T H E E C O L O G I C A L A N D S O C I O - E C O N O M I C S U S T A I N A B I L I T Y O F T H E R E E F F I S H E R I E S O F P U L A U B A N G G I , S A B A H , M A L A Y S I A by L O U I S E S.L. T E H B.Comm.(Hons.), University of British Columbia, 2000 THESIS S U B M I T T E D I N P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R OF S C I E N C E in The Faculty of Graduate Studies Resource Management and Environmental Studies T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A January 2006 © Louise S.L. Teh, 2006 A B S T R A C T Many coral reef fisheries in developing nations have been, or are currently being over-exploited, leading to negative ecological and socio-economic effects. Yet, many reef fisheries are often marginalized from policy makers, leading to a lack of effective management and a poor understanding of these valuable fisheries. For example, the small-scale reef fisheries of Pulau Banggi, Sabah, Malaysia, have neither been actively studied nor managed before. This study seeks to understand the dynamics o f Banggi's reef fisheries. I first characterize and analyze the ecological, social, and economic aspects o f Banggi's reef fisheries. This is done based on monitoring fish landings and interviewing local fishers during two field periods in 2004 and 2005. I then use an ecosystem modelling tool to explore different policy options, and identify a management strategy which is most appropriate given local social and economic conditions. This study suggests that Banggi's reef fisheries resources may still be relatively productive, but have declined significantly over time. They are likely to become depleted in the future i f current rates of exploitation are maintained. Results from an ecosystem modelling exercise, considered within the context of Banggi's socio-economic environment, indicate that: i) the establishment of a small, community-managed marine reserve within the current fishing grounds; and ii) the implementation of alternative livelihood programmes, w i l l be the most suitable courses of action for sustainable management of Banggi's reef fisheries. n T A B L E O F C O N T E N T S Abstract ii Table of Contents iii List of Tables vi List of Figures vii L i s t of A p p e n d i c e s v i i i Acknowledgements ix 1 Introduction 1 1.1 Problem statement 1 1.2 Research objectives 1 1.3 Methods • 2 1.4 Thesis outline 4 1.5 Background and literature review 5 1.5.1 Reef fisheries 5 1.5.2 Ecological and socio-economic impacts of reef fishing 6 1.5.3 Coral reef fisheries management 7 1.5.4 Marine Protected Areas 8 1.5.5 Coral reefs and reef fisheries in Sabah 9 1.6 Pulau Banggi 11 1.7 References 12 2 The Reef Fisheries of Pulau Banggi, Sabah 21 2.1 Introduction 21 2.2 Background 22 2.2.1 Site description 22 2.2.2 Coastal fishing community profile 23 2.3 Methods 24 2.4 Results 26 2.4.1 Profile of Banggi's reef fisheries 26 2.4.2 Recorded fishery data 31 2.5 Discussion 37 2.6 Summary and concluding remarks 39 2.7 References 40 i i i 3 S e a s o n a l i t y a n d T r e n d s in the R e e f F i s h e r i e s o f P u l a u B a n g g i , Sabah 44 3.1 Introduction 44 3.2 Background 45 3.2.1 Study site 45 3.2.2 Reef fisheries of Banggi 46 3.2.3 Fishing grounds 47 3.2.4 Seasonal weather patterns 47 3.2.5 Temporal trend in catch and effort data 48 3.3 Methods 49 3.3.1 Fishery data 49 3.3.2 Interviews 50 3.3.3 Data analysis 51 3.4 Results 52 3.4.1 Target species and fishing practices 52 3.4.2 Catch quantity 53 3.4.3 Catch composition 55 3.4.4 Distribution of effort 57 3.4.5 Economic benefits from fishing 59 3.4.6 Temporal trends 63 3.5 Discussion 69 3.5.1 Catches, effort, and economics 69 3.5.2 Using fishers' knowledge to explain seasonal variation 73 3.5.3 Temporal trends in catch 75 3.5.4 Management implications 79 3.6 Concluding remarks 80 3.7 References 81 4 E x p l o r i n g Management Strategies for the Reef F isher ies of Pu lau B a n g g i , Sabah 86 4.1 Introduction 86 4.1.1 Background 86 4.1.2 Study site 87 4.1.3 Reef fisheries of Banggi 88 4.2 Methods 88 4.2.1 Modell ing approach 88 4.2.2 2005 model 88 4.2.3 1980 model 93 4.3 Results 95 4.3.1 1980 model 95 4.3.2 2005 model 96 4.4 Discussion 101 4.4.1 1980 model 101 iv 4.4.2 2005 model 102 4.4.3 Marine reserves 107 4.5 Future research and concluding remarks 110 4.6 References 112 5 Conclusion 118 5.1 Summary o f main findings 118 5.2 Research approach 119 5.2.1 Potential sources of error 119 5.3 Significance of research 120 5.4 Data gaps and further research 121 5.5 Outlook 122 5.6 Concluding remarks 123 5.7 References 124 Appendices 124 v LIST O F T A B L E S Table 2.1 M e a n C P U E and catch recorded dur ing sample per iod June & Augus t 2004 31 Table 2.2 Comparison of mean C P U E by site 32 Table 2.3 E s t i m a t e d a v e r a g e f i s h i n g r e v e n u e d u r i n g l o w f i s h i n g season 35 Table 2.4 Drivers o f fishing activity in southern Banggi 38 Table 3.1 Target species and the predominant seasons/months during w h i c h they are targeted 52 Table 3.2 Da i ly catch weights hook and line, gillnet, ox-eye scad and cuttlefish spear fisheries in 4 sampling seasons 54 Table 3.3 Results of one-way A N O V A of mean catch weight for hook and line, and mean log weight for gillnet fisheries between three sampling periods 54 Table 3.4 Regression analysis of total daily landed catches by gear and wind speed 55 Table 3.5 Fish prices recorded during the first and second field seasons at three fish trading operations in southern Banggi 60 Table 3.6 Results of A N O V A o f mean log revenue for hook and line and mean revenue for gillnet 61 Table 3.7 Estimation of net monthly income and profitability for hook and line and gillnet fishers 62 Table 3.8 Reasons given by interviewees for the decline in catches 68 Table 4.1 Input parameters for 2005 model 90 Table 4.2 Estimated annual fishery landings by gear and species groups 91 Table4.3 B i o m a s s , ca tches , and effort l e v e l s for 2005 and 1980 ( d e r i v e d ) ecosystems 95 Table 4.4 Catch results under each policy 100 Table 4.5 End state economic value for gillnet, hook and line, and trap fishery under each policy 100 Table 4.6 Change in total end state biomass and economic values for target fishery groups for each policy 101 vi L I S T O F F I G U R E S Figure 2.1 Map of Sabah with Banggi off the north coast 23 Figure 2.2 Map of villages in southern Banggi 25 Figure 2.3 M a p o f M a l i a n g i n f i sh ing ground , showing spat ia l d i s t r i bu t ion o f effort 28 Figure 2.4 Composition of hook and line catch 33 Figure 2.5 Composition of gillnet catch by weight and family 33 Figure 2.6 Breakdown of daily revenue from hook and line o f 19 fishers 36 Figure 3.1 Maliangin fishing ground map showing spatial distribution of effort 46 Figure 3.2 Catch composition for hook and line landings by major fish groups 56 Figure 3.3 Catch composition of gillnet landings by major fish groups 57 Figure 3.4 Map showing seasonal usage of Maliangin fishing ground 58 Figure 3.5 Estimated annual revenue, cost, and profit for hook and line and gillnet fisheries 63 Figure 3.6 Distribution o f interview responses for the magnitude o f decrease in catch rates 64 Figure 3.7 Dai ly catch rates for 1970-2005 64 Figure 3.8 Breakdown by fishers' age for the question of when catch declines first occurred 65 Figure 3.9 Derived temporal trend for mean daily catch rates 67 Figure 3.10 Present and historic prices and average catch rates of coral groupers using hook and line and traps 69 Figure 4.1 Map of southern Banggi, showing Maliangin fishingground 87 Figure 4.2 Po l i cy 1: Status quo, fishing effort continues to increase as per historic rate 96 Figure 4.3 Policy 2: Effort increase as per historic rate, after which fishing stops 97 Figure 4.4 Pol icy 3: Effort increase equivalent to historic rate from years one to five, followed by stable effort thereafter 98 Figure 4.5 Pol icy 4: Effort declines from year one to five, followed by stable effort thereafter 99 V l l L I S T O F A P P E N D I C E S Appendix I Interview questionnaire 126 Appendix II List of commonly caught fish species 135 Appendix III Assumptions for estimates of fishing effort 137 Appendix IV Assumptions for profitability and revenue estimates 138 Appendix V Fish lengths recorded during landings surveys 141 Appendix V I U B C Research Ethics Board Certificate of Approval 142 Appendix VII Diet matrix for Ecopath model 143 vin A C K N O W L E D G E M E N T S I would like to thank my research supervisor, Rashid Sumaila, for all his effort and support for this project. I also thank my supervisory committee - Les Lavkulich, Dirk Zeller, Daniel Pauly, and Armadel Cabanban for their ongoing help, interest, and encouragement. Special thanks to Les Lavkulich, who has been there to answer my questions since the first time I stepped into the R M E S office to enquire about the programme. In Sabah, Annadel Cabanban provided valuable field guidance, support, and enthusiasm to keep the research going. I could not have completed my field work so smoothly without the help of Lydia Teh, my 'sidekick', research assistant, and dive buddy. The Borneo Marine Research Institute Seaweed Project staff at Universiti Sabah Malaysia and in Banggi saved me much trouble by arranging for my boat, house, and other logistical things in Karakit. Lastly, this thesis wouldn't be possible without the generous help of the people I met in Banggi, many of whom have become my friends.. ..maksukul. ix C H A P T E R 1 I N T R O D U C T I O N 1.1 Problem statement The majority of the world's shallow coral reefs are concentrated along the coastlines of developing nations (McManus, 1997). These biodiversity rich ecosystems harbour a wide range of fish and invertebrate species that are caught mostly by small-scale fishers using multiple gears (Munro, 1996). Coral reef fisheries provide a main source of food and income for these small-scale subsistence and artisanal fishers. However, in recent decades, burgeoning populations and poverty in the coastal zones of developing countries, particularly in Southeast Asia, have led to the overexploitation and degradation of many nearshore reef resources and habitats, putting these vulnerable ecosystems under serious threat (McManus, 1997). Yet, small-scale tropical reef fisheries are often marginalized from mainstream policy makers (Pauly, 1997), leading to less than adequate management of these valuable fisheries. The continuation of this trend is likely to lead to depleted reef ecosystems, with dire socio-economic consequences for the communities who depend on them. As such, there is an immediate need to understand the local dynamics of small-scale reef fisheries. Filling in basic data gaps can enable assessments of the state of the studied fisheries, ultimately leading to the identification of management strategies which can ensure the long-term ecological and socio-economic sustainability of reef fisheries. 1.2 Research objectives This research focuses on the small-scale, artisanal coastal reef fisheries of southern Banggi. In this thesis reef fisheries refer to the capture of demersal reef fish, as well as reef-associated species, such as the pelagic scombrids and carangids. This study represents the first attempt of understanding both the ecological and socio-economic dynamics of southern Banggi's reef fisheries.1 The goal of this research is to identify viable management strategies which will lead to the long term sustainability of the reef fisheries and associated socio-economic 1 Certain aspects of Banggi's small-scale artisanal fisheries have been previously documented by Fisher (2000), Daw et ai, (2002); Daw (2004), and Cooke (2003). 1 systems. Insights from this research can also be applied on a broader scale to other reef fisheries in the region, many o f which share similar characteristics and problems as Banggi. The three objectives of this research are: 1) Characterize and describe the ecological, social, and economic aspects of south Banggi's artisanal reef fisheries; 2) Assess the sustainability of south Banggi's reef fisheries; 3) Explore the fishery effects of potential management strategies, and identify a feasible strategy, or strategies, that w i l l contribute to the long-term ecological and socio-economic sustainability o f Banggi's reef fisheries. 1.3 Methods I used both qualitative and quantitative methods in this study to obtain a comprehensive view of Banggi's reef fisheries. Data were collected during two field seasons, which were chosen to correspond to the low and peak fishing seasons. These were June and August, 2004, and March and Apr i l 2005, respectively. The first part of my field work involved creel surveys and fishing ground surveys. The creel surveys were conducted everyday at the two main fish landing sites, during which I recorded the weight, value, and species composition of each landed catch. During each fishing ground survey, I randomly stopped fishers to record gear type, the approximate weight of fish they had caught, the species composition, and the number of hours fished. I also counted the number of fishing boats present at the fishing ground, and recorded their spatial distribution. Techniques similar to those just mentioned had been used to quantify other small-scale reef fisheries (e.g. Alcala and Luchavez, 1981; Amar et al, 1996). Furthermore, I also engaged in participatory observation by joining fishing trips in order to learn more about local fishing practices and strategies. The value of consulting fishers' local ecological knowledge as part of the process in managing marine resources has been increasingly emphasized (Johannes, 1981; Johannes et 2 al, 2000). Subsequently, the second part of the field work involved conducting semi-structured interviews with fishers. During these interviews, I asked questions pertaining to fishers' ecological and fisheries knowledge, as well as perceptions on temporal changes in the size and composition of catches. The quantitative method involved using Ecopath with Ecosim (EwE) ecosystem modelling software as a means of exploring, on a broad scale, the impact different fisheries management strategies might have on the target fish populations, catches, and economic value of Banggi's reef fisheries. Ecopath is a trophic modelling approach based on the assumption of mass balance (Christensen and Pauly, 1992). The theoretical framework for Ecopath was first established by Polovina (1984) and applied to the coral reef ecosystem o f the French Frigate Shoals. Since then, Ecopath has become a well established ecosystem modelling tool that has been applied extensively to various aquatic ecosystems (Christensen and Pauly, 1993). One of the strengths of Ecopath is that it only requires a limited number of biological and fishery input parameters. A mass balanced Ecopath model enables one to derive insights about the trophic relationships of the ecosystem. Previous Ecopath models built for coral reefs include that of Opitz (1996) for the Caribbean, A l ino et al. (1993) for the Philippines, and Arias-Gonzalez et al. (1997) for French Polynesia. More recent uses of E w E for modelling coral reef ecosystems include Tudman (2001), Gribble (2001 and 2003), and Arias-Gonzalez et al. (2004) Ecosim (Walters et al, 1997) uses the results from the static Ecopath procedure, and incorporates principles of predator-prey interactions to enable dynamic simulations, thereby allowing the model to predict temporal responses of ecosystems to exploitation. Ecosim has been used to successfully predict the ecosystem level responses brought about by changes in fishing pressure (e.g., Christensen, 1998). Ecosim has also been used to explore the implications of alternative fisheries management given uncertain ecological dynamics (e.g., Bundy, 1998). Thus Ecosim is a useful tool for investigating the ecosystem effects of 3 different management scenarios; it is particularly relevant for policy makers hoping to adequately manage the fishery induced changes in complex coral reef ecosystems. Ecospace, the spatial component of E w E , is meant to represent how biomasses vary with time as well as space (Walters et al, 1999). It allows users to define (by sketching) grids of spatial cells for exploring the potential ecosystem impacts of Marine Protected Areas (MP As). Ecospace relies on the mass-balanced Ecopath and dynamic Ecosim model for most of its parameters, with additional inputs including movement rates of functional groups between cells, habitat preferences for each functional group, and estimates of trophic interaction (Walters et al, 1999). B y incorporating fishing effect and trophic interaction considerations with spatial movement, Ecospace serves as a useful tool to evaluate the impacts of spatial management policies such as M P As . To date, there are relatively few (compared to Ecopath and Ecosim) publications o f Ecospace application in the literature. Some examples include Okey et al, (2004), Zeller and Reinert (2004), Pitcher et al, (2002), and Tudman (2001). In this thesis, only Ecopath and Ecosim were used for policy exploration. 1.4 Thesis outline This thesis is based on five chapters, three o f which (chapters 2 to 4) consist of manuscripts ready for submission (chapter 3 & 4), with one (chapter 2) successfully submitted (Teh et al., 2005). Chapter 2 provides an ecological and socio-economic characterization of Banggi's reef fisheries, based on information collected during the first field work period. A preliminary assessment o f the state of the island's reef fisheries suggests that they are in a relatively good shape compared to others in the region, but that continued current rates of exploitation w i l l l ikely lead to the depletion of fisheries resources in the future. Chapter 3 outlines the main seasonal characteristics of Banggi's coastal reef fisheries in terms of catch volume, species composition, fishing effort, and economic benefits to fishers. B y doing so, I attempt to explain what, where, when, and why fishers in southern Banggi operate. Information from fisher interviews is used to establish a preliminary trend of the temporal change in catch rates and fishing effort. One of the important findings of this study 4 is that Banggi fishers have witnessed substantial decline in catch rates over the past 20 years, and particularly in the past 5 years. In chapter 4,1 use Ecopath with Ecosim to explore, on a coarse scale, the effects of fishing on south Banggi's coral reef ecosystem. I first speculate on the probable state of the past ecosystem, given a priori assumptions about changes in historical fishing effort and catches. I then explore simulations of four potential management policies. These simulation results are then interpreted within Banggi's socio-economic context, from which I suggest that a community-managed marine reserve might be the best approach for future management of Banggi's reef fisheries. Chapter 5 then summarizes the main findings of the research, and discusses the significance and potential sources of error of this research. It concludes with the recommendation that implementing a small-scale, community-managed marine reserve appears to be the most appropriate course of action for the sustainable management of Banggi's reef fisheries in the immediate future. 1.5 Background and literature review 1.5.1 Reeffisheries Known popularly as "rainforests of the ocean", coral reefs are found mainly in shallow coastal tropical waters, and are one o f the most diverse ecosystems on earth (Bellwood and Hughes, 2001). The large variety of fish and invertebrate species that live in coral reefs and reef associated habitats are the targets of coral reef fisheries. These fisheries are dominated by small-scale subsistence or artisanal fishers, who use gears ranging from hook and line, gillnet, traps, and spears, to destructive techniques such as dynamite, cyanide, and muro-ami (Dalzell, 1996). Reef fisheries are inherently complex as they involve multiple species, gears, landing sites, and uneven effort distribution (Russ, 1991). In the Caribbean, reef fish 2 Muro ami fishing involves a line of divers swimming down a reef while lifting and dropping weighted scare-lines onto the reef in order to drive fish towards a bag net. In the process, corals are broken and damaged. 5 harvests might include around 100 species of fish, while in the Indo-Pacific reefs, 200 to 300 species of fishes may be included (Munro, 1996). Coral reefs are largely concentrated along the coastlines of populous developing nations (McManus, 1997), where reef fisheries often provide the main source of food and income to poor coastal communities (Munro, 1996; McManus , 1997; Burke et al, 2002). Reef fish yields vary due to different intensities of fishing effort, and also different methods used by investigators to calculate the yield (Dalzell, 1996). However, it is generally thought that 2 1 sustainable yields exceeding 5 tkm" yr" are possible (Dalzell, 1996). In the Philippines, small, intensively fished reefs with good coral cover are capable of maintaining sustainable 7 1 reef fish yields ranging from 15 to 20 t-km" -yr" . In contrast, degraded reefs in the Philippines yield only about 4 to 5 t-km^-yr '^Alcala and Russ, 2002). Reef associated fisheries make up approximately 10% of global marine fishery landings (Sadovy, 2005). While the absolute yield from reef fisheries may not count as significant, their relative importance to the subsistence livelihood of coastal communities is enormous (Pauly, 1997; Munro, 1996). This is especially pertinent in Southeast As ia , where coastal communities are heavily dependent on fisheries (Ablan et al, 2002). A s the number of coastal dwellers has steadily risen in these countries, so too has the pressure on reef resources, to the extent that an estimated 80% of coral reefs in Southeast As ia are at risk, with overfishing being one of the most significant threats (Burke et al, 2002). 1.5.2 Ecological and social impacts of reeffishing Fishing on coral reefs directly affects target species, and it also leads to indirect effects on non-target species and reef habitats (Jennings and Lock, 1996). A comprehensive coverage of these ecological effects can be found in Polunin and Roberts (1996). Briefly, fishing on reefs reduces the abundance of species targeted by the fishery, alters the size and age distribution of target fish, and possibly alters the trophic structure of reef communities (Jennings and Lock, 1996). Ultimately, increased fishing intensity can lead to changes in the size and composition of fishery yields (Jennings and Lock, 1996). 6 The beginning phases of a reef fishery usually involves the removal o f top predators such as groupers and sharks (McManus et al, 2000). A s fishing effort progressively intensifies and predatory fish stocks are depleted, the fisheries turn to targeting mainly herbivorous or planktivorous species such as parrotfish and wrasses (McManus et al, 2000). This sequence corresponds to the 'fishing down' phenomenon of Pauly et al. (1998). Overfishing reduces species diversity, and can cause local extinctions of both target and non-targeted species (Roberts, 1995). In recent decades, the relentless expansion of the Live Reef Fish Trade ( L R F T ) has led to serial local depletions of economically valuable, but biologically vulnerable species such as humphead wrasse (Cheilinus undulatus), and coral grouper (Plectropomus spp.) throughout the Indo-Pacific (Sadovy et al., 2003). A t the extreme, overfishing leads to the loss of key functional groups, and this eventually reduces and threatens the resilience of reef ecosystems (Roberts, 1995). Many fishers in overcrowded coastal zones turn to fishing as an occupation of last resort. From a social perspective, overfishing causes hardship for these already poor small scale fishers (Hilborn et al., 2004), who not only have to contend with less food and income, but also have to travel further to look for fish, often in small fishing boats which put them in peril from adverse weather. Furthermore, overfishing results in a net social loss. A study by Cesar et al. (1997) estimated that overfishing in Indonesian coral reefs resulted in net quantifiable losses of U S D $109,000 per k m 2 (Present value using discount rate of 10% over 25 years), while blast fishing resulted in a net loss of U S D $43,000- 476,000 per k m 2 . 