deep concerns over the development of distant water fleets 39 that create wide-scale overexploitation emerged (Richards, 2003; MRAG, 2005). Since then, there have been a number of initiatives on combating IUU fishing; in particular, the negotiations of the FAO International Plan of Action on IUU fishing (FAO, 2001). Up until then, ‘nominal landings’ were usually considered as total fisheries catch statistics and termed ‘fisheries production statistics’ in many countries. It is this production statistics that is reported to a monitoring agency, generally to a national body that itself reports to FAO (Pitcher et al., 2002). Although discards for certain gear (e.g., trawls) have been acknowledged for quite some time and have been monitored in select fisheries in the developed world, the reported nominal landings can still be erroneous (Agnew et al., 2009). With the advent of evaluating fishing impacts at an ecosystem scale (e.g., Christie, Fluharty, White, Eisma-Osorio, & Jatulan, 2007; Christensen, Aiken, & Villanueva, 2007; Andrew et al., 2007; Garcia & Cochrane, 2005), the meaning of true fisheries catch has never been more important. This is especially true since the expected benefits from IUU fishing usually far exceed the cost of being apprehended (Sumaila, Alder, & Keith, 2006). The failure to understand the full extent of fishing can result in not only ‘blind’ fisheries management that jeopardizes ecosystem structure, but also compromises fisheries benefits (Evans, 2000). In general, when people discuss IUU fishing, topics seem to revolve around largescale fisheries and/or foreign fishing fleets. In practice, IUU fishing happens in many fisheries (FAO, 2002), including domestic fishing fleets, artisanal, small-scale and medium-scale fisheries. In the developing world, such as in Indonesia, the consequences can be more serious. The absence of proper management facilities in developing world fisheries (e.g., fishers’ cooperatives that work for the benefit of the fishers, observer  39  It started with the development of the distant water fishing fleets of the Soviet Union in the 1950s, followed by the development of similar fleets by Japan, other Far Eastern states, European states and the USA in the 1970s .  49  programs, official fish log-book systems, and monitored centralized landing facilities) have exacerbated the impact of the missing quantities of the true catch. In any fisheries, quantifying the unaccounted part of the true catch can present a significant technical challenge. Regulatory agencies may hold no mandate to record the catch statistics for certain fishery/species/gear, e.g., anchovy catch by Sulawesi migrant fishers in Kabui Bay, Raja Ampat (Papua, Indonesia) that has never been officially accounted for by Raja Ampat fisheries officers (Bailey, Rotinsulu, & Sumaila, 2008). Moreover, the issue can be compounded by the politically sensitive nature of the situation, especially when the target species is an endangered species and/or if the gear used is violating existing rules. For example, humphead wrasse (Cheilinus undulatus), listed as Endangered in IUCN 40 Red List (B. Russell, 2004) and listed in Appendix II of CITES 41 as of 2005, are widely caught in Indonesia using potassium cyanide to be traded in the lucrative market of live reef food fish (LRFF) in Hong Kong (Sadovy et al., 2003). Using cyanide for fishing is illegal in Indonesia as stipulated by Fisheries Act No.9/1985 (Halim, 2002) and later replaced by Fisheries Act No.31/2004 (President of the Republic of Indonesia, 2004) and hence it is very difficult to get the actual true catch of Indonesia’s humphead wrasse. In the artisanal fisheries of developing countries, like the slerek fishery, the traditional and cultural context makes the catch data verification exercise even more difficult than it already is due to the informal, kin-ship and patron-client nature of the trade and therefore, its lack of standardized and systematic way of doing things. Therefore, in this study to get an idea about the state of the catch data enumerated, reported and used in the management framework, I implemented a mixed methods approach in the research design (sensu Creswell, 2003). This approach combines both  40  International Union for Conservation of Nature and Natural Resources.  41  Convention on International Trade in Endangered Species of Wild Flora and Fauna. Having a species listed in the Appendix II of CITES means that exports and imports of the species must be documented by government authorities as having been taken legally. In the case of Humphead Wrasse from Indonesia, the legal way to catch the fish is by using hook and line, fish trap and gill net.  50  quantitative and qualitative approaches, in both the data collection and data analysis components (see Figure 1.1 in Chapter 1).  2.4 Materials and Methods 2.4.1 Data Collection Methods A total of five-month fieldwork in the study area was carried out in 2002, 2004, 2005 and 2006. During this time, I implemented a mixed methods approach, which combines both quantitative and qualitative approaches. This was carried out by: (1) direct observation through living within the fishing communities to get local data, first-hand accounts, and to better comprehend the issues, (2) having some immersion experience, for example, by participating in fishing activities, (3) implementing individual in-depth semi-structured interviews, focus groups discussions (FGDs) and taxonomy, toponymy and etymology interviews, (4) conducting desk-top studies on various records, archives and databases, and (5) maintaining post-fieldwork personal communication with some respondents. The direct observation through living within the fishing communities had enabled me to not only have first-hand accounts on various issues in the daily life of the fishers, but also to do landing site censuses 42, opportunistic surveys of the fishing ports, fishing villages and fish markets, and take photography and video footage on relevant objects of observation. During the 5-month fieldwork, I also managed to undertake participatory fishing in a slerek fleet and in two other types of fishing gear (i.e., bagan tancap and banjang). This enabled me to get an idea of how the fishing activity was carried out, how much fish was caught in a typical fishing day, how much was landed, how and where 43 the fish were  42  Notably tallying the total numbers of government scales vis-à-vis private scales in landing sites and fishing ports.  43  In government scales versus private scales.  