1.5.3 Coral reef fisheries management The Great Barrier Reef serves as an example of effective coral reef management (Gomez, 1997), with healthy reef fisheries (McManus et al, 2000). However, the management record for reef associated fisheries in less developed nations, where the majority of coral reefs are concentrated, is dismal (Munro, 1996). Compared to other fisheries, reef fisheries are under-funded and under-monitored, thus poorly understood and largely ignored (Sadovy, 2005). Increased competition for fisheries resources has resulted in Malthusian overfishing (Pauly, 1989) in many Southeast Asian reefs (McManus, 1997), making it unlikely that the region's reef fisheries resources w i l l survive without active management (Ablan et al, 2002). Market 7 drivers exacerbate pressure on vulnerable reef resources. The boom and bust nature of the L R F T (Sadovy et al, 2003) is an example of an unsustainable fishery which destroys the very resources on which it depends. In intensively-fished systems such as the Philippines, coral reef fisheries are believed to have been on the decline since the late 1970s (Alcala and Russ, 2002). Even in places where fishing pressure is relatively lower (e.g., the Caribbean), there are few instances o f effectively managed reef fisheries (Munro, 1996).The lack of management arises from the perception among many policy makers that reef fisheries are not economically valuable enough to warrant the expenses and personnel required for collecting data on often remote and spatially spread out reef fisheries (Russ, 1991). In other cases, although small-scale fisheries are recognized as being essential for food, social, and cultural purposes, they are not recorded comprehensively in official statistics due to a lack of funding and or personnel (e.g., Zeller et al, 2005). Consequently, many small-scale reef fisheries remain data poor (DeVantier et al., 2004), as basic information on landings and fishing effort do not enter official records. The resulting lax management facilitates the continuation of excessive fishing effort and destructive fishing techniques, leading to an increasing and widespread concern about the over-exploitation of coral reef fisheries (Burke et al, 2002; McManus, 1997, Sadovy, 2005). 1.5.4 Marine Protected Areas Marine Protected Areas, or no-take marine reserves, have been advocated as potential solutions to the global problem of overfishing and habitat destruction (Gell and Roberts, 2003) . In theory, marine reserves serve two functions; that is, they provide a buffer against overexploitation, and also potentially maintain or enhance fishery yields through the spillover effect (Sale et al, 2005). However, sound evidence for the largely theoretical benefits of marine reserves remains controversial (Russ, 2002; Gel l and Roberts, 2003; Hilborn et al, 2004) . There is even less empirical data to support the spillover effect, where there is a net emigration of adults from reserves to the surrounding area (but see Abesamis and Russ, 2005) . To my knowledge, only two studies have indicated increased fisheries yield resulting from marine reserve protection; one at Merritt Island, Florida, and the other at Apo Island, 8 Philippines (Roberts et al, 2001; Russ et al, 2003a; Russ et al, 2003b). While marine reserves show great potential as a fisheries management tool, there are also large existing knowledge gaps that must be addressed i f reserves are to be used effectively (Sale et al, 2005). Nonetheless marine reserves are arguably the better option when compared to the alternative of having no management and facing inevitable depletion of reef resources. A reserve by itself does not solve the underlying drivers o f overfishing and destructive fishing. In Malaysia, M P A management has generally followed a top-down approach (Fortes et al, 2002). However, a growing number of studies indicate that a bottom-up, community-based and participatory approach in managing a marine reserve is possibly the best solution for achieving successful coral reef protection (White and Vogt, 2000; Russ and Alcala , 1999).In fact, Jones (2002) recommends a middle ground approach which involves community-based management guided by science, to be the best management approach. This involves a broader scope of activities such as education, community development programmes, and understanding the socio-economic needs of residents (Alder et al, 1994; Sadovy 2005) A s such, the ecological, social, and economic component of coral reef fisheries management are inseparable. 7.5.5 Coral reefs and reef fisheries in Sabah Seventy-five percent of Malaysia's coral reefs occur in Sabah (Burke et al, 2002), with the majority of good reefs concentrated along the northern and eastern coastlines. The reef resources of Sabah are on the decline, as many of the nearshore reefs have been overfished and damaged from destructive fishing practices (Pilcher and Cabanban, 2000; Oakley et al, 1999). Although Sabah's coastal communities have extensively harvested reef fish and invertebrates, the contribution of Sabah's reef fisheries to total marine fish production is unknown (Cabanban and Biusing, 1999). In one study done on Sabah's coral reef fisheries, Cabanban and Biusing (1999) show that coral reef fish families contributed on average 9% (ranging from 7 to 23%) to the total marine fish production of Sabah from 1980 to 1990. However, after experiencing a peak in the mid-1980s, coral reef fish landings declined through to 1993. The current status of reef 9 fish landings remains to be investigated. To my knowledge, there are a limited number of other studies on reef fisheries in Sabah. Mathias and Langham (1978) estimated the contribution of western Sabah's coral reef fishes to total marine catch. Wood (1978) reported destructive fishing methods being used in Sabah reefs. More recently, coastal resource use, including coral reef fisheries, in the Semporna Island group in south east Sabah is documented by Suliansa (2000) and Komilus et al. (2000). In Malaysia, the federal Fisheries Act 1985 governs fisheries management, development, and conservation. The Ac t also covers the establishment of protected areas or marine parks in Malaysian waters, and includes prohibitions on destructive fishing methods (i.e., blast and cyanide fishing). Under Section IV of the Act, it is an offence to fish in any marine park or marine reserve within Malaysian waters. The Fisheries Act also delineates fishing zones according to fishing method, vessel, and engine size. The Sabah Fisheries Department has responsibility for managing and developing fisheries within Sabah waters. This includes monitoring landings, issuing fishing gear licenses, assisting in the distribution of subsidized boats, engines, and fishing gear, as well as research and development activities. The conservation of Sabah coral reefs and resources fall under the mandate of three agencies, namely, the Sabah Department of Fisheries, Sabah Parks, and the Sabah Ministry of Science, Tourism, and Environment. However, these three agencies are limited by a lack of funding and trained personnel to effectively deal with management and conservation issues (Pilcher and Cabanban, 2000). Kudat, the northern most major town in Sabah, is one of the main fish producing districts in Sabah (Biusing, 2001). Being in proximity to extensive coral reefs, it is also one of the most important centres for the Live Reef Fish Trade ( L R F T ) . Daw et al. (2002) provide a comprehensive coverage of the Kudat L R F T . This lucrative trade targets specific demersal reef species, the most important of which is leopard coral grouper (Plectropomus leopardus) in terms of trade volume and value (Daw et al, 2002). Humphead wrasse {Cheilinus undulates) and humpback grouper (Cromileptes altivelis) are the most valuable species by weight; however, these species are rarely caught. 10 Intense development of the live fish industry in Sabah started in 1987 (Bentley, 1999), although Daw et al. (2002) report that the L R F T has been occurring in Kudat since the early 1970s. Live fish exports from Sabah peaked in 1993 at approximately 500 tonnes, but have since declined by over 30% (Bentley, 1999). The main live fish consumer market in Southeast As ia is Singapore, but demand from Kuala Lumpur and other cities with large Chinese populations has also been increasing (Bentley, 1999). The L R F T ' s high economic incentives have led to intense fishing pressure on target species, and the use of cyanide has proliferated as a means o f capturing these fish. The live fish trade is seen as an unsustainable fishery (Sadovy et al., 2003) which jeopardizes the population of species with naturally vulnerable life histories. There are currently few controls of the Kudat live fish trade. A s such, the trade has already taken a toll on the reef ecosystems of northern Sabah, with landings of target species declining 40 to 90% since the late 1990s. (Daw et al, 2002). 1.6 P u l a u B a n g g i Pulau (Island) Banggi is a relatively undeveloped island off the northern tip of Sabah. Banggi is administered as a sub-district of Kudat, which is the nearest mainland town approximately 30 k m across the Banggi Channel. The main island of Banggi is surrounded by approximately 50 smaller islands with a combined area of roughly 270 square miles ( IPMB, 2003). The Banggi Island group is bounded by the South China Sea to the west, and the Sulu Sea to the east. The entire northern region of Sabah, including Pulau Banggi, forms part of the larger Sulu Sulawesi Marine Ecoregion, which extends to the southern Philippines and northern Indonesian islands. This region forms the 'Coral Triangle', which harbours the highest marine biodiversity in the world (Allen and Werner, 2002). Recently, the northern marine and coastal areas of Sabah have been proposed for a marine park (Tun Mustapha Park), which when gazetted, w i l l cover an area of one mill ion hectares. However, the coastal resource uses within the proposed Tun Mustapha Park remains poorly understood. Addressing this knowledge gap for Banggi's reef fisheries is one of the motivations for this research. 11 1.7 References Abesamis, R . A . , Russ, G.R. , 2005. Density-dependent spillover from a marine reserve: long-term evidence. Ecol . App l . 15(5), 1798-1812. Ablan, M . C . A . , McManus, J.W., Chen, C . A . , Shao, K . T . , Be l l , J., Cabanban, A . S . , Tuan, V . S . , Arthana, I.W., 2002. Meso-scale Transboundary Units for the Management of Coral Reefs in the South China Sea Area. Naga, Worldfish Center Quarterly 25, 3&4, 4-8. Alcala , A . C . , Luchavez, T., 1982. Fish yield of the coral reef surrounding A p o Island, Negros Oriental, Central Visayas, Philippines. In: Gomez, E . D . , Birkeland, C .E . , Buddemeier, R .W. , Johannes, R . E . , Marsh, J .A. Jr., Tsuda, R .T . (Eds.), Proceedings of the 4th International Coral Reef Symposium. Marine Science Center, University o f the Philippines, Manila, pp.69-73 Alcala , A . C . , Russ, G.R. , 2002. Status of Philippine Coral Reef Fisheries. Asian Fisheries Science 15, 177-192. Alder, J., Sloan, N . A.,Uktolseya, H . , 1994. A comparison of management planning and implementation in three Indonesian marine protected areas. Ocean and Coastal Management 24, 179-198. Al ino , P . M . , McManus L .T . , McManus, J.W., Nanola, C L . Jr., Fortes, M . D . , Trono, G . C Jr., Jacinton, G.S., 1993. Initial Parameter Estimates of a Coral Reef Flat Ecosystem in Bolinao, Pangasinan, Northwestern Philippines. In: Christensen, V . , Pauly, D . (Eds.), Trophic models of aquatic ecosystems, I C L A R M , Manila , pp. 252-258. Al len , G.R. , Werner, T., 2002. Coral reef fish assessment in the 'coral triangle' of southeastern Asia . Environ. B i o l . Fishes 65, 209-214. 12 Amar, E .C . , Cheong, R . M . T . , Cheong, M . V . T . , 1996. Small-scale fisheries of coral reefs and the need for community-based resource management in Malalison Island, Philippines. Fish. Res. 25, 265-277. Arias-Gonzalez, J.E., Delesalle, B . , Salvat, B . , Galzin, R. , 1997. Trophic functioning of the Tiahura reef sector, Moorea Island, French Polynesia. Coral Reefs 16, 231-246 Arias-Gonzalez, J.E., Nunez-Lara, E . , Gonzalez-Salas, C , Galzin, R., 2004. Trophic models for investigation of fishing effect on coral reef ecosystems. Ecol . Model . 172, 197-212. Bellwood, D.R. , Hughes, T.P., 2001. Regional-Scale Assembly Rules and Biodiversity of Coral Reefs. Science 292 ,5521,1532-1535. Bentley, N . , 1999. Fishing for Solutions: Can the Live Trade in W i l d Groupers and Wrasses from Southeast As ia be Managed? T R A F F I C Southeast Asia , Petaling Jaya, 100 pp. Biusing, R . E . , 2001. Assessment o f coastal fisheries in the Malaysian-Sabah portion of the Sulu-Sulawesi Marine Ecoregion. W W F Malayisa, Kota Kinabalu, 719 pp. Bundy, A . , 2004. The Ecological Effects of Fishing and Implications for Coastal Management in San Miguel Bay, the Philippines. Coast. Manage. 32, 25-38. Burke, L . , Selig, L . , Spalding, M . , 2002. Reef at Risk in Southeast Asia . W R I , Washington. 72 pp. Cabanban, A . S . , Biusing, R . E . , 1999. Coral reef fisheries and their contribution to marine fish production in Sabah, Malaysia. In: Cabanban, A . S . , Phillips, M . (Eds.), Proc. Workshop on Aquaculture of Coral Reef Fishes and Sustainable Reef Fisheries. Institute of Development Studies, Sabah. pp.87-96. 13 Cesar, H . , Lundin, C . G . , Bettencourt, S., Dixon, J., 1997. Indonesian Coral Reefs - A n Economic Analysis of a Precious but Threatened Resource. Ambio 26 (6) 345-360. Christensen, V . , 1998. Fishery-induced changes in a marine ecosystem: insight from models of the G u l f of Thailand. J. Fish B i o l . 53 (Suppl A ) , 128-142. Christensen, V . , Pauly, D . , 1992. E C O P A T H II - a software for balancing steady-state ecosystem models and calculating network characteristics. Ecol Model 61, 169-185. Christensen, V . , Pauly, D . (Eds.), 1993. Trophic models of aquatic ecosystems. I C L A R M Conf. Proc. 26, 1-390. Cooke, F . M . , 2003. L iv ing A t the Top End: Communities and Natural Resource Use in the Kudat/Banggi Region of Northern Sabah. W W F Malaysia, Kota Kinabalu, 45 pp. Dalzel l , P., 1996. Catch rates, selectivity and yields of reef fishing. In: Polunin, N . V . C . , Roberts, C M . (Eds.), Reef Fisheries. Chapman & Hal l , London, U K , pp. 161-192. Daw, T., Jeffrey, L . , B i n A l i , M . A . , 2002. Preliminary Assessment of the Live Reef Fish Trade in the Kudat Region. W W F Malaysia, Kota Kinabalu, 47 pp. Daw, T., 2004. Reef Fish Aggregations in Sabah, East Malaysia. Western Pacific Fisher Survey Series: Volume 5, Society for the Conservation of Reef Fish Aggregations. 59pp. DeVantier, L . , Alcala , A l , Wilkinson, C , 2004. The Sulu-Sulawesi Sea: Environmental and Socioeconomic Status, Future Prognosis and Ameliorative Policy Options. Ambio 33, 88-97. 14 Fisher, FL, 2000. A Socio-Economic Assessment of Coastal Communities of Pulau Banggi Sabah, East Malaysia. Master o f Science Thesis, University of Minnesota. Fortes, M . D . , Djohani, R., Kahn, B . , 2002. The Regional Action Plan to Strengthen a Network of Effective Marine Protected Areas in Southeast Asia : 2002-2012. The World Commission on Protected Areas South East As i a Marine Working Group. 130 pp. Available: http://www.iucn.org/themes/wcpa/pubs/pdfs/marine/raptechnicalbackground.pdf Gel l , F. , Roberts, C M . , 2003. Benefits beyond boundaries: the fishery effects of marine reserves. T R E E 18 (9), 448-455. Gomez, E .D. , 1997. Reef management in developing countries: a case study in the Philippines. Coral Reefs 16, Suppl: S3-S8. Gribble, N . A . 2001. A Model of the Ecosystem, and Associated Penaeid Prawn Community, in the Far Northern Great Barrier Reef. In: Wolanski, E . (Ed.), Oceanographic Processes of Coral Reefs: Physical and Biological Links in the Great Barrier Reef. C R C Press, pp. 189-207. Gribble, N . A . , 2003. GBR-prawn: modelling ecosystem impacts of changes in fisheries management of the commercial prawn (shrimp) trawl fishery in the far northern Great Barrier Reef. Fish. Res. 65, 493-506. Hilborn, R., Stokes, K . , Maguire, J J . , Smith, T., Botsford, L . W . , Mangel, M . , Orensanz, J., Parma, A . , Rice, J., Be l l , J., Cochrane, K . , Garcia, S., Hal l , S., Kirkwood, G.P., Sainsbury, K . , Stefansson, G. , Walters, C , 2004. When can marine reserves improve fisheries management? Ocean Coast. Manage. 47, 197-205. 15 Institut Penyelidikan Mar in Borneo, 2003. Laporan Kajian Mengenai Pengetahuan dan Pandangan Masyarakt Pulau Banggi Terhadap Penubuhan Taman Tun Mustapha. Unpublished report. 36 pp. Jennings, S., Lock, J . M . , 1996. Population and ecosystem effects of reef fishing. In: Polunin, N . V . C . , Roberts, C M . (Eds.), Reef Fisheries. Chapman and Ha l l , London, U K pp. 193-218. Johannes, R . E . , 1981. Words of the Lagoon. University of California Press, Berkely. Johannes, R . E . , Freeman, M . M . R . , Hamilton, R.J . , 2000. Ignore fishers' knowledge and miss the boat. Fish. Fish. 1, 257-271. Jones, P.J.S., 2002. Marine protected area strategies: issues, divergences and the search for middle ground. Reviews in Fish Biology and Fisheries 11, 197-216. Komilus, C.F. , Hassan, T., Shing, W . C , Isnain, I., Martin,B., Laision, D . , Gueh, G . , Biusing, R. , Jintony, B . , 1999. The Semporna Islands Project: The Marine Fisheries Industry. A report for the Semporna Islands Project. Fisheries Research Centre, Sabah. Mathias, J .A. , Langham, N . P . E . , 1978. Coral reefs. In: Chua, T.E. , Mathias, J .A. (Eds.), Coastal Resources of West Sabah. A n Investigation into the Impact of O i l Spills. Universiti Sains Malaysia, Pulau Pinang, pp. 117-151. McManus, J.W., 1997. Tropical marine fisheries and the future of coral reefs: a brief review with emphasis on Southeast Asia . Coral Reefs 16, Suppl: S121-S127. McManus, J.W., Menez, L . A . B . , Kesner-Reyes, K . N . , Vergara, S.G., Ablan, S . M . C , 2000. Coral reef fishing and coral-algal phase shifts: implications for global reef status. ICES J. Mar. Sci . 57, 572-578. 16 Munro, J .L. , 1996. The scope of tropical reef fisheries and their management. In: Polunin, N . V . C . , Roberts, C M . (Eds.), Reef Fisheries. Chapman and Ha l l , London, U K , pp. 1-14. Okey, T. A . , Banks, S., Born, A . F . , Bustamante, R . H . , Calvopina, M . , Edgar, G.J . , Espinoza, E . , Farina, J . M . , Garske, L . E . , Reck, G . K . , Salazar, S., Shepherd, S., Toral-Granda, V . , Wallem, P., 2004. A trophic model of a Galapagos subtidal rocky reef for evaluating fisheries and conservation strategies. Ecol . Model . 172, 383-401. Oakley, S., Pilcher, N . , Atack, K . , Digges, C , Enderby, S., Mackey, G . , Clubb, R., Stapelton, K . , Toh, S . M . , Huet, C , Morton, T., 1999. Reefs under attack: the status of coral reefs of Sabah, East Malaysia. Presented In: 4 t h International Conference on the Marine Biology of the South China Sea, Quezon City, 20-22 October, 1999. Opitz, S., 1996. Trophic Interactions in Caribbean Coral Reefs. I C L A R M Tech. V o l . 43 341 pp. Pauly, D . , Silvestre, G. , Smith, I.R., 1989. On development, fisheries, and dynamite: a brief review of tropical fisheries management. Nat. Resour. Mode l 3 3, 307-329. Pauly, D . , Christensen, V . , Dalsgaard, J., Froese, R., Torres, F. Jr., 1998. Fishing down marine food webs. Science 279, 860-863. Pauly, D . , Christensen, V . , Walters, C , 2000. Ecopath, Ecosim, and Ecospace as tools for evaluating ecosystem impact of fisheries. ICES J.Mar. Sci. 57 (3), 697-706. Pauly, D . , 1997. Small-Scale Fisheries in the Tropics: Marginality, Marginalization, and Some Implicatins for Fisheries Management. In: Pikitch, E . K . , Huppert, D .D . , Sissenwine, M . P . (Eds.), Global Trends: Fisheries Management. American Fisheries Society Symposium 20, Bethesda, Maryland, p.40-49. 17 Pilcher, N . , Cabanban, A . S . , 2000. The Status of Coral Reefs in Eastern Malaysia Global Coral Reef Monitoring Network ( G C R M N ) Report. Australia Institute of Marine Science, Townsville. Pitcher, T J . , Buchary, E . A . , Hutton, T., 2002. Forecasting the benefits of no-take human-made reefs using spatial ecosystem simulation. ICES J. Mar. Sci . 59, 17-26. Pitcher, T., Buchary, E . , Trujillo, P., 2002. Spatial Simulations of Hong Kong's Marine Ecosystem: Ecological and Economic Forecasting of Marine Protected Areas With Human-Made Reefs. Fisheries Centre Research Reports V o l . 10, University of British Columbia, Vancouver, pp. 168 Polovina, J.J., 1984. Model of a coral reef ecosystem I. The E C O P A T H model and its application to French Frigate Shoals. Coral Reefs 3, 1-11. Polunin, N . V . C . , Roberts, C M . (Eds.), Reef Fisheries. Chapman & Ha l l , London, U K . Roberts, C M . , 1995.Effects of fishing on the ecosystem structure of coral reefs. Conserv. B i o l . 9 (5), 988-995. Roberts, C M . , Bohnsack, J .A. , Gel l , F. , Hawkins, J.P., Goodridge, R., 2001. Effects of marine reserves on adjacent fisheries. Science 294, 1920-1923. Russ, G.R. , 1991. Coral Reef Fisheries: Effects and Yields. In: Sale, P.F. (Ed.), The Ecology of Fishes on Coral Reefs. Academic Press, California, pp. 601-635. Russ, G.R. , 2002. Yet another review of marine reserves as reef fisheries management tools. In: Sale, P.F. (Ed.), Coral reef fishes: dynamics and diversity in a complex ecosystem, Academic Press, San Diego, C a l i f , pp. 421 - 443. 18 Russ, G.R. , Alcala, A . C . , 1999. Management histories of Sumilon and Apo Marine Reserves, Philippines, and their influence on nacional marine resource policy. Coral reefs 18,307-319. Russ, G.R. , Alcala, A . C . , Maypa, A . P . , 2003a. Spillover from marine reserves: the case of Naso vlamingii at Apo Island, the Philippines. Mar. Ecol . Prog. Ser. 264 (15), 15-20. Russ, G.R. , Alcala, A . C . , Maypa, A . P . , Calumpong, H.P. , White, A . T . , 2003b. Marine Reserve Benefits Local Fisheries. Ecol . App l . 14 (2), 597-606. Sale, P.F., Robert, C . K . , Danilowicz, B .S . , Jones, G.P., Kritzer, J.P., Lindeman, K . C . , Planes, S., Polunin, N . V . C . , Russ, G.R. , Sadovy, Y . J . , Steneck, R.S. , 2005. Critical science gaps impede use of no-take fishery reserves. T R E E 20 (2), 74-80. Sadovy, Y . J . , Donaldson, T.J. , Graham, T.R., McGilvray , F. , Muldoon, G.J . , Phillips, M . J . , Rimmer, M . A . , Smith, A . , Yeeting, B . , 2003. While Stocks Last: The Live Reef Fish Trade. Asian Development Bank, Manila , 147 pp. Sadovy, Y . , 2005. Trouble on the reef: the imperative for managing vulnerable and valuable fisheries. Fish Fish 6, 167-185. Suliansa, M . S . , 1999. Survey on the use of marine resources at the Semporna Islands. A report for the Semporna Islands Project. W W F Malaysia. Teh, L . , Cabanban, A . S . , Sumaila, U .R . , 2005. The reef fisheries of Pulau Banggi, Sabah: A preliminary profile and assessment of ecological and socio-economic sustainability. Fish. Res. 76, 359-367. Tudman, P .D. , 2001. Modell ing the trophic effects of fishing on a mid-shelf coral reef of the central Great Barrier Reef. BSc . Honours thesis, James Cook University, Townsville. i 19 Walters, C , Christensen, V . , Pauly, D . , 1997. Structuring dynamic models of exploited ecosystems from trophic mass-balance assessements. Rev. Fish. B i o l . Fish. 7, 139-172. Walters, C , Pauly, D . , and Christensen, V . 1999. Ecospace: Prediction of mesoscale spatial patterns in trophic relationships of exploited ecosystems, with emphasis on the impacts of marine protected areas. Ecosystems 2 (6), 539-554. White, A . T . , Vogt, H.P. , 2000. Philippine Coral Reefs Under Threat: Lessons Learned After 25 Years o f Community-Based Reef Conservation. Mar. Pollut. B u l l . 40 (6) 537-550. Wood, E . M . , 1978. Coral reefs in Sabah: actual damage and potential dangers. Malay. Nat. J. 31 (1), 49-57. Zeller D. , Reinert J., 2004. Modell ing spatial closures and effort restrictions in the Faroe Islands marine ecosystem. Ecol . Model . 172, 403-420. Zeller, D . , Booth, S., Pauly, D . , 2005. Fisheries contributions to G D P Underestimating small-scale fisheries in the Pacific. Fisheries Centre Working Paper #2005-05, The University of British Columbia, Vancouver, B . C . , Canada. 20 C H A P T E R 2 T H E R E E F F I S H E R I E S O F P U L A U B A N G G I , S A B A H : A P R E L I M I N A R Y P R O F I L E A N D A S S E S S M E N T O F E C O L O G I C A L A N D S O C I O - E C O N O M I C S U S T A I N A B I L I T Y 1 2.1. Introduction Most of the world's coral reefs occur in developing countries where the role of reef fisheries is crucial in providing food, income, and livelihood to millions of coastal inhabitants (Munro, 1996). Increased competition for fisheries resources has resulted in overfishing and destructive fishing methods in many Southeast Asian reefs (McManus, 1997), making it unlikely that the region's reef fisheries resources will survive without active management (Ablan et al, 2002). However, reef fisheries are generally poorly understood (Sadovy, 2005), thus hampering their sustainable management. Seventy-five percent of Malaysia's coral reefs occur in Sabah (Burke et ah, 2002), with the majority of good reefs concentrated along the northern and eastern coastlines. Destructive fishing practices have been used on these reefs since the 1970s (Wood, 1978; Pilcher and Cabanban, 2000). As a result, many of the near shore reefs have been damaged (Oakley et al, 1999), contributing to the decline in reef resources. Although Sabah's coastal communities have extensively harvested reef fish and invertebrates, the contribution of reef fish to Sabah's total marine fish production is unknown (Cabanban and Biusing, 1999). In one study, Cabanban and Biusing (1999) showed that coral reef fish families contributed on average 9% (ranging from 7 to 23%) to the total marine fish production of Sabah from 1980 to 1990. However, after experiencing a peak in the mid 1980s, coral reef fish landings declined through to 1993, and no further studies have been done to investigate this trend. In 2001 Sabah Parks, the State agency responsible for the management of marine and terrestrial parks in Sabah, proposed the establishment of Tun Mustapha Park (TMP) in north 1 A version of this chapter has been published. Teh, L . , Cabanban, A.S., Sumaila, U.R., 2005. The reef fisheries of Pulau Banggi, Sabah: A preliminary profile and assessment of ecological and socio-economic sustainability. Fish. Res. 76, 359-367. 