51  weighed, how they were enumerated and by whom, and how they were sold through various formal and informal trade channels. Ninety-two respondents (18 female and 74 male) were recruited and canvassed using a snowball sampling method (sensu Babbie, 1989, cited in Neis et al., 1999) for individual semi-structured interviews, FGDs and taxonomy, toponymy and etymology interviews. Interviews and FGDs were audio-taped and transcribed17 in the original language of the interview (viz., Indonesian). Additional methods of data collection include desk-top studies on records, archives, databases and satellite images of the study area, and maintaining post-fieldwork personal communication 44 with some respondents. Detailed descriptions on data collection methods, consenting procedure, dealing with respondents’ anonymity and postfieldwork data archiving system are given in Chapter 1 45. Data triangulation between results obtained through different sets of data collection and data analytical methods was used to confirm, cross-validate and corroborate findings, and to find convergence (see Figure 1.1 in Chapter 1). Therefore, data uncertainty in data collection and analysis, and biases in respondents’ response and memory recollection can be reduced. Data triangulation is also helpful to see pattern and propensity. 2.4.2 Analytical Methods To get a thorough understanding of the system, desk-top studies on official catch records, archives, databases and satellite images were cross-validated with groundthruthing obtained using direct observations in the field, participatory fishing and 44  45  Via cellular phone, email and text messages. The fieldwork protocols had been approved and certified by the Behavioural Research Ethics Board of the University of British Columbia (BREB-UBC). Local research permits from various administrative jurisdictions of the study area had also been acquired prior to starting the fieldwork. Copy of BREBUBC Certificate of Approval and list of local research permits are compiled in the Appendix B.  52  personal accounts from respondents (i.e., fishers, data enumerators, fish meal plant manager and labours, fisheries managers, and fisheries scientists) gained from individual semi-structured interviews, FGDs, taxonomy, toponymy and etymology interviews, and post-fieldwork personal communication. Additionally, photography and video footage analysis were also used to reflect on the information synthesized (Collier & Collier, 1986; Marshall & Rossman, 1999, p.124 - 126). Meanwhile, to cross-check whether the fish catch reporting system throughout jurisdictions and responsible agencies is consistent, I collated, compared and traced nominal catch data of S. lemuru from the Bali Strait from 1972 to 2001, across all administrative jurisdictions (i.e., village, district, regency, province, national and international). Additionally, to get an idea of the historical catch record of lemuru, I also used the village-level data set that was collected by previous local fisheries cooperatives (1950 – 1971) and was published in Ritterbush (1974). The official reported (nominal) catch data set used in the analyses mainly comprised official published government records, but it also includes material from individual scientists, researchers in government agencies and multi-agency projects, and personal journals. Time series of the median of official records and its 5th and 95th percentile were also estimated. The time series of the median of official records is used as a point of reference to estimate the true catch (see Chapter 3).  2.5 Results and Discussion 2.5.1 The Fate of the Caught Fish Once Landed During day time, dense schools of lemuru are found close to the bottom like most small clupeids, while at night they move close to the surface to feed on plankton forming scattered schools (Merta et al., 2000). Therefore, lemuru fishing with any gear in the strait is normally conducted during the night (approximately from 7pm to 5am). Ritterbush (1974, p.10) also noted that, “As the sun rises the schools migrate downward. Consequently, all fishing is carried out at night during periods of little or no moon.”.  53  This is in agreement with various narrative accounts of respondents that describe fishing only occur during the new moon phase (petengan) 46. By 6am, slerek boats arrive pair by pair at their own preferred landing sites. Based on our observations in 2004 in Negara district, only about 25% of the boats went to the government landing site, while the majority of the boats went to unofficial landing sites (which are located about 1 km from the official site). In Muncar, the unofficial landing sites are scattered along a 5.5 km beach stretch (Respondent SB75, personal communication, 13 October 2004), while the government landing site (i.e., fishing port) is located in the middle of the stretch 47. This is understandable, since there were only a handful of government scales for the easily spoiled fish that were landed in high ambient temparature 48, often in great abundance. During a census of scales on five different occasions, I found that there were only 8 government scales, compared to 47 – 65 scales at various unofficial landing sites in and around the fishing port in Muncar district. Meanwhile, in Negara district, I noted 2 to 3 unofficial scales at the unofficial landing sites and 2 government scales in the official landing site. Direct observation in the field also found that on both coasts of the strait, the unofficial landing sites were such hives of activies compared to those found in the official landing sites. Not only more fish were landed here (due to the fact that there were more scales), but services and support system were found to be more complete and diverse in unofficial landing sites: ice plants, traditional makers of fish flakes, fish vendors, food  46  Petengan comes from a Javanese term peteng, which means ‘dark’. Therefore, petengan literraly means ‘to do it in the dark’. The new moon phase is called petengan because of the absence of moonlight. The opposite of petengan is padangan, which comes from a Javanese term padang that means ‘bright’. The full moon phase is called padangan.  47  The landing of catches in the unofficial landing sites was in ‘full view’ of the fisheries field enumerators and fisheries officers. The sheer amount of lemuru landed at any given fishing-day making it impossible to have them all landed and weighed in government landing site. However, based on my field observation during participatory slerek fishing on 22 – 23 September 2004 and various opportunistic surveys of the fishing ports, the landing often occur during the time (i.e., 5 to 6am) when the fisheries field enumerators and fisheries managers were not present yet.  48  The line-up for the scales was too long in the official site, rendering spoilage of the fish and causing the price to drop from canned fish price to fish meal price.  54  and drink vendors, fish traders/dealers, pedicab drivers, rentals of pick-up trucks, boat workshops, engine spare parts shops, cigarette vendors, etc. In other words, the unofficial landing sites provide venues for the network of stakeholders to thrive and form into a community safety net for the fishing community in the area. Various respondents also admitted that the location of the unofficial landing sites is perceived to be more convenient for the fish porters (who are paid flat fee per carry-out service) to carry the fish back-and-forth from the boats to the trucks 49. One additional but main reason why fishers prefer to land their catches in unofficial landing sites is the fact that these sites are outside the responsibility area of the government enumerators (Respondent SB53, personal communication, 12 September 2004) and thus, catches are not recorded nor reported. As a consequence, fishers do not pay landing tax 50 when they land fish in these sites (Respondent SB35, personal communication, 19 August 2004; Respondent SB75, personal communication, 13 October 2004). This last issue will be dealt with in detail in Chapter 3. Assuming no discards, on an average fishing-day about 28 to 38% (mode at 30%) by weight is diverted as ‘take-home’ from the total catch prior to landing (Respondent SB53, personal communication, 3 July 2002; Respondent SB35, personal communication, 19 August 2004, 20 August 2004; Respondent SB56, personal communication, 30 June 2007; Respondent SB75, personal communication, 13 October 49  In Negara district, the new pier that was built in the government fishing port was too long and fish porters (panol) don’t like to carry the heavy fish basket that far given the flat fee which was set at Rp.50,000 or US$8.00 (2004 value) per tonne of lemuru landed (Respondent SB2, personal communication, 3 August 2004). In Muncar district, during direct observation, I noted that the location of the pier was as such that it was too high to climb up during low tides, rendering difficulties for fish porters to walk along the ramps onto the pier from the boats. Meanwhile, access from fishing boats to beach landing sites were much easier.  50  According to the Decree of the Minister of Marine Affairs and Fisheries No. PER.05/MEN/2008 (Minister of Marine Affairs and Fisheries of the Republic of Indonesia, 2008), all fishing boats < 6GT (viz., small-scale fishery) are not mandated to report their catches; meanwhile, all other boat sizes have to report their catches. The implication is that small-scale fishery sector does not have to pay landing tax. Landing tax (intres or retribusi) is regulated and coordinated at provincial level. The tax brackets vary, and they depend on the province, and scale and types of fishery. For the Bali Strait slerek fishery that generally use 25 – 30GT boats in 2004, the landing tax is 2% in Jembrana regency (Respondent SB4, personal communication, 4 August 2004), where fisher pays 1% and buyer pays 1%; and 4% in Banyuwangi regency (Respondent SB53, personal communication, 12 September 2004), where fisher pays 2% and buyer pays 2%. This tax is part of fisheries sector contribution to regional gross domestic product. See footnote 54 for details on landing tax.  55  2004). The amount may seem a lot. However, considering the number of people involved, the amount seems reasonable as incentives, in-kind payment and as ‘giveaway’ to 85 – 112 persons each time (see details in Figure 1.8 in Chapter 1 and Appendix D): 25 to 50+ crew members (pandega) in each fleet, a group of 12 fish porters (panol) per fleet, a group of 6 boat scrubbers/cleaners (penguras) per fleet, 2 – 4 fish unloaders (pengisi) per fleet, and about 40+ women and children who would swarm (ngujur) around the fleet once they drop anchor. There were also some cases when crew members (pandega) took more fish than they were entitled to and sold them off to some fish dealers/brokers (belantik) without the knowledge of the fleet owners (Respondent SB44, personal communication, 24 August 2004; Respondent SB45, personal communication, 24 August 2004). Therefore, on average, I estimate that only 62 to 72% (mode at 70%) of the total fish catch gets as far as the weighing scales (viz., landed). Using narrative accounts about unreported amount of catches (Respondent SB35, personal communication, 19 August 2004; Respondent SB75, personal communication, 13 October 2004), I then estimated that only 28 to 56% (mode at 45%) 51 of what is actually caught is landed in official landing site (i.e., government fishing port), while 14 to 42% (mode at 26%) would likely arrive at unofficial landing sites/scales (Figure 2.1). These accounts are in agreement with the accounts of the owner of the 2 – 3 unofficial scales in Negara district (Respondent SB36, personal communication, 20 August 2004). In the absence of a true fish auction usually held at official landing site, the fate of fish post-weighing is quite dynamic: some fish are forwarded to various fish plants and  51  This is for estimation of lemuru landing during 1974 – 2001, after slerek gear is introduced in the strait. For the historical landing of lemuru (1950 – 1973), using accounts from very old (80+ years old) respondents (Respondent SB19, personal communication, 8 August 2004; Respondent SB23, personal communication, 10 August 2004), I guesstimated that the take-home amount was about 3%, and that only 40 to 80% of what was actually caught was landed in official landing site (the port where the local cooperative was located), while 17 to 57% would likely arrived at unofficial landing sites. These lemuru were caught using various gear prevalent in those decades, i.e., serok, jala buang, payang oras, jaring eder, bagan tancap and banjang kerei (see text). The higher percentage of lemuru landed in official site was due to the fact that there were still very few fishers at the time and thus, tallying catches was easier.  56  cold storage plants as per initial financial contract agreements 52, some are sold to beachbased traditional fishing enterprises, and some are bought by other entrepreneurs. The fish bought by the other entrepreneurs are either shipped out of the area or re-sold back to the area to other buyers. The sardine fishery trade flow system in the Bali Strait is complex (Figure 2.1) as this is a reflection of the underlying crowded and fragile network of human interdependence as discussed in Chapter 1 (see Figure 1.8 in Chapter 1). If one needs to estimate the true catch of the fish, the only feasible option is to trace the data directly from the fishers who first caught the fish (ex-vessel), which is what was implemented in this study and detailed quantitatively in Chapter 3.  