21 Sabah. T M P , which encompasses Pulau Banggi, has the aim of alleviating overexploitation of the region's fisheries and conserving the rich biodiversity found within its coastal environment. When gazetted, the T M P w i l l cover one mil l ion hectares, and form part of the wider Sulu-Sulawesi Marine Ecoregion (SSME) , which spans across the southern Philippines, Sulawesi Island in Indonesia, and Sabah. The Sabah State Government's approval of T M P in 2003 raised the need for understanding coastal resource uses within T M P . 2 1 address this need by first providing an ecological and socio-economic characterization of Banggi's reef fisheries, based on fishery surveys and interviews carried out on site in June and August 2004.1 then do a preliminary assessment of the state of the island's reef fisheries, with the aim of contributing towards the identification of appropriate policies for the sustainable management of reef fisheries within the T M P . Insights from this study can also be applied on a broader scale to other reef fisheries in the region, many of which share similar characteristics and problems as Banggi. 2.2 Background 2.2.1 Site description Pulau 3 Banggi (7.14'N.,117.10'E) is situated off the northern tip of Sabah, East Malaysia (Fig.2.1), and is bounded by the South China Sea to the west, and the Sulu Sea to the east. Banggi is the biggest island in Malaysia. The main island is surrounded by around 50 small outlying islands, covering a total area of approximately 700 km , with a coastline of 420 km (Anon., 2003). Shallow fringing reefs border the coast of the Banggi islands, while deeper patch reefs of up to 30 m depth are found further offshore. The nearest mainland town is Kudat, which is approximately 30 km across the Banggi Channel. Banggi is administered as a sub district of Kudat. Karakit serves as the administrative centre and entry point into Banggi, and was the research base for this study. M y study site covers southern Banggi, specifically the villages which have easy, daily access to the Maliangin fishing grounds. 2 Although approved, the The Tun Mustapha Park has not yet been officially gazetted as of July 2005. It still has to go through a parliamentary process and before receiving final gazettement status from the Sabah Chief Minister. 3 Pulau is the Malay word for island. In this paper I will refer to all islands simply by name, without the Pulau prefix. 22 L Figure 2.1 Map of Sabah with Pulau Banggi situated off the North coast. 2.2.2 Coastal fishing community profile The current population of Banggi is approximately 18,000 (Biusing, 2001) to 20,000 (IPMB, 2003), with close to 1100 inhabitants in the vicinity of the study site. The majority of Banggi's coastal communities were first settled after World War II by migrants from the southern Philippines. These new settlers mainly derived their subsistence from the sea, although some were involved in small scale agriculture and coconut growing. In the past two decades, a large influx of migrants from the Philippines and neighbouring islands have come to Banggi in search for a marginally better life. Nevertheless, Banggi remains relatively undeveloped, and coastal households are considerably below the Sabah poverty line (IPMB, 2003). These communities continue to depend heavily on marine resources for their livelihood, with fishing accounting for 70% of the island's economic activity (IPMB, 2003). 23 There are 1195 recorded licensed fishers on the entire island of Banggi (Kudat Fisheries Department, unpublished data). Due to the lack of Fisheries Department personnel presence on Banggi, many local fishers do not own licenses; hence this number should only be taken as a lower reference point. There are approximately 90 fishers in the southern Banggi villages of Singgahmata, Perpaduan, Maliangin Besar, Patanunam, and Karakit. The majority of these fishers belong to the Ubian or Suluk ethnic groups. 2.3. Methods Field work to collect primary data was conducted over six weeks in June and August 2004, which corresponded to the windy southwest monsoon season. Fishery data was collected for the hook and line and gillnet fisheries, which accounted for the majority of fish captured for sale in Banggi . 4 Methods used to collect these data included fishing ground surveys and monitoring catch landings. Surveys of the Maliangin fishing grounds were conducted three times a week when weather permitted, between 7:30 to 11:00 a.m., when it was most likely that fishers would have caught some fish, and before they started to head back to land. Fishers were randomly stopped and the following was recorded: a) gear type; b) quantity o f fish caught (weight was estimated by my boatman, who was also an experienced fisher); c) number of hours fished; and d) species caught. Fish landings were recorded at the two main fish buyers' sites at Karakit and Singgahmata from June 16 t h to July 6 t h , and from August 7 t h to August 27 t h , 2004. This data collection period corresponded to the low fishing season, and was considered representative of the catch and effort for that particular time period as no extraordinary events occurred to indicate otherwise. Gillnet fishers landed their catch at the Karakit site between 8:30 to 11:00 p.m. I waited at the landing site during those times and recorded all the catches that were landed during that time frame. Hook and line catches were landed more sporadically at the Singgahmata site. Therefore sampling at this site was done opportunistically. The majority of hook and line catches were sampled during the fishing grounds survey. 4 Based on information from the two main fish buyers in southern Banggi. 24 Interviews were conducted in villages where fishers were known to frequent the Maliangin fishing ground. These interviews were semi-structured, and were carried out with 20 fishers, 2 fish traders, and 3 village leaders in the villages of Karakit, Perpaduan, Singgahmata, Maliangin Besar, and Lok Tohog (Fig.2.2). Two expert interviews were also conducted with N \J i < Figure 2.2 M a p o f v i l l a g e s i n S o u t h e r n B a n g g i . a fisheries officer in the Sabah Department of Fisheries in Kota Kinabalu, and a fisheries officer in the Kudat Fisheries office. M y assistant and I conducted all the interviews in Malay. Prepared questions pertaining to catch, effort, target species, historical trends, and perceptions on the future of fishing in Banggi were asked (See Appendix I). Fishers were given a lot of breadth in answering the questions, and those who brought up new topics or elaborated on certain questions were allowed to do so. As such, not all interviews followed the same questionnaire format. Depending on fishers' willingness to continue, interviews lasted between 15 minutes to one hour. 25 Fishers were chosen opportunistically; my assistant and I went house to house and interviewed any fisher who was at home and wi l l ing to speak to us. We did not encounter any refusals, and in general all interviewees were cooperative in answering the questions. Additional information was gathered through direct observation and participation in fishing trips and reef gleaning activities. I also carried out informal interviews whenever the opportunity arose to speak to fishers or villagers about reef resource use. Descriptive statistics of fishery data was calculated using Microsoft Excel and SPSS 12.0 software. 2.4. Results 2.4.1 Profile of Banggi's reeffisheries The reef fisheries of Pulau Banggi can be classified as artisanal, which for the purpose of this paper is characterized as being small scale; using low technology fishing gear and small fishing vessels; short fishing trips; and fishing grounds close to shore. Banggi's reef fisheries can also be considered open access in that there are no active spatial, input, or output controls. The only regulation pertains to a ban on bomb fishing (Fisheries Regulation Prohibition of Methods of Fishing, 1980), although enforcement of this law by the Malaysian Marine Police is not entirely effective.5 Fishing gear Two of the most important fishing methods in Banggi are hook and line and gillnet (Cooke, 2003), and most fishers specialize in using either one. Interviews with the two fish traders in Karakit and Singgahmata indicated that hook and line and gillnet account for the majority of fish captured for sale; therefore these were the two gears focused on for fish catch data collection and analysis. Hook and line consists of a simple handline with single baited hook. Alternatively, a series of 8-10 cloth lures are attached to the handline via small hooks. This is then used for trolling for pelagic fish, or constantly jerked up and down from a stationary point. Gillnet fishers use monofilament nets that range from roughly 150 m to 200 m in 5 1 heard 4 bombs go off in a span of three days. As well, a non-governmental group doing underwater visual surveys in the area also reported hearing frequent blasts. 26 length, and 1.5 m in height. These nets are made by joining together units o f a net of approximately 12 m by 20 m in dimension. Net mesh sizes vary from roughly 6.4 cm to 12.7 cm. Some hook and line fishers also use traps, engage in spear fishing (either free diving or with compressor) or squid jigging. Traps are set one to two times a week, although this was not readily observed. Reef gleaning for invertebrates is carried out on the exposed reef flat in front of the villages during spring tides. Blast fishing still occurs on a fairly regular basis. This is used primarily to obtain bait fish for hook and line fishing; some fishers also resort to blast fishing when a large quantity of fish is needed, such as during wedding or funeral ceremonies. Bombs were observed being thrown at reefs not more than 10 feet deep. Cyanide is used to capture live fish for the live fish trade. Longlines with 50-100 baited hooks are used by outside fishers who come to fish in the Maliangin area on multi-day (3 or 4 days) fishing trips. These outsiders come from Kudat, and from as far away as Semporna, located in south east Sabah approximately 340 km away. Fishing vessels The majority of local artisanal fishers use pump boats (pumbot), wooden boats that are powered by a modified water pump engine that is 7 or 8 horsepower. Pumpboats were introduced from the Philippines in the early 1990s, and became popular for their low fuel consumption. Pump boats range from approximately 2 m for single person boats to 7 m. Most have a small holding pond in the front section of the boat, where valuable fish species are kept alive for the live fish trade. Fibre glass boats with outboard engines are used by a small number of fishers. These motorboats are given to selected fishers as part of a poverty alleviation scheme administered by the Sabah Fishery Development and Fishers Corporation since 1996. Fishing grounds The Malaysian Fisheries Comprehensive Licensing Pol icy (FCLP) introduced in 1981 reserves the area within five nautical miles of the shoreline as a traditional fishing zone, where only traditional fishing gears are allowed. Therefore there is no commercial fishery 27 operating within the same fishing grounds frequented by south Banggi's artisanal fishers. Although more commercially-oriented operations such as the compressor diving with spear gun fishery occur within the vicinity of Banggi, their larger scale excludes them from the scope of this study. Information pertaining to these operations can be found in Daw et al, (2002) The main fishing ground studied is the Maliangin area, immediately south of Banggi (Fig.2.3). This is one of three main fishing areas in Banggi, the others being Kuambang and Sibogo in the east, and Balambangan in the west. Although fishing practices tend to vary by village, the general trends observed in the Maliangin area are common to other parts of Banggi, and can be used to broadly characterize the entire island's reef fisheries. Figure 2.3 Map of Maliangin fishing ground, showing spatial distribution of effort by gear and fishery type There is a strong spatial distinction according to gear type (Fig.2.3). Hook and line fishers fish during the day in the Maliangin area, which covers an area of roughly 17km , and is characterized by shallow reefs 3 to 5 m in depth near shore, and deeper patch reefs o f 28 15-30 m depth further offshore. Many fishers tend to aggregate at a deep reef on the southwestern side of Maliangin K e c i l , where they fish for yellowtail scads, Spanish mackerel, and coral groupers. The various actively fished spots cover a total area o f approximately 8 to 9 k m 2 . 6 Gillneters fish at more sheltered locations, and normally set their nets in the evening or early morning around Balak Balak, Kaligau, Patanunam, and the northern side of Maliangin Besar (Fig. 2.3). These fishing locations cover an area o f approximately 6 k m 2 . 6 They sometimes also travel to Lok Tohog on the west of Banggi, and to Kuambang in the north east. Further fishing grounds include Pulau Malawali , (approximately 1 to 1.5 hours away), and Sibogo, which is the furthest fishing location and about 2 hours from Karakit by pump boat. Target species Hook and line fishers rotate between two target species in a year. From A p r i l to September they use baited hooks or traps to target coral grouper (Plectropomus spp.) for the live reef fish trade (LRFT) . They then switch to Spanish mackerel (Scomberomorous commerson), which is caught by trolling or using a floating line with live bait (usually Rastrelliger kanagurta) from October to February or March. Starting in November until February, many fishers also target cuttlefish, which is caught with a trident like spear. Other target species include invertebrates such as scallops (Pectinidae), abalone (Haliotidae), sea cucumber (Holothuroidea), and squid (Cephalopoda). Collection of invertebrates intensifies during the S W monsoon, when sea water temperature rises slightly, thus facilitating diving by fishers. Fishers also go spear fishing more during the S W monsoon period. Hand jigging and trolling are the fishing techniques of choice when conditions are windy, and drop lining is done when seas are calmer. On bright nights when there is a full or partial moon, hook and line fishers go night fishing at the Maliangin fishing grounds. On moonless or dark nights and during windy conditions, fishers from the villages of Patanunam, Singgahmata, and Perpaduan fish for oxeye scad 6 Hook and line and gillnet fishing areas estimated from map of Karakit (Series T735 Sheet 7/117/13), published by the Director of National Mapping, Malaysia 1984. 29 (Selar boops) using hand lines with multiple fabric lures. This takes place off the northern and southern points of Patanunam Island, which is located across Karakit. In contrast, there is no seasonal switch in fishing technique or species observed in the gillnet fishery. Instead, gillnet fishers might change the type o f net they use according to the severity of wind. Larger mesh size nets are used during calm weather, while smaller mesh size nets are used during windy conditions because they do not tangle as easily as large meshed nets. However, both net sizes capture the same type of fish. Baitfish Hook and line fishers use squid or small species of fish for bait, usually whiptail breams (Pentapodus spp.) or fusiliers (Pterocaesio spp.). However, the preferred bait fish for targeting coral grouper are clupeids (tambari), while Indian mackerel (Rastrelliger kanagrutd) or yellowtail scad (Atule mate) is used for Spanish mackerel. Fishers normally start off their fishing trip by jigging with feathered lures to catch some bait fish first. Alternatively, squid may be used, and this is caught by squid jigging the previous evening or obtained from bigger fishing vessels that are anchored within the Maliangin fishing grounds. A n unsustainable way of catching bait fish is by blast fishing. Among a certain group of fishers in southern Banggi, this practice seems more prevalent during the S W monsoon season (June-September), as they need the bait for fishing coral grouper. Clearer water conditions during the S W monsoon might also facilitate blast fishing as fishers can spot aggregations of fish more easily. Markets and prices Banggi's fishers provide a supply of fresh fish to the domestic village market. After setting aside fish for family consumption (usually around 2.0 to 2.5 kg per day), fishers sell their catch to one of three local fish buyers who then sell to the villagers. These fish traders transport their excess or more expensive fish to Kudat, where they have established permanent relationships with wholesale fish traders. L R F T species caught in the Maliangin area are kept in a holding pen in Perpaduan village. Daw et al. (2002) provide a detailed study of the Kudat/Banggi L R F T . 30 The price that a fisher receives for his catch varies according to the overall supply of fish in the market, and also according to prices set by wholesalers in Kudat (Cooke, 2003). In general, demersal reef fish are cheap, with most species fetching only 1.5 to 2.0 R M kg" 1 . 7 Spanish mackerel and trevallies are higher priced at 4 to 6 R M kg" 1, and 4.0 to 5.5 R M kg" 1, respectively, while cuttlefish fetches 4.5 R M kg" 1. In contrast, live coral groupers (Plectropomus leopardus) are worth 42 R M kg" 1, with the price of less expensive Plectropomus species ranging from 10 to 30 R M kg" 1. A list o f recorded fish prices can be found in chapter 3, Table 3.5. 2.4.2 Recorded fishery data Catches Recorded catches of both hook and line and gillnet gears were characterized by high daily variability, regardless of fishing season. It should be noted that my recorded landings only captured successful fishing trips, and did not include the trips when fishers got no catch. It also did not capture the portion of fishers who fished near shore for food fish. Approximately 1.5 to 2 kg can be added for every catch that was sold to account for the fish kept by fishers for personal consumption. Thus results presented here can be considered a conservative estimate of total catches as well as of fishing effort (number of fishers). Hook and line landings fluctuated from 0.4 kg to 30 kg fisher"1 trip"1, with a mean of 7.1 ±7.6 kg-fisher'-trip (Table 2.1). Estimated catch per unit effort (CPUE) was 1.43 kg-fisherhr"1 (Table 2.1). Gillnet landings varied from 1.5 kg to 250 kg, with a mean of 14.8±12.7 kg-fisher"1trip (Table 2.1). This corresponded to an estimated C P U E of 3.85 kg-fisherhr"1 (Table 2.2). Table 2.1 Mean CPUE and catch recorded during the sample period June to August 2004. Gear Mean C P U E (kg h r 1 ) Standard Mean catch per Standard No. of deviation trip (kg trip') deviation samples Hook and Line 1.43 1.51 7.13 7.57 69 Gillnet 3.85 2.94 14.79 12.70 80 7 Currency exchange rate on Dec.19 2005 was 1USD = 3.78 R M 31 The C P U E for both fisheries were substantially higher when compared to the C P U E of two other comparable reef fisheries, one in south east Sabah and the other in the Philippines (Table 2.2). Although not recorded, interviews with fishers revealed that night time hook andline fishing yielded particularly large catches ranging from 5 to 30 kg-fisher"'-trip"1. Landings were not recorded for other gears; however during fishing ground surveys the highest catches were observed using long line, which ranged from 5 kg to 120 kg (n=7). Traps yielded on average 3 kg to 4 kg of fish from being submerged for three days. Blast fishing yielded about 10 kg of fish (mostly fusiliers, Pterocaesio spp.), providing two days supply of bait for three fishers. Table 2.2 Comparison of mean C P U E (kgfisherhr') by site. Gear South Banggi Semporna* Malalison Island** Hook and line 1.43 0.38 0.67 Gillnet (set) 3.85 0.63 0.43 Source: Semporna Islands Project (http://www.mcsuk.org/sempoma/resource/19.htm); **Source: Amar et al. (1996). Catch composition The multi-species nature of Banggi's reef fisheries was evident in the variety of fish present in the catch. Eighty and eighty-one identifiable species were recorded in the hook and line and gillnet catches respectively. Catch composition for the hook and line fishery was recorded according to frequency of occurrence. The most common families caught were Serranidae (36%), Lethrinidae (17%), and Nemipteridae (10%). O f the serranids caught, 10% were of the valuable Plectropomus spp. Besides demersal reef fish (which made up 71% of the catch altogether), pelagic fish such as carangids and scombrids were also regularly caught by hook and line (Fig.2.4), and accounted for 6% and 5% of the catch, respectively. Gillnet catch composition was recorded by weight. Carangidae was the dominant family (57%) in gillnet catches, with yellowtail scad (Atule mate) and trevallies accounting for 24 and 23% of these carangids, respectively. Demersal reef fish belonging to the Haemulidae, 32 40% O "D g.? CO (fl "O CD CD Q. ' E X L 1 » I 5 85 • o E 1 1 8 ui '5) = f? g (5 s y. Figure 2.4 Composition of hook and line catch by frequency of occurrence and family. Lethrinidae, and Lutjanidae families made up another 32% of the catch (Fig.2.5). A list o f commonly caught species by both gears can be found in Appendix II. 60% 50% 40% 30% 20% 10% 0% Demersal reef fish Trevallies/Scads Tunas and Mackerels Others (Carangidae) (Scombridae) Figure 2.5 Composition of gillnet catch by weight and family. Due to the limited times other gears were observed and recorded, their catch composition can only be described qualitatively in this paper. Long line catches consisted of large sized coral groupers (Plectropomus spp.), aerolate grouper (Epinephelus aerolatus), pink ear emperor (Lethrinus lentjan), and Spanish flag snapper (Lutjanus carponotatus). A variety of reef fish were caught in traps, with sweetlips (Haemulidae), fusiliers (Caesionidae), and unicornfish 33 (Naso spp.) named as the most commonly trapped fish. For the spear fishery, only unicornfish, ribboned sweetlips (Plectorhinchus polytaenia), blue spotted ray (Taeniura lymma), and lobster (Panulirus spp.) were observed among the species caught. Fishing effort There were approximately 90 fishers in the villages within the immediate vicinity of the Maliangin fishing grounds (that is, the villages of Maliangin Besar, Singgahmata, Karakit, Patanunam, and Perpaduan). Some fishers from further villages such as Lumais, Kobong, and Kaligau (Fig.2.2) also fished in the Maliangin fishing grounds. If these fishers were also taken into account, the number of fishers increased to around 120. A n average of 11 (ranging from 8 to 23) hook and line fishing boats originating mainly from Singgahmata, Maliangin Besar, Perpaduan, Lok Tohog, and Lumais were recorded at the Maliangin fishing ground. These observations took place on days of relatively calm weather interspersed among days of rough weather. Some hook and line fishers diverted effort towards night time fishing once or twice a week during bright phases of the moon. The variability of fishing effort was largely determined by the severity of wind. During windy periods hook and line fishers would make 3 to 4 fishing trips per week; the rest of the time they opted to stay at home, or would just fish near shore for subsistence needs. Fishing trips during these windy months usually lasted 6 hours, from 6 or 7 a.m. to noon or 1 p.m., although some trips may have ended prematurely due to extreme weather conditions. During calm weather, fishers said that they fish everyday and for a longer time, returning to fish in the afternoons from around 2 to 5 p.m. Based on this information, hook and line fishers fish roughly 216 to 269 days per year (Appendix III). Due to the relatively protected nature of their fishing grounds, gillnet fishers were able to go out fishing more often during the windy season, usually 4 to 5 fishing trips per week during windy periods and everyday during calm weather. Gillnet fishers fish an estimated 256 to 309 days a year (Appendix III). When weather is favourable, gillnet fishers tend to go fishing twice a day, once in the evening around 6 to 8 p.m., and once again in the early morning from roughly 4 to 6 a.m. The soak time for nets was around 2 to 3 hours. There were an 34 estimated 15 regular gillnet fishers operating in the south Banggi area, the majority of whom resided in the villages of Kaligau and Batu Layar. Fishers from other parts of Sabah made multi-day fishing trips to the Maliangin area as well , some of them coming all the way from Semporna, approximately 340 km away. During fishing ground surveys, up to three outside fishing boats were observed in the span of one week. Catch revenue and income The average daily revenue for hook and line and gillnet fishers was calculated based on their daily catch, and is presented in Table 2.3. These estimates show that hook and line fishers are substantially worse off because they were more susceptible to lost fishing days as a result of adverse weather conditions. However, hook and line requires less capital investment in gear than gillnet. Therefore, there may not be a significant difference in net income i f fishing costs were deducted. This analysis w i l l be done in chapter 3. Table 2.3 Estimated average fishing revenue (in RM) during the low fishing season (June-Sept 2004). Gear Mean daily revenue (RM) Average # of fishing days per week Estimated mean monthly revenue (RM) Hook and line 20.20 3.5 247 Gillnet 33.60 5 672 With the exception of one case, all sampled hook and line fishers who attained daily revenue higher than the mean of R M 20.20 had caught a coral grouper. Coral grouper was caught in approximately 30% of all recorded hook and line catches. The average revenue obtained from the sale of coral grouper was R M 44.46, and this accounted for 43 to 100% of the revenue earned by these fishers (Fig.2.6). This economic reliance on coral grouper is a cause for concern, as its life history makes it intrinsically vulnerable to overexploitation. Thus, the depletion of the coral grouper stock might lead to economic hardship for those hook and line fishers whose sole income source is from fishing. For the gillnet fishery, the largest contribution to revenue was derived from the sale of trevallies, which made up 38% of total revenue. 