52  In the absence of formal access to financial loans in banks, many slerek owners (juragan darat) have opted for binding interest-free financial contracts with fish plants, cold storage plants and/or other private financiers (i.e., pengambeg, see Figure 1.8 in Chapter 1, and Appendix D for details) since early 1980s. Essentially, the contract finances the fishing operations. The loan is given upfront to the slerek owners and is paid back through the price difference between the set price with the market price, which is about US$0.01/kg in 2004. The set price is unilaterally determined by the financier. As a return, once under contract, the financiers are obliged to buy the fish in both glut and low seasons, giving peace of mind to the slerek owners. Additionally, the financiers also take care by providing money in difficult times, such as when family members are sick and need hospitalization, funerals, weddings, children’s school tuition fees and other family needs. The money that was given in most cases is not expected to be paid back, as it symbolizes the deepening of the relationship between the financier (the patron) and the fleet owner (the client). It is not necessarily a loan in a true sense as there is no interest. In a few instances, when the slerek owner died before the principal amount is fully paid, the balance of the ‘loaned’ money is carried forward to the heir (beneficiary) without interest. Some relationships are so harmonious and trustful that even Islamic pilgrimage trips to Mecca are taken care of. This patron-client relationship somehow forms a social insurance system for the fishing communities-at-large (Emmerson, 1976, 1987), a form of social safety net that buffer the community during the difficult time, such as during La Niña episodes when fish is scarce. Some successful slerek owners eventually are recognized by banks and become financiers for other boats, while still keeping their status as an ambeg’an (one who is financed) of their own pengambeg (financier). By 2008, after the continuing increase of fuel price and the 2006-2007 El Niño incident that created glut (and in consequence, the ex-vessel fish price fell), unfortunately, some of these juragan darat and pengambeg went bankrupt, unable to pay their debt to the banks. The context and scope of this issue is discussed in Chapter 3.  57  Figure 2.1  Schematic diagram of the fate of caught lemuru once landed; (1) takehome fish refers to unaccounted fish given to crew members, boat cleaners, fish unloaders and fish porters as in-kind payment and incentives, and also to fish that were given-away to women and children who are ngujur (see Figure 1.8 in Chapter 1, and Appendix D); (2) including canneries, fish meal & pellet producers; (3) including brokers/dealers, buyers/processors, retailers and private financiers; (4) fish are traded either back into the system (local) or go out of the system (regional). Shaded area depicts landing beaches. The percentage values in this figure apply to 1974 – 2001 period, see footnote 51. Length & width of arrows are not significant.  2.5.2 How the Landed Fish are Enumerated and Reported Throughout the System In Indonesia’s fisheries statistical system, information on production means (fishing units, fishery establishments and fishers) is collected through frame surveys or licensing systems (especially for industrial fisheries); while quantity of catch and value are sampled in fishing ports and landing places (FAO-SEAFDEC, 2005). In Indonesia, there are two main government bodies officially responsible for collecting, enumerating and reporting fisheries statistics: (1) the Ministry of Marine Affairs and Fisheries (or DKP) through the District Fisheries Office (Cabang Dinas Perikanan Kecamatan), and (2) the Central Statistical Bureau (or BPS) through the Regency Statistics Office (Kantor Statistik Kabupaten). Both agencies have their own enumerators that are responsible for collecting, enumerating and reporting related data and information. Additionally, 58  according to FAO-SEAFDEC (2005), a third government body, the Custom Office (reports to the Ministry of Finance) also collects and reports fisheries statistics. However, data collated and reported by the Custom Office (Direktorat Jenderal Bea dan Cukai) usually comprised marine and seafood products in their final form (i.e., canned, frozen, smoked, chilled, fresh, meal, etc.) destined for overseas export and not the ex-vessel landing of catches. Therefore, the Custom Office is excluded in my analysis. The fisheries field enumerators cover fisheries statistics only, while the statistical bureau field enumerators cover data and information from various sectors, including the fisheries sector. Both types of field enumerators work at the village level (ex-vessel) (Figure 2.2). Depending on where the fish are weighed, the records are not automatically enumerated. If the fish are weighed at government-owned scales located at the official fish landing sites within a government fishing port complex, the amounts will be recorded daily (Respondent SB53, personal communication, 12 September 2004) by the fisheries field enumerators who later submit the report to the DKP’s District Fisheries Office (Figure 2.2). Fish landings at the government sites are recorded every day but during the fieldwork in the study area, I never observed any fisheries field enumerators recording landings at the scattered unofficial scales in Muncar. This finding is corroborated by the information I received from a local fisheries officer that the government fishing port in Muncar has 15 fisheries field enumerators in 2004 which are only responsible for monitoring the weighing 53 of landings at those 8 government-owned scales on a daily basis (Respondent SB53, personal communication, 12 September 2004). Meanwhile, in Negara, landing monitoring is a bit more widespread; the 4 fisheries field enumerators that were employed there say that they divide tasks between official (2 scales) and unofficial landing sites (2 – 3 scales) (Respondent SB79, personal communication,  53  In Muncar government fishing port, fish weighing is only carried on a sampling basis: 3 baskets per fleet unit, the average basket weight (viz., 80kg in 2004) will be used as a factor to estimate landing based on the numbers of baskets carried by fish porters for each fleet. The sampling system was selected to avoid long waiting in line-ups that would easily spoil the fish in the high ambient temperature.  59  August 2004). However, monitoring at unofficial landing sites do not necessarily cover all unofficial scales used and records were not necessarily synchronized between official and unofficial sites (Respondent SB6, personal communication, 5 August 2004). This is also corroborated by the fact that during the fieldwork: (1) neither my two assistants nor I saw any of the 4 fisheries field enumerators recording landings by the unofficial scales, and (2) during our several visits to the Jembrana Regency Fisheries Office in 2004, we were told by various fisheries officers that they did not have any current official records of fish catch in the regency and they did not keep any copies of their own previous records either. The only reason that they gave us for this was that all records (apparently in hardcopies only) were lost and destroyed during a move from an old office to the new office. Records from government landing sites (at village level) are collected and enumerated  daily  and  assembled  every  month  (Respondent  SB53,  personal  communication, 12 September 2004) for reporting to the DKP’s District Fisheries Office, and they are then submitted to their respective subsequent superiors (i.e., regency and provincial level) through the official channels all the way up to the national level in Jakarta (headquarters of DKP), and eventually to the international level at the headquarters of FAO-UN in Rome, Italy (Figure 2.2). At the regency level, data are initially processed; then at the provincial level, they are compiled and further processed; and finally, while at the national level, the data sets are again further compiled, processed, analyzed and then disseminated (FAO-SEAFDEC, 2005). The District Fisheries Office also has a collaborative link with the District Government (Kecamatan) Office that may include, among others, sharing of the fisheries statistics reports. Since the statistical bureau field enumerators have a larger responsibility for data enumeration and collection from various sectors, and in many cases they work alone in a district, they don’t necessarily record the fisheries statistics themselves at government landing sites. Often, the statistical bureau field enumerator himself requests data directly from the fish plants, cold storage plants and from the DKP’s District Fisheries Office (Respondent SB78, personal communication, 18 October 2004; Respondent SB80,  60  personal communication, 15 September 2004). It is also important to note here that the statistical bureau field enumerators do not record landings at the unofficial scales (Respondent SB80, personal communication, 15 September 2004). Once summarized and synthesized, they would then periodically report their statistics data (including fisheries data) to their superiors at the BPS Regency Statistics Office, which will then submit them to their respective superiors through the official channels all the way up to the national level in Jakarta (headquarters of BPS), which is the national repository for cross-sectoral statistical data and indicators (Figure 2.2). It is this data that is usually available on request by international agencies such as UNDP (United Nations Development Programme) or national agencies such as BAPPENAS (National Development Planning Coordinating Board) to implement their various mandates, such as those that fall within the Millennium Development Goals (MDG). Many international researchers also seek their relevant data and information from BPS to support various projects, such as the many projects that started up after the disastrous tsunami that struck the Aceh region in December 2004. Given the increasing awareness of regional autonomy after the implementation of the Local Autonomy Act (UU No. 22/1999) in Indonesia, following the 1998 reform, local governments have gained new authority concerning marine fisheries management (Satria & Matsuda, 2004). As such, the District Government Offices also request data from  the  statistical  bureau  field  enumerators  (Respondent  SB80,  personal  communication, 15 September 2004), which will then submit them to their respective superiors through the official channels all the way up to the national level in Jakarta (headquarters of the Ministry of Home Affairs or DEPDAGRI), to be shared with the Regional Development Planning Coordinating Board (BAPPEDA) along the way, and ultimately to the National Development Planning Coordinating Board (BAPPENAS) as well (Figure 2.2). Looking at the way the data reporting system works (Figure 2.2), in theory, results obtained through the BPS channel (enumerated by statistical bureau enumerators) should be the same as the results obtained from the DKP channel (enumerated by the fisheries  61  field enumerators) as they pertain to the same thing. In reality, they are not the same, as will be shown in the next section. The ineffectiveness of the enumeration system and the complexities of the fisheries system have resulted in some overlap (and possibly multiple-counting) of data enumeration in one area while data from other areas remains unreported. 2.5.3 Inconsistencies in Data Reporting To cross-check whether the fisheries catch statistics reporting system throughout jurisdictions and responsible agencies is consistent, I collated and compared nominal catch data of S. lemuru from the Bali Strait from 1950 to 2005 (Figure 2.3), across all government jurisdictions, caught and landed by fishers from the eastern coast of East Java and western coast of Bali. Specific attention was given for the time period of 1972 onwards when the Bali Strait catches were reported using the standardized system. The data set mainly includes official records from both the DKP and BPS channels as discussed above, and from personal journals of a retired chief of a government landing site, individual scientists, researchers in government agencies and multi-agency projects. The 1950 – 1971 data sets that was collected and enumerated (at the village level by nowdefunct local cooperatives) prior to the establishment of the standardized statistics system was presented to retrace the historical records of lemuru catch in the strait. Results show that throughout the 1972 – 2001 period no single data set fully matches any other (Figure 2.3 and 2.4), although there is agreement in some general trends. In fact, the nominal catch recorded by the government landing sites, at the village level (red circles in Figure 2.4), are quite different from the median of all reported nominal catch (solid line in Figure 2.4). It appears that different levels of jurisdiction report their own versions of the data, with a tendency that values reported from village level (i.e., the government landing sites at the fishing ports) are the lowest of all (Figure 2.3 and 2.4).  62  Figure 2.2 Schematic diagram of the slerek fishery system in the Bali Strait that shows fate of landed lemuru, how landed lemuru are officially enumerated and how the statistics are reported and used throughout the official channels. Licensing procedures are excluded in this diagram. In brackets and italics are jurisdiction names in Indonesian. 63  Five reasons may explain why these data reporting inconsistencies throughout jurisdiction may be happening; (1) significant catches are landed in unofficial sites, and those that are landed in official sites are either under-reported or misreported, (2) adjustments to catch and landed value statistics for political reasons, (3) misidentification of common names of fish, (4) misidentification of scientific names, and (5) continuous change in administrative boundaries that impacts the boundaries of fisheries statistical areas. 2.5.3.1 Significant Catches are Landed in Unofficial Sites As elucidated in Figure 2.1, significant amount of catches are landed in unofficial landing sites. Even for landings made at the government landing sites, under-reporting and misreporting are quite common as previously noted by the various accounts of respondents during interviews. In the day-to-day life of the fishing communities, given the uncertainty in fish catch, lack of any government social safety net, lack of trust and respect for the authority, an almost non-existent financial access for the fishers, and the obligation to pay landing taxes 54, it appears that catch reporting is being used by the fishers as a leverage to hedge against uncertainty and poverty (see detail in Chapter 3). In other words, after unreported take-home amount of catches have been accounted for amongst the crew members, boat cleaners and fish porters, etc., fleet owners very rarely report their true landings to the local fisheries authority.  