35 160 • Other reef fish El Coral grouper | i 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Fisher Figure 2.6 Breakdown of daily revenue from hook and line of 19 fishers, showing the proportion due to the capture of coral grouper and of other reef fish, in June and August 2004. Fish lengths N o fish length measurements were carried out this field season in 2004. 8 However, based on observations, serranids, lutjanids, lethrinids, nemipterids, and caesionids caught by hook and line generally ranged from roughly 15 to 25 cm. Epinephelus fasciatus, the most commonly caught species, also seemed to be the smallest fish, with observed lengths ranging from around 15 to 20 cm. Other commonly caught hook and line species such as Lethrinus ornatus, Lethrinus lentjan, Epinephelus aerolatus, Cephalopholis miniata, and Lutjanus vitta were approximately 20 to 25 cm in size. For the gillnet catches, most of the trevallies (Carangidae) were not smaller than approximately 25 cm. Sweetlips (Plectorhinchus picus and Plectorhinchus pictus) were normally around 30 cm long. Two size ranges of Lutjanus fulviflamma, Lutjanus vitta, and Lethrinus lentjan were commonly observed. Smaller sizes ranged from around 10 to 15 cm in length, while larger sizes ranged from approximately 15 to 25 cm. The Karakit fish trader differentiated between two size categories of demersal reef 8 Fish length measurements were carried out in the second field season in 2005 (not covered in this chapter), and are presented in Appendix V. 140 S 120 100 c CD I 80 re o re a 60 40 20 36 fish. Reef fish that were considered 'small ' were less than around 20 cm in length, whereas 'b ig ' reef fish were those above 20 cm. 2.5 Discussion The emerging picture of Banggi's reef fisheries is that it has not yet reached the depleted state characterized by other Southeast Asian fisheries (e.g., Talaue-McManus and Kesner, 1995). C P U E estimates obtained in this study are substantially higher than those documented for two other heavily fished small scale reef fisheries in south east Sabah and the Philippines. The catch composition for both hook and line and gillnet is still dominated by high trophic predatory fish. According to McManus et al. (2000), this type of species composition is characteristic of reef fishing that is considered in a 'manageable stage'. Subsequently, it also suggests that the "fishing down marine food webs" (Pauly et al, 1998) phenomenon occurring in global fisheries is currently not applicable here. Furthermore, the fact that outside fishers are wil l ing to make multi-day journeys to fish at Maliangin suggests that fish stocks in Banggi are relatively abundant compared to other areas of Sabah. Excessive exploitation may be restricted by seasonal weather patterns, whereupon strong south west monsoon winds reduce fishing effort from June to September. This natural barrier may indirectly act to conserve fish stocks, as noted also by Alcala and Luchavez (1982) about the Apo Island fisheries in the Philippines. On the other hand, it was also apparent that the open access nature of Banggi's reef fisheries has led to some extent of Malthusian overfishing (Pauly et al, 1989). Anecdotal evidence suggests that the number of fishers has risen markedly in the past ten to fifteen years, corresponding to a two to four time decrease in individual catch rates (various fishers, pers. comm.). A t the same time, dynamite and cyanide fishing, hallmarks of Malthusian overfishing, are a regular occurrence in southern Banggi (Koh et al, 2002; Daw et al, 2002). The commonly observed range of fish lengths (15 to 25 cm) straddles the stages of ecosystem and Malthusian overfishing, as defined for Southeast Asian reef fisheries by McManus et al. (2000). Furthermore, biological overfishing is l ikely applicable to L R F T 37 species such as the humphead wrasse (Cheilinus undulatus), and humpback grouper (Cromileptes altivelis), whose populations are severely depleted in Banggi (Oakley et al, 1999). Banggi fishers also attested to the rarity o f capturing these species. In particular, the economic reliance of hook and line fishers on the live coral grouper fishery is a cause for concern, as it intensifies fishing pressure on a species whose life history makes it intrinsically vulnerable to exploitation (e.g., Sadovy et al, 2003).This highlights the urgency to initiate management of the coral grouper fishery while it has not yet collapsed. The prevailing socio-economic conditions in Banggi also act as contributing factors to unsustainable fisheries. Poor economic development limits opportunities for non-fishing employment; population growth augmented by migrants from other parts of Sabah or the Philippines increases pressure on coastal fisheries resources (Biusing, 2001). This trend has already led to overfishing throughout Southeast As i a (Ablan et al, 2002). Other factors such as cultural ties to fishing and lack of conservation ethic (Fisher, 2000) have the potential to maintain or increase entrants to the fishery. These social and economic factors also influence the decision of youths to stay in or leave Banggi, thereby impacting the future fishing population (Table 2.4). Table 2.4 Summary of drivers of fishing activity in southern Banggi. Driver Impact on fishing activity Pathway Human population growth Increase Influx of migrants from neighbouring islands Maintain/Decrease Youths leaving for urban centres Economic incentives Increase High monetary incentives from Live Reef Fish Trade Subsidies Increase Decrease cost of fishing Culture Maintain/Increase Cultural connection to fishing lifestyle; low conservation ethic Seasonal weather Decrease South west monsoon winds prevent fishing Globally, fisheries subsidies have been shown to have negative impacts on fisheries sustainability (Munro and Sumaila, 2002). The motorboat aid scheme implemented by the Malaysian government can actually lead to overexploiting the resource base in the long term by increasing fishing efficiency. A form of indirect subsidy exists in non-fishing family 38 members entering the workforce. B y contributing to household income, they enable fishers to continue fishing even when it is unprofitable to do so (Pauly, 1997). This is exemplified by family members who work in urban centers and send money home, or women participating in seaweed cultivation (see Cooke, 2003). The various drivers of fishing effort are summarized in Table 2.4. 2.6 Summary and concluding remarks At this point, it is not possible to determine conclusively the current status o f Banggi's coral reef fisheries. Banggi's reefs appear productive enough to meet the needs of the present population; how long this can last is unknown. Fishing effort is influenced by a broader realm of social, ecological, and economic drivers which have to be explicitly considered when pursuing further assessment. The overall impression is that the continuation of present exploitation rates w i l l lead to declining resources in the long term. While fisheries productivity has not yet declined to a critical level, the opportunity should be taken for fisheries managers to initiate precautionary policies which wi l l conserve reef resources while enabling sustainable use in the long term. The results in this study provide a snapshot of Banggi's reef fisheries. However, in the absence of any data, this study serves as an important starting point for understanding the dynamics of Banggi's reef fisheries. Future work includes: i) Expanding sampling locations and incorporating seasonal catch and fish size data to further assess exploitation status; ii) using Ecopath with Ecosim (Pauly et al, 2000) ecosystem modeling software to undertake preliminary exploration of the biological and socio-economic impact of different management strategies; iii) identifying best policy option (s) which wi l l balance the need for conserving reef fisheries resources with the need to derive economic benefits from the ecosystem. 39 2.7 References Ablan, M . C . A . , McManus , J.W., Chen, C . A . , Shao, K . T . , Be l l , J., Cabanban, A . S . , Tuan, V . S . , Arthana, I.W., 2002. Meso-scale Transboundary Units for the Management of Coral Reefs in the South China Sea Area. Naga, Worldfish Center Quarterly 25, 3&4, PP.4-8. Alcala , A . C . , Luchavez, T., 1982. Fish yield of the coral reef surrounding Apo Island, Negros Oriental, Central Visayas, Philippines. In: Gomez, E .D . , Birkeland, C .E . , Buddemeier, R . W . , Johannes, R . E . , Marsh, J .A. Jr., Tsuda, R .T . (Eds.), Proceedings of the 4th International Coral Reef Symposium. Marine Science Center, University of the Philippines, Manila, pp.69-73 Amar, E . C . , Cheong, R . M . T . , Cheong, M . V . T . , 1996. Small-scale fisheries of coral reefs and the need for community-based resource management in Malalison Island, Philippines. Fish. Res. 25, 265-277. Annon., 2003. Profail Daerah K e c i l Banggi Tahun 2003. Unpublished report. 25 pp. Burke, L . , Selig, L . , Spalding, M . , 2002. Reef at Risk in Southeast Asia . W R I , Washington. 72 pp. Biusing, R . E . , 2001. Assessment of coastal fisheries in the Malaysian-Sabah portion of the Sulu-Sulawesi Marine Ecoregion. W W F Malayisa, Kota Kinabalu, 719 pp. Cabanban, A . S . , Biusing, R . E . , 1999. Coral reef fisheries and their contribution to marine fish production in Sabah, Malaysia. In: Cabanban, A . S . , Phillips, M . (Eds.), Proc. Workshop on Aquaculture of Coral Reef Fishes and Sustainable Reef Fisheries. Institute of Development Studies, Sabah. 40 Cooke F . M . , 2003. L iv ing A t the Top End: Communities and Natural Resource Use in the Kudat/Banggi Region of Northern Sabah. W W F Malaysia, Kota Kinabalu, 45 pp. Daw, T., Jeffrey, L . , B i n A l i , M . A . , 2002. Preliminary Assessment of the Live Reef Fish Trade in the Kudat Region. W W F Malaysia, Kota Kinabalu, 47 pp. Fisher, FL, 2000. A Socio-Economic Assessment of Coastal Communities of Pulau Banggi Sabah, East Malaysia. Master o f Science Thesis, University of Minnesota. Institut Penyelidikan Mar in Borneo ( IPMB), 2003. Laporan Kajian Mengenai Pengetahuan dan Pandangan Masyarakt Pulau Banggi Terhadap Penubuhan Taman Tun Mustapha.Unpublished report. 36 pp. K o h , L . L . , Chou, L . M . , Tun, K . P . P . , 2002. The status of coral reefs o f Pulau Banggi and its vicinity, Sabah, based on surveys in June 2002. R E S T Technical Report 2/02. Department of Biological Sciences, National University of Singapore, Singapore. 35 pp. McManus , J.W., 1997. Tropical marine fisheries and the future of coral reefs: a brief review with emphasis on Southeast Asia . Coral Reefs 16, Suppl: S121-S127. McManus, J.W., Menez, L . A . B . , Kesner-Reyes, K . N . , Vergara, S.G., Ablan, M . C . , 2000. Coral reef fishing and coral-algal phase shifts: implications for global reef status. ICES J. Mar. Sci . 57, 572-578. Munro, J.L., 1996. The scope of tropical reef fisheries and their management. In: Polunin, N . V . C . , Roberts, C M . (Eds), Reef Fisheries. Chapman & Hal l , London, pp.1-14. 41 Munro, G . , Sumaila, U .R . , 2002. The impact of subsidies upon fisheries management and sustainability: the case of the North Atlantic. Fish Fish 3 4, 233-250. Oakley, S., Pilcher, N , Atack, K . , Digges, C. , Enderby, S., Mackey, G . , Clubb, R., Stapelton, K . , Toh, S . M . , Huet, C. , Morton, T., 1999. Reefs under attack: the status of coral reefs of Sabah, East Malaysia. Presented In: 4 t h International Conference on the Marine Biology of the South China Sea, Quezon City, 20-22 October, 1999. Pauly, D . , Silvestre, G . , Smith, I.R., 1989. On development, fisheries, and dynamite: a brief review of tropical fisheries management. Nat. Resour. Model 3 3, 307-329. Pauly, D . , 1997. Small-Scale Fisheries in the Tropics: Marginality, Marginalization, and Some Implicatins for Fisheries Management. In: Pikitch, E . K . , Huppert, D .D . , Sissenwine, M . P . (Eds.), Global Trends: Fisheries Management. American Fisheries Society Symposium 20, Bethesda, Maryland, p.40-49. Pauly, D . , Christensen, V . , Dalsgaard, J., Froese, R., Torres, F . Jr., 1998. Fishing down marine food webs. Science 279, 860-863. Pauly, D . , Christensen, V . , Walters, C , 2000. Ecopath, Ecosim, and Ecospace as tools for evaluating ecosystem impact of fisheries. ICES J. Mar . Sci . , 57 3, 697-706. Pilcher, N , Cabanban, A . S . , 2000. The Status of Coral Reefs in Eastern Malaysia Global Coral Reef Monitoring Network ( G C R M N ) Report. Australia Institute of Marine Science, Townsville. Sadovy, Y . J . , Donaldson, T.J. , Graham, T.R., McGi lvray , F. , Muldoon, G.J . , Phillips, M . J . , Rimmer, M . A . , Smith, A . , Yeeting, B . , 2003. While Stocks Last: The Live Reef Fish Trade. Asian Development Bank, Manila, 147 pp. 42 Sadovy, Y . , 2005. Trouble on the reef: the imperative for managing vulnerable and valuable fisheries. Fish Fish 6, 167-185. Talaue-McManus, L . , Kesner, K . , 1995. Evaluation of a Philippine municipal sea urchin and implications of its collapse. In: Juinio-Menez M . A . , Newkirk, G.F . (Eds.), Philippine coastal resources under stress. Selected papers from the Fourth Ann . Common Property Conf , Manila , Quezon City, Philippines, June 16-29,1993. Marine Science Institute University Philippines, Quezon City, Philippines, pp. 229-239. Wood, E . M . , 1978. Coral reefs in Sabah: actual damage and potential dangers. Malay. Nat. J. 31 (1), 49-57. 43 C H A P T E R 3 S E A S O N A L I T Y A N D T R E N D S I N T H E R E E F F I S H E R I E S O F P U L A U B A N G G I , S A B A H , M A L A Y S I A 3.1 Introduction Multispecies tropical reef fisheries are dynamic, and often characterized by distinct seasonality with regards to catch rates, distribution of effort, and target species (e.g., see Alcala and Luchavez, 1981). These spatial and temporal patterns of catches and effort are dictated by a range of biophysical, ecological, and economic factors, such as monsoon winds, spawning aggregations, and pursuit of highly valuable fish species for trade. To reef resource managers, understanding the spatial and temporal trends of a fishery is important for assessing the impact of the fishery in terms of both ecology and socioeconomics. Understanding what, when, where, and why fishers fish is necessary for designing appropriate management policies for the fishery. Uncovering historical trends in catches is further needed to highlight changes in the state of the fishery so that underlying causes can be identified, thereby bringing to light issues for management to act upon. Being aware of temporal changes and trends in a fishery is particularly pertinent in Southeast Asia , where overfishing in recent decades has occurred in tandem with the widespread practice of blastfishing, leading to a decline in fish catches (McManus, 1997). This situation has been exacerbated by the lack of long-term monitoring data for the region's reef fisheries (DeVantier et al, 2004); thus many of these fisheries which are dominated by small-scale fishers escape management attention. This study focuses on the reef fisheries of Pulau 1 Banggi, in Sabah, Malaysia. I first quantify the seasonal variation in catches, effort, and species composition of the island's artisanal reef fisheries, based on fishery surveys carried out in two different seasons in June/August 2004 and March/Apri l 2005.1 then utilize information from interviews with fishers to establish a preliminary trend of the temporal change in catch rates and effort for the island's reef ' Pulau is the Malay word for island. 44 fisheries. Insights derived from examining these trends can help in the development of more enduring and relevant fisheries policies because they adhere to local fishing practices and are thus in tune with local weather cycles, fish behaviour, and socio-economic needs. In light of Banggi being included in the recently approved Tun Mustapha Park (TMP) , lessons learned from this study can ultimately contribute to the development of zoning plans within the Tun Mustapha Park (TMP) that meet fisheries conservation objectives as well as accommodate livelihood needs of coastal communities. 3.2 Background 3.2.1 Study site Pulau Banggi (7.14'N, 117.10'E) is situated off the northern tip of Sabah, East Malaysia (see chapter 2, Fig.2.1), and is bounded by the South China Sea to the west, and the Sulu Sea to the east. Banggi is the biggest island in Malaysia, covering a total area of 700 k m 2 , with a coastline of 420 km (Anon., 2003). The nearest mainland town is Kudat, which is around 30 km across the Banggi Channel. This study focuses on southern Banggi, with the research base located in Karakit, the administrative centre and main entry point to the island. Together with the adjoining villages of Singgahmata and Perpaduan, Karakit comprises the population centre and main socio-economic hub of southern Banggi. The primary fishing ground studied is the Maliangin area, immediately south of Karakit (Fig.3.1). This is one of three main fishing areas in Banggi, the others being Kuambang and Sibogo in the east, and Balambangan in the west. The population of Banggi is estimated to range from around 18,000 (Biusing, 2001) to 20,000 (Institut Penyelidikan Marin Borneo, 2003). There are around 1100 inhabitants in the vicinity of the study site (Teh et al, 2005). Banggi remains relatively undeveloped, and coastal households are considerably below the Sabah poverty line ( I P M B , 2003). Fishing, on which these communities continue to depend heavily for livelihood, accounts for about 70% of the island's economic activity ( I P M B , 2003). There are 1195 fishers on the entire island of Banggi (Kudat Fisheries Department, unpublished data); the population of fishers in the main 45 fishing villages in proximity to Karakit (i.e., Singgahmata, Perpaduan, Patanunam, and Maliangin Besar) is estimated to number around 90. 2 In Kobong and Lumais, two other villages whose fishers frequent the Maliangin fishing grounds, there are approximately 40 to 50 fishers. Figure 3.1 Maliangin fishing ground map showing spatial distribution of effort by gear and fishery type. 3.2.2 Reeffisheries of Banggi In this study, reef fisheries refer to the capture of both demersal reef fish, as well as reef-associated and coastal species such as scombrids and carangids. A description of Banggi's reef fisheries can be found in Chapter 2 and Teh et al. (2005). Briefly, the reef fisheries of Banggi are small-scale, artisanal, and can be considered open access with little regulation. Fishing is done with traditional gears, with hook and line and gillnets being the two most important fishing gears in terms of fish landings (Chapter 2 and Teh et al, 2005). Other fishing methods include the use of traps, jigs, spears and spear guns, explosives, cyanide, and 2 This number is estimated based on information supplied by a local district administrator, and confirmed by personal observations and conversations with fishers in the villages. 46 reef gleaning. The majority of local fishers use wooden boats that are powered by modified water pump engines (pump boats). A small number operate outboard engine fiberglass boats that are provided as part of a poverty alleviation scheme administered by the Sabah Fisheries Development and Fishers Corporation. 3.2.3 Fishing grounds The major southern Banggi fishing ground is the Maliangin area, which is approximately 5 km from Karakit. The total fishing ground is roughly 17 k m 2 in size; within this area, actively 2 3 fished spots cover approximately 8 km . Maliangin is a traditional fishing zone where commercial fleets such as trawlers and purse seiners are prohibited from fishing. 4 It is characterized by shallow reefs 3 to 5 m in depth near shore, and deeper patch reefs of 15 to 25 m depth further offshore. Fishing spots are highly gear specific. During the day, hook and line fishers aggregate around the southern part of the Maliangin area, where they fish over patch reefs up to 30m in depth (Fig.3.1). Gillnet fishers fish over sandy bottom habitat, and have more dispersed fishing grounds that cover areas around Balak Balak and Kaligau, the northern side of Pulau Maliangin Besar, and Lok Tohog (Fig.3.1). These frequently visited fishing spots cover a total area of approximately 6 km . Further fishing grounds include Pulau Malawali , Kuambang, and Sibogo. Sibogo, the furthest away fishing location, is approximately 2 hours away from Karakit by pump boat, and fishing trips to this spot usually last 2 days. The decision to go to these far out fishing grounds are influenced by localized fishing conditions, for example, when the water is "dirty" (i.e. highly turbid), and also by economic factors such as the availability of fuel and ice. When going to further fishing grounds, gillnet fishers from the same village tend to go out together for safety reasons.5 3.2.4 Seasonal weather patterns Sabah's climate is characterized by the Asian southwest (SW) and northeast (NE) monsoons. The N E monsoon prevails from November until March, and the S W monsoon dominates 3 Hook and line and gillnet fishing areas estimated from map of Karakit (Series T735 Sheet 7/117/13), published by the Director of National Mapping, Malaysia 1984. 4 The Malaysia Fisheries Comprehensive Licensing Policy designates the area within five nautical miles of the shoreline as a traditional fishing zone. 5 Fishers in southern Banggi have recently been robbed of motors and boats while at sea, and are weary of traveling too far from their home village, and of traveling alone at night. 47 from M a y until September. Two inter-monsoon periods between A p r i l and May, and September and October bring calmer weather. The N E monsoon brings maximum annual rainfall from November to January, after which ensues a period of strong north easterly winds. Starting in the spring (March), the winds change to a prevailing south westerly direction, and the calmest weather during the year occurs from March/Apr i l until June. Banggi experiences frequent stormy and gusty weather from June to September, and gusts of up to 98 km h"1 have been recorded ( http://www.weatherunderground.com/) during this time. However, these winds do not blow continuously, and there are intermittent periods of two or three calm days before the winds pick up again. In Banggi, the fishing seasons are largely determined by the severity of monsoon winds. The peak fishing season occurs during the calm period in the beginning of the S W monsoon, between March and May/June. The low fishing season follows suit, coinciding with the windiest months of the S W monsoon from June until September, when the hostile weather prevents regular fishing activity. 3.2.5 Temporal trend in catch and effort data The activities of Banggi's artisanal reef fisheries are not recorded in the official statistics for the Kudat region, of which it is a sub-district. There is thus no reliable time series catch data for Banggi's reef fisheries. In the one available study which examines temporal trends in catches of Sabah's coral reef fish, Cabanban and Biusing (1999) indicate that after experiencing a peak in the 1980s, coral reef fish contribution to total marine landings have been declining through to the end of the analysis period (1991). Similarly, Daw et al. (2002) report that between 1997 to 2002, fishers involved in the Kudat Live Reef Fish Trade ( L R F T ) experienced 50% to 90% reduction in yields of grouper species (Plectropomus and Epinephelus spp.). A t the same time, L R F traders have seen a ten fold decline in catches of the highly valuable humphead wrasse (Cheilinus undulatus) between 1995 and 2003 (Helen Hendry, unpublished data cited in CITES , 2004). 48 3.3 Methods Field work to collect primary data was conducted in two periods in 2004 and 2005. The first field season lasted six weeks during the S W monsoon windy period in June and August 2004. The second field season was another six weeks in March and Apr i l 2005. This period was selected as it is normally the calm period; however, winds were unusually strong this year, and delayed the anticipated arrival of calm weather. A s a result, recorded catch data might have been an underestimate of normal peak season levels. 