54  In Indonesia’s fishery system, the actual catch reported to the authority will be used to determine the amount of landing tax (locally known as intres or retribusi) that the fishers need to pay. The landing tax for slerek fishery has been set by regional regulation (perda) to 5%, where 2.5% is supposed to be paid by the fisher and 2.5% by the buyer. However, the 1998 reform era stipulated various packages of tax deregulation. For some time, the landing tax was completely abolished. Then in early 2000, it was reintroduced and prompted complaints from fishing communities in Jembrana regency, while in Banyuwangi regency, fishermen ignored it altogether. Following demonstrations critizing the reintroduction of the landing tax, the tax level was reduced, in Jembrana to 2% (Respondent SB4, personal communication, 4 August 2004), where fishers pay 1% and buyer pays 1%, and in Banyuwangi to 4% (Respondent SB53, personal communication, 12 September 2004), where fishers pay 2% and buyer pays 2%. However, in practice, the collection of landing tax is sporadic, both in spatial and temporal terms; and sometimes it is not collected at all.  64  Figure 2.3 Profile of reported/nominal annual catch (tonnes) of Lemuru (Sardinella lemuru) from the Bali Strait, 1950 - 2005, as recorded by various sources and agencies across jurisdictions. Data sources (see legend): (1) various unpublished landing record archives of KUD Mina Blambangan Fishers’ Cooperative, 1975 – 1996 (Respondent SB75, 1997b), and Merta et al. (2000, Table 4); (2) Merta et al. (2000, Table 24 and 37); (3) Cabang Dinas Perikanan Daerah Kabupaten Banyuwangi (1990, 1991, 1992), Dinas Perikanan Daerah Kabupaten Daerah Tingkat II Banyuwangi (1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000), Dinas Perikanan dan Kelautan Kabupaten Banyuwangi (2001, 2002, 2003), and Merta et al. (2000, Table 25); (4) Badan Pusat Statistik Kabupaten Banyuwangi & BAPPEDA Kabupaten Banyuwangi (2002), BAPPEDA Kabupaten Banyuwangi & Badan Pusat Statistik Kabupaten Banyuwangi (2003, 2004), BAPPEDA Kabupaten Banyuwangi & Kantor BPS Kabupaten Banyuwangi (2001), BAPPEDA Kabupaten Dati II Banyuwangi & Kantor BPS Kabupaten Banyuwangi (2000), BAPPEDA Kabupaten Dati II Banyuwangi & Kantor Statistik Kabupaten Banyuwangi (1995, 1997), Kantor Statistik Kabupaten Banyuwangi (1992, 1994, 1996), Kantor Statistik Kabupaten Banyuwangi & BAPPEDA Kabupaten Dati II Banyuwangi (1998, 1999), Kantor Statistik Kabupaten Jembrana (1987, 1988, 1990), Kantor Statistik Kabupaten Jembrana & Kantor Bupati KDH Tk. II Kabupaten Jembrana (1992, 1993); (5) Dinas Perikanan Daerah Propinsi Daerah Tingkat I Jawa Timur (2000), Dinas Perikanan Daerah Tingkat I Propinsi Bali (2000); (6) Merta et al. (2000, Table 5, 6 and 9); (7) Fakultas Perikanan UNIBRAW (1999), Fakultas Perikanan UNIBRAW, Dinas Kelautan dan Perikanan Propinsi Jawa Timur, & Dinas Kelautan dan Perikanan Propinsi Bali (2007); (8) Ritterbush (1974), and Ghofar & Mathews (1996); (9) Directorate General of Capture Fisheries (2001, 2002, 2003, 2004, 2005, 2006, 2007), Directorate General of Fisheries (1974, 1979, 1980a, 1981, 1982, 1985a, 1985b, 1986, 1987, 1988, 1989, 1990a, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, no date-a, no date-b, no date-c).  65  Figure 2.4 Distribution of reported/nominal annual catch (tonnes) of the Sardinella lemuru from the Bali Strait, 1950 – 2001. Each open dot (○) represents a reported landing for a particular year by a particular jurisdictional authority. The red circles (●) represent reported landings from the combined government landing sites located at the lowest level of jurisdiction, the village. The line represents the median, while error bars denote 5th and 95th percentiles. Grey shading depicts possible range of landing distribution. Slerek started in 1974. The 1950 – 1971 period represents historical catch of lemuru, while the 1972 – 2001 period represents data sets collected using standardized statistical system.  The level of actual reporting varies according to whether or not they made money the previous month (Respondent SB35, personal communication, 19 August 2004). On average, after 30% 55 take-home amount, fishing fleet owners would only report about 70% of what they actually landed, the minimum amount that they would report is 50% and the maximum amount that they would report is 75% (Respondent SB35, personal communication, 19 August 2004). The amount of misreporting in the study area varies from fisher to fisher; however, this account may shed some light on the uncertainty of the nominal catch records.  55  Take-home is estimated to be about 28 to 38% (mode at 30%).  66  Meanwhile, realizing these tendencies, local fisheries authorities appear to employ some adjustment measures to account for the missing and misreported data. As seen in Figure 2.3, the nominal catch data from the District reports are much higher that those reported by the government landing site reports at village level. Adjustment of the data to be reported to the higher authority is not an uncommon practice, though often not readily admitted. Local fisheries authorities in the Bali Strait did not openly acknowledge this, although they acknowledged the amount of catch data that they believed were missing from monitoring (Respondent SB53, personal communication, 12 September 2004; Respondent SB75, personal communication, 13 October 2004). Other local fisheries authorities such as those in eastern Indonesia (where I implemented a similar suite of methods for a study not included in this thesis) openly admitted these adjustment measures, as they realized the fact that there were fish landings that were out of their control given the scattered landing locations (Respondent KM60, personal  communication,  11  December  2004;  Respondent  KM118,  personal  communication, 2 February 2005; Respondent KM113, personal interview, 4 February 2005). The magnitude of the adjustment is not always the same for every location and every time, and is usually based on the best guesstimation by the officers in question. However, such adjustment can be re-assessed, with difficulty, by tracing the fish flow carefully from when and where it’s landed (ex-vessel). However, data adjustment that is solely based on non-technical reasons, which is hard to trace, leads us to a second reason. 2.5.3.2 Adjustments to Catch and Landed Value Statistics for Political Reasons In the context of Indonesia’s sectoral development system, when it comes to progress evaluation of sectoral development, success level is usually defined by the size of production generated by the sector; and in the case of fisheries, it’s the size of production and production value of the landing (S. Nurhakim23, personal communication, 20 November 2007). Sectoral development success in a region (i.e., regency level) will be used as a bargaining tool by the regency in question to propose an annual budget to the provincial and central governments for next year’s fiscal year (S. Nurhakim23, personal communication, 20 November 2007). 