3.3.1 Fishery data Quantitative fishery data was collected for the hook and line and gillnet fisheries, which account for the majority o f fish captured for sale in Banggi . 6 Methods used to collect these data included fishing ground surveys and monitoring catch landings. Surveys of the Maliangin fishing grounds were conducted three times a week when weather permitted, between 7:30 to 11:00 a.m., when it was most likely that fishers would have caught some fish, and before they started to head back to land. Fishers were randomly stopped and the following was recorded: a) gear type; b) quantity off ish caught (weight was estimated by my boatman, who was also an experienced fisher); c) number of hours fished; and d) species caught. Fish landings during my first sampling season were recorded at the two main fish buyers' sites at Karakit and Singgahmata from June 16 t h to July 6 t h , and from August 7 t h to August 27 t h , 2004. This data collection period corresponded to the low fishing season, and was considered representative of the catch and effort for that particular time period as no extraordinary events occurred to indicate otherwise. Gillnet fishers landed their catch at the Karakit site between 8:30 to 11:00 p.m. I waited at the landing site during those times and recorded all the catches that were landed during that time frame. Hook and line catches were landed more sporadically at the Singgahmata site. Therefore sampling at this site was done opportunistically. However, the majority of sampling of hook and line catches during 2004 was taken during the fishing grounds survey. 6 Based on information from the two main fish buyers in southern Banggi. 49 During my second sampling season, catch landings were recorded from March 6 to A p r i l 12 t h , 2005. This was supposed to represent the calm weather period, but the continuation of strong N E winds might have caused catches to be lower than normal. The Singgahmata fish buyer had moved his operation to Maliangin Besar in 2005; therefore I traveled to the island two to three times a week to record landings, inspect the catch composition, and measure fish lengths. Receipts covering parts of November 2004 to February 2005 (corresponding to N E monsoon), and mid Apr i l -May 2005 (typically the peak season) were kindly supplied by the Karakit fish trader. Receipts covering mid-Apr i l to M a y 2005 were also supplied by the Maliangin fish trader. In total, data covering three and four sampling periods were available for the hook and line and gillnet fisheries, respectively. Due to the possibility that peak season catches for both these fisheries might have occurred during the A p r i l - M a y sampling period, those receipts covering mid A p r i l to M a y were treated as a separate data set from those I recorded on site from March to mid Apr i l . It should be noted that the sampling frequency for all three data collection periods were not the same. Fish lengths (total length) were measured in 2005 at the two fish landing sites, using a i m long fish measuring board. I attempted to measure every species in the catch, and as many samples of fish in each catch as possible. However, this was difficult to do as customers were frequently waiting to buy fish, and would select the fish they wanted as soon as the fish were landed. In addition, it was sometimes not convenient to measure the fish as they had to be put into an ice box immediately. Consequently, the total sample sizes of some measured species were low, and are presented only for descriptive purposes. 3.3.2 Interviews Interviews were conducted in villages where fishers are known to frequent the Maliangin fishing ground. In 2004, interviews following the semi-structured format were carried out with 20 fishers, 2 fish traders, and 3 village leaders. These interviews took place in the villages of Karakit, Perpaduan, Singgahmata, Maliangin Besar, and Lok Tohog. Two expert interviews were also conducted with a fisheries officer in the Sabah Department of Fisheries in Kota Kinabalu, and a fisheries officer in the Kudat Fisheries Office. In addition, a 50 participative approach (Reason, 1994), in which I joined in fishing trips and reef gleaning activities, was undertaken as a means of furthering my knowledge about local fishing practices. During 2005, a total of 30 additional interviews were conducted with fishers in the villages of Perpaduan, Maliangin Besar, Kobong, Lumais, and Patanunam. Altogether, the number of interviews in 2004 and 2005 represented roughly 30 to 40% of the total fisher population in these villages. M y assistant, Lydia Teh, and I conducted all the interviews in Malay. Prepared questions pertaining to catch, effort, target species, historical trends, and perceptions on the future of fishing in Banggi were asked (See Appendix I). Fishers were given a lot of breadth in answering the questions, and those who brought up new topics or elaborated on certain questions were allowed to do so. A s such, not all interviews followed the same questionnaire format, and the number of respondents varied for each question. In this study, the number of total respondents to specific questions is indicated in brackets. The questions asked in 2005 were more detailed following revision of 2004 questions and subsequent identification of issues that needed further iteration. Depending on fishers' willingness to continue, interviews lasted between 15 minutes to one hour in 2004, and at least half an hour in 2005. Fishers were chosen opportunistically; my assistant and I went house to house and interviewed any fisher who was at home and wi l l ing to speak to us. We did not encounter any refusals and, in general, all interviewees were cooperative in answering the questions. Additional information was gathered through informal interviews conducted whenever the opportunity arose to speak to fishers or villagers about reef resource use. Interview data was also used to recreate a historical trend in mean daily catches, a procedure outlined in section 3.4.6. 3.3.3 Data Analysis Data analysis was done using Microsoft Excel and SPSS 12.0 software. Student's t-test was used for comparing differences between mean revenue derived from the sale of different target species (Section 4.5.1). Catch and revenue data collected for the hook and line and gillnet fisheries were tested for normality. If at least one of the assumptions for normality were satisfied (i.e., data corresponded to a theoretical normal distribution or homogeneity of variances was justified), one way A N O V A was considered robust enough a test (Underwood, 51 1981) to detect inter-seasonal differences in mean catch weight and revenue within gear types. If both assumptions for normality were not satisfied, the data were transformed to log scales, and analyzed using A N O V A . Tukey's multi-comparison test was used to identify significant differences between group means. A l l comparison results were significant at the 95% level. Linear regression analysis (Section 3.4.2) between weather and daily catch quantity was conducted to determine the effect of weather on catches. Data for daily wind speed was extracted from an internet weather service provider (www.weatherunderground.com). 3.4 Results 3.4.1 Target species and fishing practices The coral grouper (Plectropomus spp.) and Spanish mackerel (Scomberomorous commersori) were the two most important target species for hook and line fishers. The former was caught predominantly during the S W monsoon, whereas the latter was targeted during the N E monsoon. Gillnet fishers targeted trevallies (Carangidae) year round. The main target species and gears used to capture them are presented in Table 3.1. Starting in November until February, many fishers also targeted cuttlefish (Sepia spp.), which was caught nearshore either by diving with a spear, or from a boat using a trident-like spear. Once cuttlefish were out of season, squid (Loligo spp.) became abundant during the S W monsoon months, and were caught with squid jigs (using plastic prawn lure) or with trident-like spears. Table 3.1 Target species and the predominant seasons/months during which they are targeted. Gear Target Species Season Months Hook and line - Trolling with live bait Spanish mackerel N E Oct - Feb/March - Trolling with lures Kawakawa (Scombridae) N E Jan - March - Jigging with lures Yellowtail scad N E Jan - March - Baited hook Coral grouper SW April - Sept. Gillnet Trevallies N E & S W Year round Juvenile black tip sharks SW July/August Trap Coral grouper NE/SW Jan - Sept Spear Cuttlefish N E Nov - Feb Squid jig Squid SW June - Sept Diving/gleaning Invertebrates SW -52 Other target species included invertebrates such as scallops (Pectinidae), abalone (Haliotidae), sea urchins (Echinoidea), and sea cucumber (Holothuroidea). The collection of these invertebrates also intensified during the S W monsoon, when sea water temperatures became warmer and fishers were able to dive comfortably. Reef gleaning occurred on the exposed reef flat in front of Karakit and Singghamata during spring low tides of the summer months. During a span of roughly 3 weeks in late July or August, a few fishers targeted juvenile black tip reef sharks which aggregated in shallow waters off small beaches on Banggi's four southern offshore islands. These sharks were caught using drifting gillnets; at one location, approximately 80 kg of juvenile sharks were netted in less than 10 minutes. Lunar phases also affected fishing conditions and target species in Banggi. Banggi fishers claimed that night fishing yields better catches, especially of trevallies (Carangidae) and pink-ear emperor (Lethrinus lentjan). On nights of a new moon during windy conditions, fishers fished for ox-eye scad (Selar boops), using lamps to attract the fish and fishing with lures. These small schooling pelagic species occurred in great abundance off the sheltered northern and southern points of Patanunam, a small island across from Karakit. Kawakawa (Scombridae) was also targeted during windy periods by trolling with a hand line baited with multiple fabric lures. The onset of strong winds during the S W monsoon often resulted in a lack of bait fish, as fishers did not go fishing, or just fished long enough to obtain fish for their own consumption. Since bait fish was not readily sold locally, fishers could not go out to fish for coral grouper when a fair weather day arrived. Consequently, this led some fishers to turn to blast fishing in order to obtain the bait (mainly fusiliers Ptereocaesio spp.) required to fish for coral grouper. 3.4.2 Catch quantity Mean catch for the hook and line fishery in June and August 2004 was 7.13±7.57 kg-fishef'-day"1 while it was 11.17±9.47 kg-fishef'-day"1 in March and A p r i l 2005 (Table 3.2). For mid-Apr i l to M a y 2005, mean catch was 11.57±12.23 kg-fisher"1-day '. It should be noted that only non-zero catches were included for analysis in all seasons. This is because it was not possible to record zero catches for the gillnet landings, while zero catches for the 53 hook and line fishery were not recorded consistently. Mean catches differed between sampling periods (Table 3.3), with 2004 mean catches being significantly lower than March-Apr i l 2005 (Tukey test, p=0.026) and Apr i l -May 2005 (Tukey test, p=0.023). The two periods in 2005 had mean catch weights which did not differ. Table 3.2 Daily catch weights (Mean ± S.D. and median (kg-day"1)) for hook and line, gillnet, ox-eye scad and cuttlefish spear fisheries in 4 sampling seasons. Gear June/Aug 2004 Nov 04-Feb 05 March/April 2005 April /May 2005 Mean Media Mean Median Mean Median Mean Median Hook & Line 7.1±7.6 • 4.1 - - 11.2±9.5 6.8 11.6±12.2 8.8 Gillnet 14.8 ±12.7 11.5 13.4±12.0 10 16.4±14.8 11 18.8±24.0 11 Oxeyescad . . . . 14.1±8.1 13 Cuttlefish spear - - 49 .8±54.8 . . . . . Table 3.3 Results of one-way A N O V A of mean catch weight (kg-day"1) for hook and line, and mean log weight (kg-day"1) for gillnet fisheries between three sampling periods. Gear d.f. M S F P Hook and line Between seasons 2 424.88 4.18 0.016* Within seasons 288 101.59 - -Total 290 - - -Gillnet Between seasons 3 0.225 1.381 0.248 ns Within seasons 517 0.163 - -Total 520 - - -ns = not significant Mean gillnet catches in 2004 were 14.79 ± 12.70 kg-fisher^-day"1 (Table 3.2), while available receipts covering November 2004 to February 2005 showed that mean catch during the N E monsoon was 13.36±12.00 kg-fisher"'day"1. During this period many fishers also specialized in capturing cuttlefish. Mean catch weight for this cuttlefish fishery was 49.8±54.8 kg-fisher'-day"1 (Table 3.2). During March-Apri l 2005, mean gillnet catch was 16.39±14.82 kg fisher"1 day"1, and rose to 18.82±22.98 kg-fisher^-day"1 during mid Apr i l -May (Table 3.2). Mean gillnet catches in all four seasons did not differ significantly (Table 3.3). 54 Regression analysis suggests that during March-April 2005, maximum wind speed was a significant (p<0.05) explanatory variable for the variation in total daily fish catches landed for the hook and line fishery. Minimum wind speed, however, was not a significant explanatory variable. For the gillnet fishery, neither maximum nor minimum wind speeds could explain the variability in total daily fish landings (Table 3.4). Table 3.4 Regression analysis of total daily landed catches by gear and wind speed (df = 32 for all analyses). Gear Variable F Significance Hook and line Min wind speed 0.93 0.342 ns Max wind speed 4.22 0.05 * Gillnet Min wind speed 2.96 0.095 ns Max wind speed 3.20 0.083 ns ns = not significant 3.4.3 Catch composition The catch composition for both gears displayed inter-seasonal variation. For the hook and line fishery, catch composition was recorded using frequency of occurrence during 2004, and using weight in 2005. In 2004, demersal reef fish were the most frequently occurring group, making up 76% of a typical catch. Within this group, serranids were the dominant family (36%), followed by lethrinids (17%), nemipterids (10%), and lutjanids (8%) (Fig.3.2).Other families in the catch consisted of carangids (6%) and scombrids (5%). During March-April 2005, the proportion of demersal reef fish in the total catch decreased significantly, comprising only 32% of the catch. Pelagic fish, namely carangids and scombrids became more important, making up 34% and 30% of the catch, respectively. Yellowtail scad was the single biggest contributing species to the catch (31%), and its abundance accounted for the significant increase in contribution of carangids (Fig.3.2). Spanish mackerel (19%) and kawakawa (11%) were responsible for the increase in scombrids. The months of March and April were considered the end of the Spanish mackerel 55 0.8 0 6 0.7 • Jun-Aug 2004 • Mar-Apr 2005 • Apr-May 2005 c 0.5 o 1 O 0.4 — o £ 0.3 J [ L O Fish group Figure 3.2 Catch composition for hook and line landings by major fish groups for 3 sampling seasons. season; thus it is likely that this species contributed more than 19% to the catch during the peak of the Spanish mackerel season from November to January. During April-May 2005, demersal reef fish began to dominate the catch again, rising to 44% of total catch. The proportion of Spanish mackerel remained steady at 18%, while there was a big increase in carangids from 3% to 15% (Fig.3.2). Gillnet catch composition during both sampling periods was recorded by weight. Although gillnet composition was more consistent than hook and line, inter-seasonal variation still existed (Fig.3.3). Carangidae was the dominant family in 2004, making up 57% of the total catch. Demersal reef fish (mainly Haemulidae, Lethrinidae, and Lutjanidae) made up 32% of the catch. In March-April 2005, both carangids and demersal reef fishes constituted less of the catch, accounting for 35% and 27% of the catch, respectively (Fig.3.3). The proportion of scombrids increased from 7% to 21%, due mainly to an abundance of kawakawa. Like hook and line catches, the proportion of demersal reef fishes increased in April-May 2005, from 31% to 40% (Fig.3.3). One distinct change was the abundance of stingray during April-May 2005 which increased to 16% from less than 5% in the other seasons (Fig.3.3). In November 2004 to February 2005, there was the largest proportion of demersal reef fishes and smallest proportion of carangids among all sampling periods (Fig.3.3). Although 'expert' cuttlefish 5 6 0.5 • Jun-Aug 2004 • • N o v 0 4 - F e b 0 5 Figure 3.3 C a t c h c o m p o s i t i o n o f g i l l ne t l a n d i n g s b y m a j o r fish g roups fo r 4 s a m p l i n g seasons . fishers could make large catches of cuttlefish ranging from 6 to 97 kg per trip during this period, squid and cuttlefish together only made up 1.65% of overall gillnet catches (4.48 ± 3.40 kg of cuttlefish per trip). 3.4.4 Distribution of effort The highest observed aggregation of effort in Maliangin occurred during calm days of the SW monsoon. An average of 11 (ranging from 8 to 23) hook and line fishing boats per day were observed during fishing ground surveys of Maliangin in 2004. In contrast, only six fishing boats (ranging from 4 to 12) were recorded per day during March-April 2005. In general, fishers fished up to six hours a day during windy conditions. In calm weather fishers made two fishing trips a day, fishing up to 9 hours daily. During the calm season fishers went fishing daily, whereas during windy periods they made only three to four fishing trips a week. This was usually most prevalent during the SW monsoon winds (June-September); however this was also the case during March-April 2005 due to the unseasonably strong winds. On nights of a new moon during windy conditions, effort was concentrated on fishing for ox-eye scad (Selar boops) off the northern and southern points of Patanunam Island. Interviewed 57 fishers who fished for ox-eye scad (n=3) indicated that approximately 10 and up to 20 fishers were involved in this fishery, although the normal range of boats I counted was six to eight. Hook and line fishers were observed fishing at more dispersed locations in 2004. Besides aggregating at the southern end of Maliangin, they also fished at reefs on the north facing side of Maliangin Besar Island (Fig.3.4). In contrast, during March-Apr i l 2004, fishers kept to a much smaller spatial scale, keeping to an actively fished area of approximately 8 k m 2 . 0 0 . 5 1 1.5 2 Figure 3.4 Map showing seasonal usage of Maliangin fishing ground by hook and line fishers. There were around 15 regular gillnet fishers operating in the southern Banggi area. Like the hook and line fishery, gillnet fishers were also limited by the severity of the winds. However, 7 It was possible to keep track of ox-eye scad fishing boats by counting the number of illuminated boats from shore. However, this is likely an underestimate as not all boats use lamps. 58 since gillnetting grounds were more sheltered, gillnet fishers could make at least four to five fishing trips per week during windy periods, and fish daily during calm weather. During favourable weather, gillnet fishers set their nets twice a day, once in the evening from around 6 to 8 p.m. and again in the early morning from 4 to 6 a.m. 3.4.5 Economic Benefits from fishing Fish prices Fish prices tend to increase during the northeast monsoon due to the decreased fishing activity (Biusing, 2001), although this trend might vary according to species and fish buyer. During the S W monsoon 2004 field season, the prices paid by the Karakit fish buyer for trevallies (Carangidae) was 4.5 R M k g " 1 , 8 whereas during March-Apr i l 2005 the price rose to 5.5 R M kg" 1. The price for large demersal reef fish and crabs also rose from 1.7 R M R M kg" 1 and 2.5 R M kg"1 in 2004 to 2 R M kg"1 and 3 R M kg"1 respectively in 2005. In contrast, the price for sting ray fell from 1 R M kg"1 to 0.7 R M kg" 1. The Singgahmata fish trader bought demersal reef fish for 1.5 R M kg"1 in 2004, but this was lowered to 1.4 R M kg"1 during 2005. Table 3.5 provides a list o f recorded fish prices in 2004 and 2005. Fishing revenue Average revenue estimated from daily hook and line catches in 2004 (coral grouper season) was R M 20.20 ± 28.72 (n=60). Twenty-eight percent of the sampled catches included a live coral grouper. Within this group, the average revenue obtained from the sale of coral grouper was R M 44.46; this made up on average 71% (ranging from 43 to 100%) of daily total catch revenue. During March-Apr i l 2005, average daily revenue for hook and line catches was R M 25.46 ± 23.80 (n=139). Spanish mackerel was present in 28% of total recorded catches, and of these catches, made up on average 76% (range 29 to 100%) of the fisher's total daily revenue. Overall, Spanish mackerel accounted for 32% of total daily revenue recorded for this season. Receipts from A p r i l - M a y 2005 showed that average daily revenue was R M 29.98 ± 34.57 (n=90). The contribution of Spanish mackerel dropped slightly, accounting for 27% The exchange rate on December 12, 2005 was 1 USD = 3.78 R M 59 of overall total daily revenue. Revenue data was transformed to a log scale for statistical analysis. Mean log revenue from 2004 differed significantly from both sampling periods in Table 3.5 Fish prices recorded during 2004 and 2005 at three fish trading operations in southern Banggi. Species/group Price (RM kg"1) Price (USD kg"1) Karakit 2004 2005 2004 2005 Trevallies 4.5 5.5 1.2 1.5 Thicklip and giant trevally 3.5 4.0 0.9 1.1 Yellowtail scad 2.5 2.5 0.7 0.7 Spanish mackerel (big) - 6.0 - 1.6 Spanish mackerel (small) 4.0 4.0 1.1 1.1 Other pelagics (scads, tuna) 1.5 1.5 0.4 0.4 Reef fish (big) 1.7 2.0 0.4 0.5 Reef fish (medium) 1.5 1.7 0.4 0.4 Reef fish (small) 1.5 1.5 0.4 0.4 Stingray 1.0 0.8 0.3 0.2 Crab 2.5 3.0 0.7 0.8 Singgahmata/Maliangin Trevallies 4.5 5.0 1.2 1.3 Thicklip and giant trevally - 4.0 1.1 Yellowtail scad 2.0 1.6 0.5 0.4 Spanish mackerel 4.0 5.0 1.1 1.3 Dolphin fish, barracuda - 1.0 0.3 Reef fish (all sizes) 1.5 1.4 0.4 0.4 Live fish pond Plectropomus leopardus 42.0 - 11.1 --oversize (>2kg) 50 per fish - 13.2 -Plectropomus maculatus 30.0 - 7.9 -Plectropomus oligocanthus, 10.0 - 2.6 -P.areolatus Epinephelus fuscoguttatus 10.0 - 2.6 -Cheilinus undulatus 90.0 - 23.8 -2005 (Tukey's test, p<0.001 ), but there was no significant difference between the mean log revenue of the two 2005 sampling periods. It should be noted that April-May 2005 revenues might have been an underestimate due to the fact that the sale of coral grouper (which was beginning to be targeted around this time) was not captured in these receipts, as live fish were sold to a live fish pond in Perpaduan. When considering the cases where only a target species (either coral grouper or Spanish mackerel) was caught, there was also no significant difference between mean revenue derived from the sale of Plectropomus spp. or Spanish mackerel (t-test, p=0.208, d.f.=20). 60 The average da i l y revenue for gi l lnet f ishers i n 2004 was R M 33.57 ± 29.11 (n=68). The largest contr ibut ion (38%) to revenue was der ived f rom the sale o f treval l ies (Carangidae). Ave rage da i l y revenue dur ing M a r c h - A p r i l 2005 was R M 42.95 ± 5 6 . 1 7 (n=186), and was R M 43.27 ± 44.83 (n=54) dur ing A p r i l - M a y 2005. D u r i n g the N o v e m b e r 2004 to February 2005 per iod , the mean da i l y revenue f rom gi l lnet catches was R M 38.88 ± 39.86 (n=203). There was no signi f icant di f ference i n revenue among a l l samp l ing per iods for gi l lnet f ishers (Table 3.6). T a b l e 3.6 Results of A N O V A of mean log revenue (RMday 1 ) for hook and line (3 sampling periods) and mean revenue (RMday 1 ) for gillnet (4 sampling periods). d.f. M S F P Hook and line Between seasons 2 2.373 10.181 O.001 Within seasons 286 0.233 - -Total 288 - - -Gillnet Between seasons 3 1629 0.773 0.509 ns Within seasons 507 2107 - -Total 510 - - -ns = not significant The mean da i l y revenue generated f rom the cutt lef ish f ishery f rom N o v e m b e r to February was R M 224.25 ± 246.80 (n=15). H o w e v e r , this was heav i l y b iased b y two extremely large catches o f cutt lef ish w h i c h generated revenues o f R M 603 and R M 886. I f these two outl iers were e l iminated, mean da i l y revenue dropped to R M 125.44 ± 109.79, st i l l a remarkab ly h igh value w h i c h was at least three to f ive t imes the average revenue acquired f rom the hook and l ine and gi l lnet f isheries. Fishing costs Petro l consti tuted the largest f i sh ing associated cost for B a n g g i ' s sma l l scale f ishers, and accounted for approx imate ly 33 to 45 % and 27 to 3 7 % o f gross revenue for hook and l ine and gi l lnet f ishers, respect ively. Pet ro l is transported f rom Kuda t to B a n g g i v i a cargo ship, 61 and sold to fishers through the fish trader, or from small sundry shops. Ten litres of petrol costs approximately RM 10. Aside from petrol, hook and line fishers had minimal operating costs (see Appendix IV, Table 1). Bait was not readily available in Karakit, and was not frequently purchased by hook and line fishers. Instead, fishers started off their fishing trips by fishing for bait fish first. Therefore the petrol and time spent on fishing for bait could be considered to represent the cost of bait; however, time was not explicitly accounted for in the ensuing analysis. The majority of fishers owned their own pump boats, which have a life of 5 to 10 years. Pump boat engines cost from RM 600 for a 5 HP to RM 1000 for a 7 HP model, while the cost of the boat itself ranged from RM 700 to RM 1500, depending on size and workmanship. Gillnet fishers incurred high capital cost for their nets. Depending on the desired length and type of net, the price of a net ranged from RM 100 for a 6.25 cm mesh size net of approximately 200 m, to RM 400 for a black drift gill net. A standard 6.25 cm mesh net could be used for up to one year, while the drift gill net could last for a substantially longer time. Gillnet fishers spent RM 6 for a block of ice, which was sufficient to last them two days at sea for the longer fishing trips to Sibogo. Net Income An estimation of monthly net income according to each sampling period is presented in Table 3.7 for both the hook and line and gillnet fishery (see Appendix IV for assumptions). Table 3.7 Estimation of net monthly income (RM) and profitability for hook and line (HL) and gillnet (GN) fishers. Season/ Gear June/Aug 2004 Nov 04-Feb 05 March/Apr i l 2005 Apr i l /May 2005 H L G N H L G N H L G N Oxscad H L G N Gross daily revenue 17.62 33.57 38.88 26.26 42.95 14.14 29.98 43.27 Gross monthly revenue 281.92 671.40 - 933.12 525.20 1030.80 113.12 719.52 1038.48 Variable costs 171.79 320.00 - 357.65 214.74 357.65 43.79 257.68 357.65 Net revenue 110.13 351.40 575.47 310.46 673.15 69.33 461.84 680.83 Depreciation costs 16.25 24.58 24.58 16.25 24.58 20.31 16.25 24.58 Net monthly income 93.88 326.82 - 550.89 294.21 648.57 49.02 445.59 656.25 Profitability * 0.33 0.49 0.59 0.56 0.63 0.43 0.62 0.63 *Ratio of Net income to gross revenue. Note that labour costs are not considered. 