67  Consequently, the fisheries annual production report is expected to almost always show positive growth. A frequent incidence of negative growth (both for production and landed value) would be used as an indication that the fisheries sector is not productive and therefore does not warrant enough funds to be allocated for the development or management budget. This is when various adjustments to the fish landings and landed value statistics occur (S. Nurhakim23, personal communication, 20 November 2007) and it is almost impossible to assess or estimate its magnitude as it is never disclosed officially. 2.5.3.3 Misidentification of Common Names of Fish Indonesia has the world’s richest marine biodiversity. For example, with more than 2,000 species of fish recorded (Froese, Luna, & Capuli, 1996) and more than 700 distinct languages spoken 56 (Gordon, 2005) by over 700 ethnic groups 57 (Joshua Project, 2008), there is an enormous diversity of common names for the organisms with which people interact (Buchary, Wahyuningsih, & Pauly, 2003). The Bali sardine is known to local (mainly Muslim 58) fishing communities in Bali and East Java by four different names, depending on the age-classes as determined by the average total length: (1) sempenit for juvenile stage, approx. 3 to 10 cm; (2) protolan for sub-adult stage, approx. 10 to 13 cm, (3) lemuru for adult stage, approx. 13 to 18 cm, and (4) lemuru kucing for senescent stage, approx. 18 to 25 cm (Respondent SB53, personal communication, 3 July 2002). However, in day-to-day parlance, fishermen and fish traders assign the fish into four different categories based on its price schedule (Respondent SB2, personal 56  The number of languages listed for Indonesia is 742. Of those, 737 are living languages, two are categorized as second language without mother-tongue speakers, and three are categorized as extinct (http://www.ethnologue.com/show_country.asp?name=ID).  57  Joshua Project (http://www.joshuaproject.net/) recorded 767 ethnic groups in Indonesia.  58  The Balinese Hindu fishing communities have a different naming system for the fish. They call both ‘lemuru’ and ‘lemuru kucing’ as kocing and bei kocing, respectively; wherein bei = fish, and kocing is simply a fish name. While ‘sempenit’ and ‘protolan’ are called the same, there is another name for a size-class between ‘sempenit’ and ‘protolan’. This is called ‘lentreng’ and essentially is a larger sempenit (approx. 8 – 10 cm).  68  communication, 3 August 2004; Respondent SB6, personal communication, 5 August 2004; Respondent SB56, personal communication, 20 September 2004; Respondent SB67, personal communication, 30 September 2004; Respondent SB70, personal communication, 5 October 2004): (1) sardén 59 for lemuru and lemuru kucing, those high quality larger-size fish which fetch high prices and are processed for canning; (2) tepungan or ikan tua 60, for smaller-size fish (i.e., sempenit and protolan) and spoiled larger-size fish (i.e., lemuru and lemuru kucing) which fetch lower prices and are processed for fish meal at the plants; (3) pilihan 61, for a mixture of different species (other than the sardines and mostly are not economically important) and age-classes, which also fetch low prices, similar to tepungan and are usually sold to the fish meal plants, and (4) apkir 62, for fish (sardines or otherwise) of reject quality by fish meal plants and canneries, which are therefore sold to beach-based traditional fish meal entrepreneurs for making flake meal. It is with these price-based common names that fishermen record their catches in their personal log-books (buku bakul) 63. Nevertheless, due to the difficulties in observation and enumeration in the hustle and bustle of the landing sites, fisheries and statistical field enumerators and local fisheries officers do not record daily landings 64 based on local biological traits (i.e., ageclasses: sempenit, protolan, lemuru and lemuru kucing) nor local trade parlance preferred by the fishermen (i.e., sardén, tepungan or ikan tua, pilihan, and apkir). Everything is lumped into one category and called ‘lemuru’. Consequently, the published reported catch statistics for lemuru have also never been differentiated based on age-classes. 59  Sardén = sardines.  60  Tepungan = of meal quality; ikan tua = old fish, refers to its nature that rots easily, in the absence of enough icing.  61  Pilihan = of something that needs to be sorted and weeded.  62  Apkir = of reject quality, something that is rejected.  63  As observed in the personal log-books of Respondents SB70, SB56 and SB73, all slerek owners. They kindly donated these log-books to me during the fieldwork.  64  As observed in the various unpublished landing record archives (1975 – 1996) of KUD Mina Blambangan Fishers’ Cooperative, which is the government-appointed authority that kept and maintained official landing records at the village level in Muncar district. Respondent SB75, who was a former manager of the cooperative and a former chief of government fish landing site, kindly donated these archives to me during the fieldwork. They were his own original personal copies.  69  Whether these lemuru statistics refer to data from ‘lemuru-size fish only’, or also from lemuru of other age-classes and of other price-schedules, is not clear. Respondents’ answers were also not clear. 2.5.3.4 Misidentification of Scientific Names As ‘lemuru’ is also the standard Indonesian name for S. lemuru set by the national statistics system (Dwiponggo, 1987), it is under this name that the statistics are summarized and reported from the landing sites on the coasts, to the district office, regency office, provincial office and all the way to the national office. These lemuru statistics from the Bali Strait will then be lumped together with other ‘lemuru’ statistics that come from other regions in Indonesia, and later reported to FAO as the production of lemuru (Sardinella lemuru) from Indonesia. However, this is a serious misnomer as will be explained below. While lemuru from the Bali Strait is taxonomically identified as Sardinella lemuru (Gloerfelt-Tarp & Kailola, 1984), not all fish identified as lemuru from other areas in Indonesia are from the same species, although S. lemuru also occur in some other parts of Indonesia, such as in south of Ternate (near Ambon), Jakarta Bay and off Central Java in the Java Sea (Gloerfelt-Tarp & Kailola, 1984). According to Burhanuddin, Hutomo, Martosewoyo, & Moeljanto (1984, cited in Merta, 1995), in Indonesia’s fisheries statistics system, the common name ‘lemuru’ apparently also includes four other taxonomically distinct species: Sardinella longiceps (English: Indian oil sardines; local names: tembang moncong, temban monco, bete lalaki, lemuru, tembang mata kucing, malaka, mapikal, teli, maa pirale), S. aurita (E: round sardinella), S. leiogaster (E: smooth-belly sardinella; LN: ciro, siro, lemuru, sibula, sembula, tanjan, sabula laisi), S. clupeiodes (E: Bleeker smooth-belly sardinella; LN: tengkurung, siro, ciro, cincin, lemuru, sabula laisi, sembula, tanjan), which then in the data aggregation process would all be lumped together with Amblygaster sirm (E: spotted sardinella, LN: ciro, siro, lemuru, lengoro, lenguru, sabola, sardin, sembula, sibula, tanjan, tengkurung, teros). The  70  five distinct species 65 have their own local common names as noted above, viz., Schuster & Djajadiredja (1952), which is updated by Buchary et al. (2003) and incorporated into www.fishbase.org. However, they seem to be lumped together into “lemuru” in the interests of simplicity. Hence, Indonesia’s lemuru statistics as reported by the national fisheries office and later reported to FAO cannot be used at face value as Sardinella lemuru for reasons stated above. Fortunately, realizing this situation, from 2004, Amblygaster sirm is being reported separately from the Sardinella spp. (I.G.S. Merta 66, personal communication, 27 June 2007). It is unclear, however, whether the five sardinella species (S. longiceps, S. lemuru, S. aurita, S. leiogaster and S. clupeiodes) are now reported separately from each other. The Indonesian fisheries statistics system developed by Dr Tadashi Yamamoto in the early 1970s (Yamamoto, 1980) created a standardized set of statistical forms with Indonesian common names, both for species names and fishing gear (Directorate General of Fisheries, 1975a, 1975b, 1980b, 1990b). It is these forms that are then distributed to all regency and district fisheries offices throughout Indonesia for data collection procedure. While the creation of these forms is very good for an efficient standardized system for a country the size of Indonesia, unfortunately, not all fisheries field enumerators understand the Indonesian common names listed in the form (due to language barriers and the multi-species and multi-gear nature of the fisheries), leading to misallocation of 65  The following is the geographic distribution of these five species using common name accounts documented by Schuster & Djajadiredja (1952): (1) Sardinella longiceps is found in South Sulawesi, Jakarta Bay, Central Java, and Maluku; however, according to FishBase (www.fishbase.org), studies and reports pertaining to S. longiceps from Indonesia probably refer to S. lemuru; (2) S. leiogaster is found in South Sulawesi, Jakarta Bay, Central Java, and West Java; (3) S. clupeiodes are found in South Sulawesi, Jakarta Bay, Central Java, and West Java; (4) Amblygaster sirm is found in Jakarta Bay, Central Java, West Java, Madura, and South Sulawesi; meanwhile, according to Gloerfelt-Tarp & Kailola (1984), A. sirm is found from southwest Sumatra to Timor Sea; and (5) S. aurita, however, is not documented in Schuster & Djajadiredja (1952). Clupeoid taxonomy revision by Wongratana (1983) have allocated S. aurita in Indonesia into S. lemuru (I.G.S. Merta, personal communication, 30 July 2009). Additionally, FishBase (www.fishbase.org) does not list S. aurita to occur in Indonesia either.  66  Senior fisheries scientist at the Research Centre for Capture Fisheries of Indonesia’s Ministry of Marine Affairs and Fisheries, and one of the members of my thesis supervisory committee.  71  fish species and fishing gear into the wrong categories 67 (E. Buchary, personal observation in eastern Indonesia, October 2004 to February 2005). 2.5.3.5 Continuous Change in Administrative Boundaries The last and fifth reason for the inconsistencies in landing data reporting is the continuous change in the fisheries statistical area boundaries, due to changes in Indonesia’s geopolitical map that affect administrative boundaries (Table 2.1) and recent developments in fisheries management zones.  Table 2.1.  No.  67  The effects of Indonesian geopolitical map changes to the Indonesian fisheries statistical areas: 1973 to present. Data sources: (Directorate General of Capture Fisheries, 2001, 2007; Directorate General of Fisheries, 1974, 1980a, no date-a). Period  Number of provinces  Number of statistical areas  Notes  1.  1973 - 1974  26  26  Standardized fisheries statistical system established in 1973. Statistical areas followed exactly provincial boundaries.  2.  1975 - 1977  26  32  Improvement in data collection system by accounting for locations of the coastal areas in all provinces had increased the number of statistical areas; but they were still located within respective administrative boundaries.  During a fieldwork (not part of this thesis) in West and East Nusa Tenggara provinces in 2004-2005, I observed accounts of species and gear misallocation in the standardized statistical forms completed by fisheries field enumerators. Unfortunately, this is not an uncommon case, as field enumerators are more familiar with common names in their own languages (Respondent KM21, personal communication, 29 November 2004; Respondent KM22, personal communication, 29 November 2004; Respondent KM24, personal communication, 29 November 2004; Respondent KM25, personal communication, 29 November 2004). They acknowledged this problem and admitted that some of them had repeatedly asked their main offices for some fish and gear posters (which are published by DKP) to cross-check the common names with (Respondent KM2, personal communication, 29 November 2004; Respondent KM23, personal communication, 29 November 2004), but none had arrived at the time.  72  Table 2.1.  No.  The effects of Indonesian geopolitical map changes to the Indonesian fisheries statistical areas: 1973 to present. Data sources: (Directorate General of Capture Fisheries, 2001, 2007; Directorate General of Fisheries, 1974, 1980a, no date-a). Period  Number of provinces  Number of statistical areas  Notes  3.  1978 - 1998  27  33  East Timor was annexed by Indonesia in 1975 and became the 27th province in 1976. The statistical area system was modified by 1978. Everything remained stable during the New Order regime of Suharto.  4.  1999 - 2004  26  32  Democracy reform took place in 1998, ousting the three decades of Suharto regime. East Timor became a sovereign state in 1999: one less province, one less fisheries statistical area.  5.  2005 - present  33  41  Following 1998 reform, an enactment of Local Autonomy Act (UU No. 22/1999) had spurred euphoria of devolution of power, resulting in the creation of new provinces in almost every quarter.  A first attempt to improve the statistical system had been initiated in 1999 with the enactment of the Decree of the Minister of Agriculture, KepMentan No. 995/Kpts/IK.210/9/99 when those 41 statistical areas (Table 2.1) were rearranged into 9 large fisheries management zones (Figure 2.5a), but it was not implemented until 2004 with the stipulation of Fisheries Act (UU No.31/2004) (Sulistiyo 68, personal communication,