62 The profitability of hook and line fishing was lower than gillnet in all seasons. However, hook and line profitability in April and May could potentially increase from 62% to 74%, thereby surpassing gillnet profitability, i f revenue from coral grouper sales were included (Appendix IV, Table 3).9 The biggest gap between gear incomes occurred during the windy SW monsoon period in 2004, when fishing by gillnet yielded 16% higher profitability than by hook and line. On an annual basis, profit derived from gillnet was approximately double that of hook and line (Fig.3.5). Profitability in this case was equal to gross revenue less fishing related costs, and represented the return to the amount fishers spent to fish. Labour cost was not explicitly deducted from this analysis because of the lack of alternative employment opportunities on Banggi; therefore it was considered reasonable to assume that the opportunity cost of labour was zero. 12 10 4 -• Profit • Cost Hook & Line Gillnet Gear Figure 3.5 Estimated annual revenue, cost, and profit (RM) for hook and line and gillnet fisheries. 3.4.6 Temporal Trends There was a consensus among fishers, regardless of what gear they used, that individual catch rates had declined from the past. Only 2% (n=45) of respondents said that they could still get the same amount of fish now as before. Fifty-nine percent of fishers (n=22) mentioned a 3 or 4 time decrease in catch rate, with one gillnet reporting the largest decrease, a 20 time reduction. Overall, fishers used to catch on average three times more fish than they can now, with the largest proportion of fishers (45%, n=22) suggesting that catch rates had 9 Receipts in April-May 2005 did not include potential coral grouper sales, which would have been sold to a live fish operation. 63 decreased between 3 to 4 times (Fig.3.6). Maximum daily catches used to range from 10 to 0.50 -1 0.45 • B 0.40 -c o •D 0.35 -C O D. 0.30 -10 £ 0.25 -o C o 0.20 -t o CL 0.15 -O 0.10 -0.05 -0.00 --K i l l 1 -1.99 2-2.99 3-3.99 4-4.99 >5 Magnitude of decrease (times) Figure 3.6 Distribution of interview responses for the magnitude of decrease in catch rates. 50 kg for hook and line in the 1980s (Fig.3.7), compared to reported maximum catches between 4 to 30 kg in 2005. Gillnet fishers used to be able to catch up to 200 kg offish per 50 £ 2 0 o y = -0.39x + 813.b^ ^ R 2 = 0.0903 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Year Figure 3.7 Daily catch rates (kg-fisher"' day"1) for 1970-2005, as reported by hook and line fishers during interviews. Regression line y=-0.4x+813.8, R2=0.09, N = 70 data points. 64 day, while they reported that they could only get a maximum of 30 kg per day now. Regression analysis showed a declining temporal trend in catch rates (Fig. 3.7), and suggested that time was a significant explanatory variable for the variation in daily catch rates (F=6.85, df=70, p=0.01). A one sided t-test showed that the mean of past catches was significantly higher (p<0.0005) than the mean of present catches. There was no significant correlation (Pearson correlation coefficients. 138, p=0.529) between the age of the respondent and the magnitude of decrease mentioned. The majority (60%, n=28) of fishers mentioned that catches just started to decline after 2000, and this was consistent among all gears. Of this group, 32% said that catches started to decline between 2003-2005, and another 29% said between 2000 and 2002. The next most commonly mentioned time period of decline was the 1990s (25%), followed by the 1980s (7%). Other responses included not sure (4%), and no change from before (4%). As recently as 2002 and 2003, two fishers reported being able to catch up to 50 kg a day (Fig. 3.7). A breakdown of fishers by age showed that all the fishers in the two oldest age groups said that catch decreases started in the two most recent time periods (Fig.3.8) Respondents were 1.0 09 0.8 | 0.7 I 0.6 in m 1 0.5 o 2 0.4 •c o a. 0.3 o £ 0.2 0 1 0.0 til • 1980s • 1990s • 2000-2005 IV Age category Figure 3.8 Breakdown by fishers' age for the question of when catch declines first occurred . Age categories are divided as follows I = 20-30 yrs; II = 31-40 yrs old; III = 41-50 yrs; IV = 51-60 yrs; V = >60 yrs. 65 in general not able to quantify the number of hours they used to fish in the past, compared to the present. Instead, 70% (n=10) said that it was easier to get fish last time. O f this group, 43% said that they used to spend less time fishing compared to the present. Both these responses suggested individual catch rates had decreased from the past. Although there are acknowledged issues regarding the use of fishers knowledge, some of which w i l l be discussed in a later section, the fact that such a large proportion of respondents mentioned 2000 onwards as the starting period of noticeable catch decline implied that the magnitude of decrease has been greatest in the last 5 years. Based on this line of thought, a temporal trend of catch rates could potentially be reconstructed the following way: 1) Each respondent was asked to identify the year in which individual catch rates first started to decline; 2) Assume that each respondent represented one unit of magnitude of change in catch rate, in this case the direction of change was a decrease; 3) The proportion of respondents mentioning any one year represented the degree to which catch rates that year had declined from the previous year. For example, i f 50% of respondents mentioned a given year (year x) as the starting year of decline, then catches in that year had declined by 50% from year x-1; 4) One could then start constructing the temporal curve, using 2005 as the starting point and working backwards. The year 2005 was chosen, as quantitative catch data were available for this year; 5) Two temporal curves were built, based on the hook and line mean and median catch rates averaged over all 3 sampling periods (Table 3.2). Only the hook and line fishery was considered here due to data constraints; 6) The largest proportion (67%) of respondents identified the years from 2000 to 2005 as the period when catches started to decline. Thirty percent stated the 1990s, and 4% said the 1980s. A s such, the steepness (slope) of the downward curves increased through time. 66 This procedure produced two temporal trend curves which showed that individual catch rates had declined by roughly 3 times (or 70% decline) over the past two decades (Fig.3.9). This echoed the findings from interviews discussed previously, and illustrated that in this case, even without explicit knowledge about time series information on catch per unit effort, one could roughly arrive at a temporal trend based only on information about when catches first started to decline. 35 -, 0 -I , , , , , , -, 1975 1980 1985 1990 1995 2000 2005 2010 Year Figure 3.9 Derived temporal trend for mean daily catch rates (kgfishef'trip"'). The starting point for both curves is 2005, the value of which is based on either the median or mean (averaged over 3 sampling seasons) daily catch rate obtained from this study. The decrease in catches were mainly attributed to 3 reasons by the interviewees: 1) purse seine vessels from Kudat and other parts of Sabah encroaching upon the traditional fishing zone and catching all the fish; 2) increase in the number of fishers; and 3) use of more efficient or destructive fishing techniques (Table 3.8). Fishers had not noticed a change in the catch of fish at the broad family or group level (e.g., demersal reef fish). In terms of change in specific species, 79% (n=28) of respondents (all of whom used hook and line, trap, or both gears) said that catches of leopard coral grouper (Plectropomus leopardus) had declined significantly from the past, with the remaining saying that pelagic species such as yellowtail scad, Indian mackerel, and Spanish mackerel had 67 declined a lot. In particular, catches of humphead wrasse (Cheilinus undulatus) and humpback grouper (Cromileptes altiveles) were extremely rare. Trap fishers (n=8) used to Table 3.8 Reasons given by interviewees for the decline in catches. Reason for decline Number of times mentioned Purse seines 15 Increase in number of fishers 15 Bomb fishing 11 Use of compressors & spear guns 3 Use of traps 3 Use of nets 4 Use of cyanide 2 Trawlers 3 Do not know 1 catch up to a maximum of 30 kg week"'of coral grouper in the 1980s, compared to a maximum of 10 kg week"1 in 2005. The corresponding figures for hook and line fishers (n=7) were a maximum of 20 kg day"1 in the past (1970s) compared to a maximum of 5 kg day"1 now. Therefore catches of coral grouper by trap had dropped by a factor of 6 times over the past 20 years, while catches by hook and line had declined by a factor of 12 times in the past 30 years (Fig.3.10). The decrease in coral grouper catch rates was mirrored by an increase in the price of the fish. Coral groupers used to be fished and eaten locally just as any other reef fish. The earliest reported date for when coral grouper started to be caught for sale was 1972, during which the fish was priced at RM 2 kg"1 .The price increased steadily through the next decade, increasing to RM 6 kg"1 in 1974 and reaching RM 12 kg"1 by 1988. Fishers identified 1989 as the year when the demand for live coral groupers intensified, and the reported price increased accordingly to RM 20 kg"1. After that, prices increased to RM 25 kg"1, RM 35 kg"1, and to the present price of RM 42 kg"1. 68 30 25 20 .2> 15 10 trap catch 1970s 1980s 1990s 2002-2003 2005 Period 45 40 35 30 I o 25 -g a. a> + 20 §• o ai 15 n 6 o 10 5 0 Figure 3.10 Present and historic prices and mean catch rates of coral groupers using hook and line and traps, as reported by fishers. Units are kg-day"1 for hook and line, and kgweek"1 for trap. Sixty five percent (n=17) of the respondents said that fish sizes had decreased from before. Of these, 64% specified decreased sizes in coral groupers, while the rest referred to emperors, Spanish mackerel, and trevallies. Fishers said that the coral groupers they caught used to commonly weigh 1 to 2 kg, whereas the fish only weighed 0.1 to 0.3 kg or 0.5 to 0.8 kg now. One fisher recounted how he had never seen a Spanish mackerel that was smaller than elbow length (approximately 40 to 45 cm) before 2003. However, since purse seiners started becoming an issue around that time, he had seen Spanish mackerel the length of his index finger (roughly 7 to 8 cm) being caught. Fish length measurements taken in 2004 indicated that the average total length of fish caught by hook and line, excluding Spanish mackerel, was 24 cm (Appendix V, Table 1). For the gillnet fishery, average total length of landed fish was 30 cm (also excluding Spanish mackerel) (Appendix V, Table 2). 3.5. Discussion 3.5.1 Catches, effort, and economics The monsoon driven seasonality of fish yields in southern Banggi was similar to those observed in other tropical fisheries in the region (e.g., see Alcala and Luchavez, 1981). The 69 monsoon weather not only dictated fishers' ability to go fishing, but also had a direct bearing on bio-physical parameters which ultimately influenced the amount o f fish caught. On one hand, this variability was predictable based on temporal seasonal patterns; on the other hand, daily fluctuations in environmental and physical factors (e.g., wind speed, water clarity, currents, moon phase) within monsoon periods made fishing activity and catches highly variable on a day-to-day basis. Nonetheless, it was possible to detect broad seasonal scale patterns in catches and effort for the hook and line fishery, in which mean catch weights between the S W and N E monsoons differed significantly. Seasonal trends of catch quantities and target species were weaker for the gillnet fishery, with no significant difference in catch weights between all seasons. In fact, the median, or ' typical ' gillnet catch remained fairly constant throughout the year, although scombrids showed a much stronger presence in the catch composition in March and Apr i l 2005 than in 2004. This suggested that the gillnet fishers were better able to maintain relatively constant catches year round, even in the face of inclement conditions. One could attribute this to the physical location of gillnet fishing grounds, which tend to be more sheltered (see Fig. 3.2), thus less susceptible to windy conditions than hook and line fishing spots. Furthermore, hook and line fishers need to 'feel' their lines for bites, and windy conditions hampered their ability to do so. Regression analysis supported this contention, as maximum wind speed was a significant explanatory variable for the variation in total daily catch for the hook and line, but not gillnet, fishery. It must be noted that wind data used in this analysis originated from the Kudat airport, which is at least 30 km from southern Banggi, but were considered representative. When fishing conditions were poor in southern Banggi, gillnet fishers were wil l ing to risk traveling to further fishing grounds in Sibogo to get better catches or bigger fish than what would be obtainable in southern Banggi. In contrast, hook and line fishers tend to fish closer to shore, rather than go beyond the immediate Maliangin area when winds were strong. Fishing activity occurred year round at the study site due to the absence of alternate employment opportunities. Total fishing effort (i.e., number of fishers) remained fairly 70 constant throughout the year, albeit distributed differently in terms of time and space. Changes in fishing effort around Maliangin were driven by a mixture of economic and environmental factors that influenced fishers' decision to go fishing or not. The lucrative live fish trade was a direct economic incentive driving the high hook and line effort observed in 2004, even though winds were at their strongest during this period. However, results in this study suggested that fishers might be operating under a false notion about the profitability of the L R F T , as there was no significant difference between the revenue derived from coral grouper and Spanish mackerel. In fact, it was interesting to note that both coral grouper and Spanish mackerel were caught in approximately 30% of recorded catches, and that Spanish mackerel accounted for a slightly higher percentage (76% versus 71%) of total daily revenue. The lower hook and line fishing effort observed during March-Apr i l 2005 could be considered a reflection of fishers trying to minimize economic risks. The months of March and Apr i l were a transition period in terms of being the end of the Spanish mackerel season but not yet the full blown start of the coral grouper season. Fishing effort was reduced during this period of flux as many fishers said that they were waiting for the arrival of southwest winds before resuming normal fishing activity. Unseasonably strong winds added another hindrance to fishing. Consequently, fishing during the day was mostly done near shore to meet subsistence needs only. Fishers from villages other than Maliangin Besar 1 0 diverted their effort to fishing grounds nearer to the southern Banggi villages of Singgahmata and Perpaduan. This shrinking of a fisher's resource space during inclement weather was previously noted (Pet-Soede et al, 2001), and appeared to be a common strategy employed by Banggi's fishers. Instead of making the journey to open sea in Maliangin during unfavourable weather and face uncertain fishing success (fish were not "biting" during this period), they focused on the lower value but seemingly more certain night time ox-eye scad fishery, thereby minimizing both economic and safety risks. B y pursuing ox-eye scad during this season, fishers were able to achieve a profitability (43%) that was higher than that in 2004, although it did not match the higher profitability (56%) of day fishing. Results obtained here seemed 1 0 Fishers from Maliangin Besar are closest to the Maliangin fishing ground, and make trips there everyday. 71 to indicate that the ox-eye scad fishery was more dependable but less profitable, although inconsistency had previously been noted as a feature of ox-eye scad landings elsewhere (Dalzell et al, 1996). The profitability analysis indicated that hook and line fishers' practice of targeting seasonal species in fact yielded poorer returns than that of gillnet fishers, who did not target any specific species. In fact, there was no significant difference between seasonal gillnet revenue. Estimated profitability for the hook and line fishery was lowest during the windy S W monsoon period in 2004 (33%), due to a combination of low catches and disproportionately higher fuel cost (57% of gross revenue). This has serious implications for fishing households, which are at their most financially vulnerable state during this time period due to lost fishing days. Moreover, local shopkeepers do not extend credit to fishers, which can lead to potential social problems with regards to nutrition and health. Fuel was the biggest operating cost for fishers; a 10% increase in fuel price would have caused a 3 to 5 % decline in profitability, depending on season. Fuel usage was exacerbated by hook and line fishers' fishing strategy, which involved moving frequently between different reefs within their fishing grounds in order to find a spot where fish, and in particular target species, were 'biting'. Hook and line fishers who sold to the Maliangin Besar fish buyer also received a lower price compared to the Karakit fish buyer, to whom all gillnet fishers sold. However, there were around seven regular fishers from Maliangin Besar who obtained a share in the profits from the fish buying operation, which was run by an extended family member.' 1 Therefore although they received lower unit prices, their actual profitability might have been higher than that estimated here. Besides weather and catchability offish, social reasons also affected fishers' decisions on whether or not to fish. For example, three fishers who fished regularly in 2004 found construction jobs during my second field season in 2005, and therefore stopped fishing for the duration of their job. Another regular fisherman stopped fishing temporarily in order to 1 ' Other fishers who sell to the Maliangin buyer do so for convenience, as they are mostly fishers who live on Maliangin Besar Island, and do not want to travel to Karakit. 72 prepare for the wedding of his child. Consequently, even though total fishing effort reflected anticipated changes in synchrony with different fishing seasons, it was also subject to the variability of a separate set of socio-economic factors. 3.5.2 Using fishers' knowledge to explain seasonal variation A study by Poizat and Baran (1997) demonstrated that fishers' knowledge o f spatial and temporal patterns of fish occurrence corresponded well to scientific sampling results. Indeed, the value of consulting fishers' local ecological knowledge as a process in managing marine resources has been increasingly emphasized (Johannes et al, 2000). Interviews with fishers in Banggi likewise revealed many ecological observations by which they explained seasonal patterns. It is l ikely that further investigation of the various explanations discussed below w i l l provide valuable information to reef resource managers regarding how environmental parameters, abundance, and behaviour of target species influence the seasonal fishing strategies previously outlined. Fishers attributed the dip in mid-year catches to an abundance of small pelagic fish12 during this period; therefore adult fish had sufficient food resources, and did not take bait. In contrast, a lack of small pelagic fish earlier in the year (March/April) resulted in adult fish readily taking bait. This explanation suggested a link between seasonal fluctuations in catch and spawning and/or recruitment cycles o f fish, and in fact Pauly and Navaluna (1983) demonstrated a relationship between recruitment patterns and monsoon winds for pelagic fish in the Philippines. Fishers reported that fishing during the full moon yielded large catches, particularly of trevallies (Carangidae), with catches of up to 20 to 30 kg per trip obtainable even now. Similarly, the largest catches of a Caribbean carangid species were reportedly made after sunset on a full moon (Johannes, 1981). In Palau, Johannes (1981) documented that the best catches of carangids were also made around the full moon, and suggested that this was due to possible lunar spawning patterns in carangids. 1 2 Fishers referred to "small fish that are on top of the water", which I assume mean small pelagic fish or the young pelagic stages of reef fish. 73 Oceanographic conditions also had a direct impact on the catchability of fish. For example, fishers explained that the lack of fish during the continuous N E winds was due to the presence of two currents, one at the surface and one undercurrent, going in opposing directions. A s such, fish were swept haphazardly along, and are unable to take bait. Another likely reason was that the catchability of demersal reef fish decreased because the continuous winds and unsettled currents prevented fishers from feeling their line for bites. A s a result fishers tended to j i g or troll for pelagic fish instead during windy periods. The increased catch of Plectropomus spp. during the S W monsoon was due to the water becoming clearer beginning in March/Apr i l , thereby allowing the coral groupers to see the bait. Also , the rise in sea surface temperature during the S W monsoon enabled more fishers to go spear fishing or to set traps. It was not clear whether the switch in fishing effort from coral grouper to Spanish mackerel was due more to economic incentives or to seasonal shifts in fish abundance or catchability. This distinction is important to know because each cause suggests different management policies. In the small scale tropical fishery of Spermonde, Indonesia, Pet Soede et al. (2001) found no relationship between observed aggregations of fishing effort and patterns in fish abundance. The contrary seems to be true in Banggi, as fishing effort, particularly for valuable target species, was concentrated at distinct times and places where these fish appeared most abundant.1 3 For instance, the profusion of scads (both yellowtail and ox-eye) and kawakawa caught during the N E monsoon certainly suggested that there was a healthy stock of these pelagics in the Maliangin area during this time. However, during calm conditions these species do not bite, hence were absent from catch landings. Therefore whether or not they remained in the Maliangin area year round cannot be ascertained. It was surprising that none of the fishers mentioned that migration was a factor in determining Spanish mackerel seasonality, even though long migrations by this species have been documented (Collette and Nauen, 1983). Consequently, it was not certain whether Spanish mackerel were still present during the S W monsoon (some were still recorded being 1 3 It is noted that fishers' perception of the presumed availabilityof fish might have led me to make an overly optimistic conclusion as to the actual measured abundance of fish. 74 caught in June and August 2004), but just passed over for the more lucrative coral grouper fishery instead. In Australia's Northern Territory, the Spanish mackerel fishery peaks in September to November, a trend that is thought to be related to reproductive activity (Buckworth & Clarke, 2001). In another study o f the Spanish mackerel fishery in the G u l f of Oman, Claereboudt et al. (2005) suggested that the low point in the fishery during Apr i l and M a y coincided with the fish migrating out of the G u l f waters during their reproductive season. For the coral grouper fishery, fishers suggested that the fish were there throughout the year, and that catches peaked because of improved catchability conditions (warmer and clearer water). Surprisingly, none of the respondents mentioned targeting spawning aggregations as the reason for increased catches of coral grouper during the season, even though Daw (2004) found that fishers around the Banggi islands knew o f such spawning \ aggregation sites. 3.5.3 Temporal trends in catch A t overexploited rates of fishing, fishers first experience a decline in catch per unit effort (CPUE) of target species, followed by decline in overall catch rates (Jennings and Lock, 1996). This situation is certainly applicable to Banggi's coastal fisheries, where almost all respondents indicated a temporal decline in catch rate. Within the context of widespread overfishing of reef and coastal fisheries in Southeast Asia , the three fold decline in catch rates over the past two decades suggested by Banggi fishers appears reasonable. This is comparable to the small scale municipal hook and line and gillnet fishery in Lingayen G u l f in the Philippines, where catch rates decreased by 124.5% and 38.2% respectively between 1985 and 2001 (P. Al ino , unpubl. data, available: www.unuftp/is/Philippine%20Fisheries.pdf). When compared to Olango Island in Cebu, Philippines, where C P U E (kg-person^-day"1) declined by a factor of five between 1970 and 1998 (White & Cruz-Trinidad, 1998), Banggi may have experienced a lesser magnitude of decrease. In fact, daily catch quantities for the gillnet and hook and line fishery in Banggi were higher than those recorded during the early 1990s for another small-scale Philippine reef fisheries in Malalision Island (Amar et al, 1996). 75 Anecdotal evidence pointed to the fact that fishers had expanded the boundaries of their fishing grounds due to overexploitation. Many fishers mentioned how they used to be able to catch big sized fish of species such as emperors and trevallies from fishing near shore or from the beach. However, only small fish species are left near shore now, and fishers have to go to deeper areas to find bigger fish. In addition to overfishing, the lack of big fish near shore might be attributed to degraded coral reef habitat caused by blast fishing. Technology has played a part in this decline. Boats equipped with compressors have enabled fishers to dive to greater depths and easily catch large quantities of fish, especially groupers, and invertebrates, during the night. Furthermore, the introduction of fuel-economizing pump boats in the early 1990s has allowed fishers to access deeper reefs further offshore in the Maliangin area, which were previously hard to reach by paddled canoes. A notable aspect about the temporal trend in southern Banggi is that compared to neighbouring countries, declines appeared to have only happened relatively recently. While coastal inshore fisheries in the Philippines and Indonesia had been considered heavily and over-exploited since the 1970s, (Alcala and Russ, 2002; Pet-Soede et al, 1999) the earliest decade mentioned by fishers here was the 1980s. This paralleled a report by Cabanban and Biusing (1999), who used official statistics to show that reef fish landings in Sabah peaked in the mid 1980s, and then declined up to end of the analysis period in 1991. The relatively recent decline in catch rates could be due to the fact that Sabah remained relatively unpopulated compared to other Southeast Asian coastal regions until recently, resulting in a corresponding lower demand for fish. In fact, the population density of Sabah was only 24 persons km" 2 in 1991 (Anon., 1994). In comparison, even Palawan, one of the most sparsely populated Philippine provinces, had a population density of 34 persons km" 2 in 1991, while populous coastal centres such as Pangasinan and Cebu provinces had densities of up to 346 and 513 persons km" 2 in 1991, respectively ( P O P C O M online database: www. popcorn.gov.ph/dseis/datafmder.php). Fishers had not noticed a change in the type of fish caught, except for valuable commercial species mentioned earlier. This was comparable to a study of South Pacific small scale 76 fisheries (Kitalong & Dalzel l , 1994), which found little change in the composition of landings from 14 years of catch time series data, except for declines of certain species only. The fact that species composition of fish consumed for subsistence had not changed significantly from the past suggested that "fishing down the food web" (Pauly et al, 1998) may not yet have occurred in full fledge in southern Banggi; rather it may be at an early stage thereof. A potentially disconcerting finding was that the majority o f respondents indicated that drastic decreases in catch rate only started to occur since 2000, with the primary reasons being an increase in fishers and intrusion by purse seine vessels into the traditional fishing zone. The number of registered seine nets in Kudat has increased since 1992 (A.Cabanban, unpubl. data), thus lending credibility to the fishers' allegations. Regression analysis (Fig. 3.7) suggested a declining trend in catch rates, thereby supporting fishers' assertion about a temporal decrease in catch rates. However, the regression had a low R 2 value; this was possibly due to the wide spread in catch rates recorded for each year, as well as extreme values at either end of the time scale. One of the impediments of using information derived from fisher interviews is that it is subject to memory loss (Neis et al, 1999). This might partially explain the lack of recorded past catches that were less than 10 kg per fisher per day in Figure 3.7. Knowledge tends to be unevenly distributed among the fishers, in general being more concentrated among older fishers (Neis et al, 1999). Therefore the makeup of my respondent pool might have affected the outcome of this question in that interviewing a larger proportion of young or less experienced fishers could have led to more recent years being mentioned. However, this was not the case as the oldest fishers in fact identified 2000 as the year catches started to decline. The fact that all the fishers with long fishing experience singled out the last five years as the period of decreasing catches signified the severity of the potential decline, and should be taken as a serious alert by fisheries resource managers. In particular, the influx of purse seines in traditional fishing grounds has the potential to adversely affect small scale fisheries, 77 as happened with the traditional troll and gillnet fishery for small tunas on the east coast of Malaysia (Chee, 1995). The coral grouper fishery was another cause for concern. Fishers' statements regarding the decline in size and catch rate of Plectropomus spp. in southern Banggi paralleled a study by Daw et al. (2002) that reported about 50 to 90% reductions in yields of L R F T species in Kudat, and indicated that the stock may be biologically overfished. Biological overfishing had already depleted the population of another valued L R F T species, the humphead wrasse (Cheilinus undulatus), throughout the northern region of Sabah (CITES, 1994). On a local scale, the decline in Plectropomus spp. stocks had forced some fishers from southern Banggi to follow 'piskadul' vessels (mother-dory operations, where a large vessel tows several pump boats) to reefs further north in Sibogo and Tigabu, where it was supposed to be easier to catch coral groupers than in the Maliangin area. This suggested a spatial serial depletion of coral groupers, yet another indication of the lack of sustainability of the live fish trade. Kudat is one of the major live fish trade centres in Sabah; estimates of the live fish trade volume range from 70 to around 300 tonnes per year (Daw, 2002). Most of the live fish from Kudat are exported to international markets such as Hong Kong, Singapore, and Taiwan, with some being consumed within Malaysia (Daw, 2002). Regionally, increasing demand for live fish has already led to overfishing and depletion of coral grouper populations in parts of the Philippines and Indonesia (Bentley, 1999). A t the same time, the practice of poison fishing to supply the L R F T has grown in synchrony with exploitation rates since the 1990s (DeVantier et al, 2004), putting the physical health of coral reefs in jeopardy. Warren-Rhodes et al. (2003) showed that the live trade of groupers sold up to four times the estimated sustainable yield of Southeast Asian reefs. Moreover, the L R F T imposed a net welfare loss to the exporting country (Warren-Rhodes et al, 2003), and particularly to the individual small scale fishers, who not only received a minute portion of profits from the trade, but also suffered the consequences of present and future lost fisheries production and degraded reef habitats. Clearly, the targeting of coral grouper in Banggi seems to be on an unsustainable path. However, given Kudat's importance in the live fish trade in Sabah, 78 demand for delivery of live fish is not likely to drop until wi ld stocks become thoroughly depleted. 3.5.4 Management implications The seasonal difference in catches, effort distribution, and target species implied that fish stocks within the study area were not subject to the same fishing pressure year round. Naturally arising constraints such as weather and fish movement, as well as economic incentives compelled fishers to adopt varying fishing strategies during different times of the year. The spatial and temporal patterns presented in this paper have the following implications for management decision making: 1) Declining catch rates and fish sizes, particularly for the coral grouper fishery, indicate the need for catch, size, and or spatial limits. However, a conventional top-down approach w i l l l ikely not be effective in enforcing these regulations. Instead, self-regulation achieved via a participatory process in which fishers are involved in the planning, implementation, and enforcement of such policies should be adopted. This community-based approach has proven to be successful at Apo Island in the Philippines (Alcala, 1998). 2) The effects of zoning (i.e., spatial closures) should be considered within the context of gear use. Since hook and line fishers mostly concentrate their effort in one location, they w i l l be more drastically affected by spatial closures compared to gillnet fishers, who have more spatially dispersed fishing sites; 3) That overall fishing effort remains fairly constant throughout the year reflects the importance of Banggi's coastal reef fishery as an income source, and also the lack of alternative full time employment. A n y policy that restricts fishers' ability to fish should therefore have to be complemented by an alternative income generation scheme; 4) The impact of commercial purse seine operations intruding into the traditional fishing grounds should not be taken lightly, as it has substantial negative effects on both the fisheries resource, as well as economic state of the artisanal fisheries. 79 3.6 Concluding remarks This study outlined the main seasonal characteristics of Banggi's coastal reef fisheries, in terms o f catch volume, species composition, fishing effort, and economic benefits to fishers. It also shed some light on what, where, when, and why fishers in southern Banggi fish. One of the important findings of this study was that Banggi fishers have experienced substantial decline in catch rates over the past 20 years, and particularly in the past 5 years. On a positive note, Banggi's reefs did not seem to have been subject to the same intensity of fishing as other Southeast Asian reefs in the past. Subsequently, there is a possibility that productivity can be maintained or improved i f actions are taken to alleviate and minimize current drivers of overexploitation now. The lack of official records to substantiate the temporal decline in catches underlines the importance of undertaking precautionary management under data poor conditions (Johannes, 1998; Sadovy and Cheung, 2003). The applicability of this study's findings to local reef resource management within the context of a Marine Protected Area (the proposed Tun Mustapha Park) can be enhanced by future work on i) establishing long term monitoring of catch landings and effort; ii) quantitative spatial analysis o f effort allocation (e.g. Pet Soede et ai, 2001); i i i) investigating variables affecting the behaviour of artisanal fishers, for example see Salas et al. (2004); and iv) biological studies of abundance, movement, and reproductive cycles of local target fish populations. 80 3.7 References Alcala, A.C., 1998. Community-based coastal resource management in the Philippines: a case study. Ocean Coast. Manage. 38, 179-186. Alcala, A.C., Luchavez, T., 1982. Fish yield of the coral reef surrounding Apo Island, Negros Oriental, Central Visayas, Philippines. In: Gomez, E.D., Birkeland, C.E., Buddemeier, R.W., Johannes, R.E., Marsh, J.A. Jr., Tsuda, R.T. (Eds.), Proceedings of the 4th International Coral Reef Symposium. Marine Science Center, University of the Philippines, Manila, pp.69-73 Alcala, A.C., Russ, G.R., 2002. Status of Philippine Coral Reef Fisheries. Asian Fish. Sci. 15, 177-192. Amar, E.C., Cheong, R.M.T., Cheong, M.V.T., 1996. Small-scale fisheries of coral reefs and the need for community-based resource management in Malalison Island, Philippines. Fish. Res. 25, 265-277. Anon., 1994. Yearbook of Statistics Malaysia 1993. Anon., 2003. Profail Daerah Kecil Banggi Tahun 2003. Pejabat Daerah Banggi. Unpublished report. 25 pp. Bentley, N . , 1999. Fishing for solutions: can the live trade in wild groupers and wrasses from Southeast Asia be managed? TRAFFIC Southeast Asia, Petaling Jaya. 100 pp. Biusing, R.E., 2001. Assessment of coastal fisheries in the Malaysian-Sabah portion of the Sulu-Sulawesi Marine Ecoregion. WWF Malayisa, Kota Kinabalu, 719 pp. 81 Buckworth, R . C . , Clarke, R., 2001. Fishery Assessment Report for the Northern Territory Spanish Mackerel Fishery 1999, Rep. No . 52. Northern Territory Department of Primary Industry and Fisheries. Cabanban, A . S . , Biusing, R . E . , 1999. Coral reef fisheries and their contribution to marine fish production in Sabah, Malaysia. In Cabanban, A . S . , Phillips, M . (Eds), Workshop on Aquaculture o f Coral Reef Fishes and Sustainable Reef Fisheries, Institute of Development Studies, Kota Kinabalu, Sabah, pp. 87-96 Chee, P .E. , 1995. Tuna Fisheries Interactions in Malaysia. In: Shomura, R.S. , Majkowski, J., Harman, R.F . (Eds), 2nd F A O Expert Consultation on Interactions of Pacific Tuna Fisheries. F A O , Shimizu, Japan. CITES , 2004. 13th meeting of the Conference of the Parties. Amendments to Appendices I & II of CITES . Available at http://international.fws.,gov/pdfcopl3/Cheilinus%undulatus%20proposal.pdf. Claereboudt, M . R . G . , Mc l lwa in , J., A l -Ouf i , H.S. , A m b u - A l i , A . A . , 2005. Patterns of reproduction and spawning of the kingfish (Scomberomorus commerson Lacepede) in the coastal waters of the Sultanate of Oman. Fish. Res. 73, 273-282. Collette, B . B . , Nauen, C .E . , 1983. F A O Species Catalogue. Vol .2 . Scombrids of the world. A n annotated and illustrated catalogue of tunas, mackerels, bonitos and related species known to date. F A O . Dalzel l , P., Adams, T.J .H. , Polunin, N . V . C . , 1996. Coastal Fisheries in the Pacific Islands. Oceanogr. Mar. B i o l . , 34, 395-531. Daw, T., 2004. Reef Fish Aggregations in Sabah, East Malaysia. Western Pacific Fisher Survey Series: Volume 5, Society for the Conservation of Reef Fish Aggregations. 59 pp. 82 Daw, T., Jeffrey, L . , B i n A l i , M . A . , 2002. Preliminary Assessment of the Live Reef Fish Trade in the Kudat Region. W W F Malaysia, Kota Kinabalu, 47 pp. DeVantier, L . , Alcala , A . C . , Wilkinson, C , 2004. The Sum-Sulawesi Sea: Environmental and Socioeconomic Status, Future Prognosis and Ameliorative Policy Options. Ambio 33, 88-97. Institut Penyelidikan Mar in Borneo, 2003. Laporan Kajian Mengenai Pengetahuan dan Pandangan Masyarakat Pulau Banggi Terhadap Penubuhan Taman Tun Mustapha. Unpublished report. 36 pp. Jennings, S., Lock, J . M . , 1996. Population and ecosystem effects of fishing. In: Polunin, N . V . C . , Roberts, C M . (Eds.), Reef fisheries. Chapman and Ha l l , London, pp. 193-218. Johannes, R .E . , 1981. Words of the Lagoon. University of California Press, Berkley, C A . Johannes, R . E . , 1998. The case for data-less marine resource management: examples from tropical nearshore fisheries. T R E E 13, 243-246. Johannes, R .E . , Freeman, M . M . R . , Hamilton, R.J . , 2000. Ignore fishers' knowledge and miss the boat. Fish Fish. 1, 257-271. Kitalong, A . H . , Dalzel l , P., 1994. A preliminary assessment of the status of inshore coral reef fish stocks in Palau. Technical Document No . 6. South Pacific Commission, Noumea, New Caledonia. McManus, J.W., 1997. Tropical marine fisheries and the future of coral reefs: a brief review with emphasis on Southeast Asia . Coral Reefs 16, S121-S127. 83 Neis, B . , Schneider, D . C . , Felt, L . , Haedrich, R . C . , Fischer, J., Hutchings, J .A. , 1999. Fisheries assessment: What can be learned from interviewing resource users? Can. J. Fish. Aquat. Sci . 56, 1949-1963. Pauly, D . , Christensen, V . , Dalsgaard, J., Froese, R., Torres, F.J. , 1998. Fishing Down Marine Food Webs. Science 279, 860-863. Pauly, D . , Navaluna, N . A . , 1983. Monsoon-Induced Seasonality in the Recruitment of Philippine Fishes. In: Proceedings of the Expert Consultation to Examine Changes in Abundance and Species Composition of Neritic Fish Resources, F A O Fisheries Report No. 291. Pet-Soede, C , Machiels, M . A . M . , Stam, M . A . , van Densen, W . L . T . , 1999. Trends in an Indonesian coastal fishery based on catch and effort statistics and implications for the perception of the state of the stocks by fisheries officials. Fish. Res. 42, 41-56. Pet-Soede, C , van Densen, W . L . T . , Hiddink, J .G. , K u y l , S., Machiels, M . A . M . , 2001. Can fishers allocate their fishing effort in space and time on the basis of their catch rates? A n example from Spermonde Archipelago, S W Sulawesi, Indonesia. Fisheries Manag. Ecol . 8, 15-36. Poizat, G . , Baran, E . , 1997. Fishermen's knowledge as background information in tropical fish ecology: a quantitative comparison with fish sampling results. Environ. B i o l . Fish 50, 435-449. Reason, P., 1994. Three Approaches to Participative Inquiry. In: Denzin, N . K . , Lincoln, Y . S . (Eds.), Handbook of Qualitative Research. Sage Publications, Thousand Oaks, California, pp. 324-337. Sadovy, Y . , Cheung, W . L . , 2003. Near extinction of a highly fecund fish: the one that nearly got away. Fish Fish. 4, 86-99. 84 Salas, S., Sumaila, U . R . , Pitcher, T., 2004. Short term decisions of small-scale fishers selecting alternative target species: a choice model. Can. J. Fish. Aquat. Sci . 61, 374-383. Teh, L . , Cabanban, A . S . , Sumaila, U .R . , 2005. The reef fisheries of Pulau Banggi, Sabah: A preliminary profile and assessment of ecological and socio-economic sustainability. Fish. Res. 76, 359-367. Underwood, A . J . , 1981. Techniques of analysis of variance in marine biology and ecology. Oceanogr. Mar. B i o l . Ann . Rev. 19, 513-605. Warren-Rhodes, K . , Sadovy, Y . , Cesar, H . , 2003. Marine Ecosystem Appropriation in the Indo-Pacific: A Case Study of the Live Reef Fish Food Trade. Ambio 32, 481-488. 85 C H A P T E R 4 E X P O L O R I N G M A N A G E M E N T S T R A T E G I E S F O R T H E R E E F F I S H E R I E S O F P U L A U B A N G G I , S A B A H 4.1 Introduction 4.1.1 Background Fishing on coral reefs leads to some well known effects, among them, changes in the abundance of target species (Russ and Alcala , 1989), changes in size and composition of fishery yields (Jennings and Lock, 1996), and possible alteration of trophic structure (Jennings and Lock, 1996) or shifts in reef community structure (McManus et al, 2000). A n understanding of direct and indirect fishing effects is essential for assessing the sustainability of fishing practices, and for determining the ecological consequences of fishing policies (Jennings and Polunin, 1996). This is particularly relevant to the often overexploited reef fisheries of many developing countries, where the feasibility of many fishing policies tends to be limited, and also where policies directly affect the socio-economic well being of fishing dependent coastal communities. In this study, I use the ecosystem modelling tool Ecopath with Ecosim (EwE) to do a preliminary analysis of the effects of fishing on a coral reef ecosystem in Banggi, an island off northern Sabah, Malaysia. I first speculate on the probable changes that historical fishing patterns might have brought upon the ecosystem. I then explore simulations of potential future management policies, including: i) maintaining the status quo; ii) increasing effort for five years then closing the fishery; iii) increasing effort for five years then stabilizing effort thereafter; and iv) decreasing effort for five years then stabilizing effort thereafter. The model presented in this chapter is simple due to the lack of local biological and fisheries time series data. However, I consider this preliminary model sufficient for the purpose of investigating broad decadal trends in fishing impacts, and their consequent implications for the future direction of management. I then apply the model results within Banggi's social, economic, and political context (Chapter 1 and Teh et al, 2005) to assess the possibility of 86 establishing a community-based marine reserve as a means of achieving sustainable reef fisheries management. 4.1.2 Study site Pulau Banggi is the largest island in Malaysia, and is located off the northern tip of Sabah, on the island of Borneo (see chapter 2, Fig.2.1). Banggi is separated from mainland Sabah by the Banggi Channel, and is surrounded by approximately 50 smaller islands with a combined area o f 700 k m 2 . The region is characterized by shallow seas, with strong tidal currents and turbid waters occurring near the coast. Figure 4.1 Map of southern Banggi, showing Maliangin fishing ground. Reef surveys conducted by Greenforce were carried out within this area as well. The model area covers approximately 17 km of southern Banggi (Fig. 4.1), where the Maliangin fishing ground (see chapters 2 and 3) is located. Reef surveys carried out in southern Banggi found a few intact reefs with good coral cover ranging from 50% to 75% (Koh et al, 2002). However, the majority of coral reefs in Banggi were considered "poor" (Harding et al, 2001). There was a lack of large, predatory fish species observed during 87 surveys, with the reef ecosystem being dominated by herbivores (Harding et al, 2001; K o h et al, 2002). 4.1.3 Reeffisheries of Banggi Small scale, artisanal reef fisheries operate in southern Banggi. In this study, reef fisheries refer to the capture of both demersal reef fish and reef-associated, coastal species, notably scombrids and carangids. The main gears used are hook and line, gillnet, trap, and spear. Longlines are used by outside fishers who come to Banggi from other parts of Sabah (see Ch.2 and Teh et al, 2005). Destructive fishing techniques using bombs and cyanide have caused extensive damage to the island's coral reefs (Harding et al, 2001; K o h et al, 2002). Fishers target a wide variety o f demersal reef fish, as well as reef associated and pelagic fish such as carangids and scombrids. The most valuable target fish are Spanish mackerel (Scomberomorus commerson), coral grouper (Plectropomus spp.), and trevallies (Carangidae). A more detailed description of Banggi's reef fisheries can be found in chapters 1 and 2, and in Teh et al, (2005). There is currently no active management of Banggi's artisanal fisheries. 4.2 Methods 4.2.1 Modelling approach The Ecopath with Ecosim (EwE) modelling approach (Pauly et al, 2000) was used to simulate the coral reef ecosystem of southern Banggi. Theoretical concepts and procedures related to E w E can be found in Christensen et al (2004). Application of E w E to coral reef ecosystems has been done by various authors, for example Opitz (1993), Arias-Gonzalez et al (1997), Tudman (2001), Gribble (2003 and 2001), and Arias-Gonzalez et al (2004). Given that E w E is a well-established ecosystem modelling framework that has been applied extensively, I proceed with a description of the specific application here. 4.2.2 2005 model A mass balanced model reflecting the 2005 state of the southern Banggi reef ecosystem was built using Ecopath. The model was based on reef survey data collected by Greenforce, a 88 non-profit non-governmental organization, using the Reef Check method (www.reefcheck.org) over a period of fifteen months from July 1999 to September 2000 (Harding et al, 2001). The coral reef sites surveyed by Greenforce included: 1) shallow fringing reefs reaching a depth of 10-15 metres; 2) steep rocky slopes with coral cover on high islands in south-west Banggi; and 3) patch and bank reefs in the Banggi Channel. Corals of the latter two reef types extended to 20-25 metres depth. Sites that were not included in the surveys were certain deep reefs that were frequently fished (pers. obs.), some of which extended beyond 30 metres depth. The model consisted of 33 functional groups; that is, 30 consumer groups (22 fish groups), two producer groups, and a detritus group (Table 4.1). The producer groups included phytoplankton and benthic producers. Benthic producers covered macroalgae, seaweed, and seagrasses. The heterotrophic and autotrophic components o f corals were not separated, and corals were treated as a consumer group. A s the purpose o f this model was to examine the effects of fishing, fish groups were formed according to their importance to reef fisheries, and also according to the availability of reef survey data. The remaining fish were grouped based on similarities in diet and feeding habits, habitat, and body size. Fisheries landings were estimated based on catch landing surveys carried out on site during two field sampling periods in June/August 2004 and March/Apr i l 2005 (see chapters 2 and 3, and Teh et al, 2005). The model incorporated landings from five fisheries - hook and line, gillnet, trap, longline, and cuttlefish.1 Hook and line and gillnet were the two gears that accounted for the majority of landings (Chapter 2). The estimated annual yield from the southern Banggi fishing grounds was approximately 25 t-km"2 (Table 4.2). Economic value for each fishery referred to the landed value of catches, and was based on market prices 2 o f fish recorded during the two field sampling periods (see chapter 3, Table 3.5). 1 Fish landings data were collected only for the hook and line and gillnet fisheries. The other fisheries landings were estimated at a coarser scale as they were based on infrequent personal observations or receipts. A small spear fishery was not included explicitly due to lack of information on which to base estimates. 2 The monetary unit used is the Malaysian Ringgit (RM). The exchange rate on Dec.20, 2005 is 1 USD $ 1 = 3.78 R M 89 Table 4.1 Input parameters for 2005 model. Input parameters estimated by Ecopath are in italics. Group No. Group Name Biomass P/B Q/B E E (t km"2) (year 1) (year 1) 1 Sharks 0.20 0.25 1.25 0.73 2 Large carangids and scombrids 3.19 0.70 3.50 0.95 3 Medium carangids and scombrids 8.12 1.63 8.15 0.48 4 Small pelagics 15.51 3.00 15.00 0.95 5 Benthopelagic feeders 0.45 2.00 10.00 0.84 6 Demersal zoobenthos feeders 0.49 2.00 10.00 0.95 7 Labridae 19.36 1.51 7.55 0.27 8 Lutjanidae 6.77 0.70 3.50 0.35 9 Plectropomus spp. 0.75 0.44 2.20 0.95 10 Other Serranidae 2.49 0.65 3.25 0.70 11 Caesionidae 60.73 2.76 10.00 0.04 12 Chaetodontidae 0.51 1.9 11.00 0.59 13 Pomacentridae 7.55 3.80 30.00 0.31 14 Pomacanthidae 3.43 1.60 12.80 0.41 15 Mullidae 2.25 1.92 8.50 0.17 16 Lethrinidae 3.13 0.64 5.10 0.95 17 Haemulidae 1.50 0.57 4.80 0.58 18 Scaridae 15.34 1.05 26.43 0.25 19 Siganidae 0.61 1.82 32.80 0.41 20 Acanthuridae 1.12 1.30 19.70 0.65 21 Balistidae 0.32 2.15 10.75 0.31 22 Nemipteridae 0.99 1.41 11.50 0.90 23 Large benthic invterbrates 43.97 4.00 21.90 0.85 24 Mollusks/worms 102.59 4.50 22.00 0.95 25 Cephalopods 1.30 3.10 11.70 0.77 26 Sea cucumbers and sea urchins 1.80 5.98 23.63 0.90 27 Other echinoderms 6.44 1.30 15.00 0.90 28 Zoobenthos 55.63 10.00 50.00 0.95 29 Zooplankton 33.21 67.00 280.00 0.90 30 Corals 100.00 1.20 7.40 0.54 31 Benthic autotrophs 250.00 15.35 - 0.19 32 Phytoplankton 16.96 450.00 - 0.95 33 Detritus 150.00 - - 0.73 Other biological parameters such as P /B (production/biomass ratio), Q/B (consumption/biomass ratio), and diet were initially based on literature (e.g., A l ino et al, 1993; Opitz, 1993; Tudman, 2001; Garces et al, 2003) or estimated from Fishbase (www.fishbase.org). Some of these values were subsequently adjusted in order to balance the model. See Table 4.1 and 4.2 for input parameters and Appendix V I for diet matrix. Table 4.2 Estimated annual fishery landings by gear and species groups (t-krn2yr"') Group Name Gillnet H o o k & Traps Longline Cuttlefish Total Line Sharks 0.01 0.02 - - 0.03 Large carangids and scombrids 1.30 0.76 - 0.02 2.08 Medium carangids and scombrids 1.14 0.92 - - 2.06 Small pelagics* - 11.88 - - 11.88 Benthopelagic feeders 0.18 0.04 0.18 - 0.40 Demersal zoobenthos feeders 0.16 0.04 0.20 0.01 0.41 Labridae - 0.19 - - 0.19 Lutjanidae 0.56 0.30 0.02 0.03 0.90 Plectropomus spp. - 0.17 0.10 0.03 0.30 Other Serranidae 0.05 0.96 0.04 0.04 1.09 Caesionidae - 0.06 0.61 - 0.67 Chaetodontidae - - 0.04 - 0.04 Pomacanthidae - - 0.04 - 0.04 Mullidae 0.06 - - - 0.06 Lethrinidae 0.06 0.64 0.02 0.01 1.41 Haemulidae 0.74 0.05 0.08 0.01 0.48 Scaridae 0.33 0.02 0.04 - 0.11 Siganidae 0.05 - 0.04 - 0.04 Acanthuridae - 0.04 0.31 - 0.34 Balistidae - 0.07 - - 0.07 Nemipteridae 0.02 0.37 0.31 - 0.69 Large benthic invterbrates 0.26 - - - 0.26 Cephalopods 0.03 0.03 - 1.23 1.29 Total 4.90 2.04 16.55 0.16 1.23 24.87 * The high landings of small pelag ics by hook and line consist of ox -eye scad, which are cauj ;ht during the night during windy periods (see chapter 2). 91 Ecosim, the dynamic component of E w E , was used to simulate the impact of four potential management policies on the ecosystem. A main uncertainty about using E w E to model ecosystems is the type of flow dynamics, simulated via a 'vulnerability' variable, which occurs between species (Walters et al., 1997). In this model I set the vulnerabilities to a default of 2 to reflect a mixture of top-down and bottom-up controls, which I am aware is an oversimplification of the true ecosystem trophic interactions. Policy simulations were run for 25 years from the 2005 start time. Given the current lack of active management of Banggi's reef fisheries, the following two policies were considered to be plausible: 1. The status quo was maintained; i.e., no management was imposed, and fishing effort continued to increase linearly as per historic rates. Three scenarios using estimated historic rates of increase of 200%, 250%, and 300% between 1980 and 2005 were simulated (equivalent to annual rates o f 4%, 6%, and 8%, respectively). These rates were selected based on information obtained from interviews with local fishers (Chaper 3); 2. Fishing effort increased as per historic rate from 2005 to 2010, after which fishing stopped due to the establishment of a marine reserve. A s in Pol icy 1, three scenarios using the estimated increase rates of 4%, 6%, and 8% per year were simulated. Under Section IV of the Malaysian Fisheries Act (1985), it is an offence to fish in any marine park or marine reserve within Malaysian waters. This top-down policy is likely i f southern Banggi follows in the footsteps of other Marine Parks in Peninsular Malaysia and Sabah, where no-fishing zones have been established in previously active fishing grounds. It should be noted that management of marine parks in Sabah is hampered by a lack of enforcement resources (both personnel and financial) and commitment, and exacerbated by local fishers' indifference to fishing boundaries (Pilcher and Cabanban, 2000). In addition, two alternative policies that involved effort restrictions were investigated to provide a means of comparison to policies 1 and 2. 92 3. Fishing effort continued to increase as per historic rate scenarios (i.e. 4%, 6%, and 8% per year), between 2005 and 2010. Thereafter, effort limiting mechanisms were imposed, and effort was maintained at a constant level until the end of the simulation period (2030); 4. Fishing effort decreased from 2005 to 2010, after which it was maintained at the 2010 level until the end of the simulation period (2030). Two scenarios with 25% and 50% rates of decrease between 2005 and 2010 were run (i.e., 5% and 10% per year, respectively). These policies were assessed and compared with regard to changes in species biomass, catches, and catch value, and then qualitatively evaluated in light o f the socio-economic and political context of the local fishing communities. 4.2.3 1980 model A second Ecopath model assumed to approximate the 1980 system was also built. A s no data were available for the 1980 model, input parameters were largely derived from the 2005 model based on an approach described in the below section. Ecosim was then used to investigate the effect of fisheries development (changing effort levels) on the important fishery target groups. The approach taken to build the 1980 model was to try to match, via iterated simulation runs, the end biomasses and catch levels of selected fish groups in the 1980 model as closely to those in the 2005 balanced model as possible. This was done by varying three parameters -fishing effort, catch, and species biomass. For simplicity, other parameters such as P / B , Q /B , ecotrophic efficiency (EE), and vulnerability were assumed to remain constant, although it was likely that these parameters would have changed over time i f the ecosystem underwent some change. I did several runs of the 1980 model based on a priori, assumed levels of effort and catch. Given these effort and catch rates, biomass was adjusted in order to produce ending biomass and catch levels that were compatible with the 2005 starting levels. The groups of primary focus were those targeted by the reef fisheries, including the large carangids and scombrids (LCS) , Plectropomus spp, other serranids, lethrinids, haemulids, 93 lutjanids, and medium carangids and scombrids ( M C S ) . In order to achieve the desired compatability to the 2005 model, the following assumptions were made regarding the 1980 model: Biomass Interviews with fishers suggested that the abundance of all fished species had decreased from the past. The biomass of all target groups was therefore assumed to be larger in 1980 than 2005. In particular, Plectropomus biomass was assumed to be 3.2 t-km"2, which resembles a lightly fished Plectropomus population in the central Great Barrier Reef (Tudman, 2001). This was reasonable since high demand and targeting of live fish did not start until the late 1980s and early 1990s in Sabah. Catches Total catches of all groups, with the exception of L C S and Plectropomus, were assumed to be less in 1980 than 2005. This is justified considering the artisanal nature of southern Banggi's reef fisheries; therefore presumably, the lower population in Banggi during the 1980s would have corresponded to a lower number of fishers. Three different levels of catches were investigated: 25%, 50%, and 75% of 2005 catch levels. L C S and Plectropomus catches were assumed to be at least equal to 2005 catch levels. This assumption was made because of their relatively higher importance as the key target species for hook and line fishers (Chapter 2 and 3). Fishers indicated that catchability and individual catch rates o f these two species was a lot higher in the 1980s (Chapter 3), even more so than the presumed increase in fishing effort (therefore total catches might have remained approximately equal). The catches of L C S were assumed to be equal to 2005 catches, while Plectropomus catches were assumed to be 25% higher. These catch levels were determined after many runs in Ecosim showed that lower catch levels, in conjunction with the assumed effort levels, would not yield an ending catch quantity that was comparable to the 2005 starting level. Effort Fishing effort levels were assumed to have increased linearly between 1980 and 2005. Effort increases of 2, 2.5, and 3 times in magnitude were tested (i.e., per annum rates of increase of 94 4%, 6%, and 8%). This effort level applied to all fish groups except Plectropomus, for which these effort levels were too low. Instead, overall Plectropomus fishing mortality was assumed to have increased substantially from 1990-2005 (approximately 15% per year), reflecting the development of the live fish trade, which in Sabah started to intensify in the late 1980s and early 1990s (Daw et al, 2002). Changes in fishing effort were applied to three fisheries -gillnet and hook and line, which were the most widely used gears (Chapter 1), and trap. Two other fisheries, longline and cuttlefish 3, were assumed to have remained constant throughout the simulation period due to lack of information regarding their trend over time. 4.3 Results 4.3.11980 model Due to the coarse nature o f the 1980 model, results are, in most cases, not reported in absolute values, but rather in relative terms only, as they pertain to the assumptions made in Section 4.3.2. Under these assumptions, the exercise suggested that: Biomass Biomass levels for those fish groups which were both heavily targeted and had high fishing mortalities in 2005, namely Serranidae, Lethrinidae, and Haemulidae were approximately 50% higher in 1980, while the biomass of L C S was approximately 7 t-km"2 (Table 4.3). The biomass of heavily targeted groups was increased by 150% because that was the lowest level Table 4.3 Biomass, catches, and effort levels for 2005 and 1980 (derived) ecosystems. LCS = large carangids and scombrids; M C S = medium carangids and scombrids. Parameter 2005 1980 Change (t-km2) (t-km2) (%) Biomass: Plectropomus spp. 0.75 3.2 430 LCS 3.19 7.0 220 Serranidae 2.49 3.74 150 Haemulidae 1.50 2.25 150 Lethrinidae 3.13 4.70 150 Lutjanidae 6.77 6.77 No change MCS 8.12 8.12 No change Catches 26.2 13.3-18.8 50-75 Effort increase (1980-2005) n/a 2-3 200-300 3 Cuttlefish are caught with spears while diving, trident-like spears from above water, or with gillnets. 95 which would generate results comparable to 2005 biomasses and catches, given the pre-determined effort and catch rates. The biomass levels of two other important fishery groups, Lutjanidae and M C S , were assumed to be equal to the 2005 level given that their fishing mortalities were low in the 2005 model, and thus not very sensitive to fishing pressure. L C S biomass of 7 t-km"2 was arrived at after numerous runs in Ecosim showed that in order for the ending biomass and catch of L C S to be within reasonable range of 2005 levels under the assumed historic exploitation rates and catches, the fishing mortality of L C S had to be approximately 0.4 of total mortality, with ecotrophic efficiency (EE) of around 0.6. L C S biomass was subsequently increased from a minimum of 3 tkm" z (2005 balanced level) until these criteria were met, with the result of 7 t-km" . 4.3.2 2005 model Policy 1: Status quo Continued increase o f fishing effort resulted in decreasing biomass and catches of main target groups. For example, Fig.4.2 shows the simulation result for the two time increase in 0.5 • 0.0 r , , , r 1 6 11 16 21 26 Simulat ion year Figure 4.2 Policy 1: Status quo, showing a two time increase in fishing effort over 25 years (4% increase per year). LCS=large carangids and scombrids; MCS=medium carangids and scombrids. 96 effort scenario. Other scenarios followed the same general trend over time, and are not shown. LCS experienced the greatest and fastest decline, followed by Lethrinidae, Plectropomus, and Serranidae. Under the two time effort increase scenario, LCS was 'extinct', while Plectropomus, Lethrinidae, Serranidae, and Haemulidae were close to being fished out at the end of the simulation period. The catches of all target fish groups decreased to a small quantity by 2030 (Table 4.4), resulting in negligible economic value for these target groups. In contrast, MCS and Lutjanidae increased in biomass, and catches and values of these two groups increased. Economic values fell by up to 50%, 80%, and over 95% for the gillnet, hook and line, and trap fisheries, respectively (Table 4.5). The large value decreases were due to reduced catches of the most valuable species (LCS and Plectropomus spp.). Policy 2: Effort increase for five years, thereafter no fishing The biomass of all groups declined as fishing effort increased for the first five years (Fig.4.3), leading to corresponding decreases in catches. When fishing ceased, the biomass of all t 6 II 16 21 26 Simulation year Figure 4.3 Policy 2: Effort increase as per historic rate (4% per year in this figure), after which fishing ceases. LCS=large carangids and scombrids; MCS=medium carangids and scombrids. 97 groups increased immediately. This rapid recovery should be interpreted with caution, and will be discussed in more detail later. LCS and Plectropomus exhibited the sharpest increases, followed by Lethrinidae and Serranidae. Policy 3: Effort increase for five years, thereafter stable effort Biomass of targeted groups decreased as fishing effort continued to increase for the first five years (Fig.4.4). As under Policy 2, LCS, Plectropomus, Lethrinidae, and Serranidae experienced the biggest declines. After fishing effort was maintained at the 2010 level, biomass for all groups levelled out, although none of the species experienced a recovery. Catches of all target groups declined by the end of the simulation period except for MCS and Lutjanidae, which increased (Table 4.4). Economic values for all three fisheries declined by 20 to 40% under the 2 time effort scenario, by 40 to 50% under the 2.5 time effort scenario, and up to 70% under the 3 time effort scenario (Table 4.5). 1.5 Figure 4.4 Policy 3: Historic rate of effort increase (here, 4% per year) for 5 years, then stable effort thereafter. LCS=large carangids and scombrids; MCS=medium carangids and scombrids. 98 Policy 4: Effort decrease for five years, thereafter stable effort All groups started to recover slightly as fishing effort decreased between year one and five. After effort was held level at year five, biomass for LCS, Plectropomus, and Lethrinidae continued to recover before levelling out around year 12, at levels above that in 2005 (Fig.4.5). A decline in fishing effort from years one to five resulted in increased catches of LCS and Plectropomus (Table 4.4). In contrast, catches of Serranidae, Lethrinidae, and Haemulidae responded less to the decline in fishing effort, showing only minor changes from the starting level (Table 4.4). Values of all three fisheries rose slightly following a decrease in fishing effort due to the increased catches of valuable LCS and Plectropomus (Table 4.5) Simulation year Figure 4.5 Policy 4: Effort declines from year one to five, followed by stable effort thereafter. This shows an effort decline of 25%. LCS=large carangids and scombrids; MCS=medium carangids and scombrids. Overall, policies that imposed a restriction on fishing effort, such as Policy 2 and 4, were able to increase biomass of target fishery groups by the end of the simulation period (Table 4.6). As well, Policy 4 generated increased economic benefits to the fishers, while Policy 2 could potentially do so once non-market and non-consumptive values were considered (Table 4.4 and Section 5.1). 99 Table 4.4 Catch results under each policy (scenario indicated after policy number). Results are given as a percentage of the 2005 starting catches. L C S Plec Serr Leth Haem Lutj M C S Policy* Scenario (% of 2005 starting levels) 1 200% 0 10 20 10 40 210 160 1 250% 0 0 0 0 20 240 150 1 300% 0 0 0 0 0 250 130 2 Al l - - - - - - -3 200% 50 60 80 70 90 30 120 3 250% 30 40 70 50 90 140 130 3 300% 10 30 50 30 70 40 50 4 25% 130 130 100 110 90 70 80 4 50% 140 110 80 90 70 50 60 *Policy 1: status quo; Policy 2: effort increase for five years, thereafter no fishing; Policy 3: effort increase for five years, thereafter stable effort; Policy 4: effort decrease for five years, thereafter stable effort; Table 4.5 End state economic value* for gillnet, hook and line, and trap fishery under each policy (scenario indicated after policy number). Results are given as a percentage of the 2005 starting values: Gillnet = 12.53 10 3 RMkm" 2 ; Hook & Line = 33.27-103RMkm"2; Trap = 4.53 1 0 3 R M k m 2 . Policy Scenario Gillnet Hook Trap Average & Line all gears 1 200% 50 80 10 50 1 250% 50 20 0 20 1 300% 50 20 0 20 2 Al l - - - -3 200% 80 70 60 70 3 250% 60 50 50 50 3 300% 60 40 30 40 4 25% 110 120 120 120 4 50% 110 110 120 110 *Economic value refers to the landed value of catches expressed in RM-km"2, and includes only the selected target fishery groups. Prices remain constant for the duration of the simulation. T a b l e 4.6 Change in total end state biomass and economic values for target fishery groups, relative to starting levels, for each policy. The outcome column indicates which of biological or economic value is greater for each policy. Policy Biomass E c o n o m i c V a l u e Outcome 1 - - B>E 2 + - * B>E 3 - - B>E 4 + + B>E *Non-market and non-consumptive values are not incorporated. For example, future economic benefits derived from potential marine park related tourism development are not included in this analysis. 4.4 Discussion 4.4.11980 model The main finding derived from the 1980 model was that, given the a priori assumptions about fishing effort and catch rates, biomass of target fishery groups were probably higher in the 1980 ecosystem than in the 2005 ecosystem. Although the 1980 model was largely a modification of the 2005 model, the conclusions drawn from the model were nevertheless in broad agreement with available knowledge on general trends in the development of southern Banggi's reef fisheries. The main point illustrated by the 1980 model was that, given a relatively lightly exploited fishery in the 1980s, fishing effort could feasibly have increased two to three times in magnitude (4% to 8% per year) over the past 25 years, possibly almost doubling total catches by 2005. Given the small scale artisanal nature of Banggi's fisheries, this rate of increase is reasonable as it is comparable to the approximate growth in population in southern Banggi over the past 25 years.4 In particular, the model's high catches and intense fishing rate for Plectropomus was consistent with the intensification of the live fish trade starting in the 1990s (Daw et al, 2002). On the other hand, Cabanban and Bisuing (1999) suggested that total catches of coral reef fish in Sabah increased from 1980 to 1987, but declined from 1988 to 1993. This does not coincide with the model's trend for Banggi, which suggested increasing catches from 1980 to 4 No official population data for Banggi was available for 1980 and 2005. Boutin (1990) estimated that the population on Banggi was around 6000 in 1980. The current population of Banggi is approximately 18000 to 20000. 101 2005. The discrepancy might be due to the fact that Cabanban and Biusing based their analysis on data extracted from the Annual Fisheries Statistics of Sabah. These statistics likely did not capture the small-scale, subsistence and artisanal catches landed in remote sites like Banggi, therefore leading to an underestimate of total reef fish catches. In addition, their analysis only covered five reef fish families, and did not include pelagic species, which made up a substantial portion of catches in Banggi. A s Cabanban and Biusing's analysis did not go beyond 1993, it is not possible to speculate whether the five-year decline suggested by their study was a short term dip, or whether total catches continued on a decreasing trend from 1993 onwards. 4.4.2 2005 model The declining trend under Policy 1 came as no surprise, as all these target groups were already under high fishing pressure at the start of the simulation period, and any increase in fishing would expectedly have led to a further decline in biomass. However, an unexpected outcome was that the decline rate of L C S was greater than that of Plectropomus. This was not consistent with the life history o f large scombrids and carangids such as Spanish mackerel, trevallies, and tuna, whose life history traits make them more resilient to fishing pressure than Plectropomus (www.fishbase.org). This inconsistency could have resulted from a weakness of the underlying model. Therefore, the rates and magnitude of response of L C S to management simulations in Section 4 were most likely too extreme to be applicable for decision making. The high vulnerability of L C S to fishing could be attributed to the underlying assumption of the model study area. In its current form, the model treated all fish stocks as being essentially closed populations, and confined to the model area. This is likely true for groupers, which are fairly sedentary and have limited distribution (Zeller, 1997), and therefore not likely to move outside the model area. At the same time, the model was also potentially 'open' with regard to recruitment for demersal reef species (Sale, 1991 and 2002). The migratory large scombrids and carangids however, have a much wider range and distribution. For example, Spanish mackerel in Australia undertake large scale migrations along the east and west coasts of the country ( D E H Australia, 2004). The Banggi study area therefore likely constituted 102 only a small portion of the total distributional range for the L C S group. This aspect of their population dynamic was not captured by the model. In reality, the movement o f species into and out of the study area would likely decrease the observed rate and magnitude of decline. The Spanish mackerels' migration patterns might also partially explain the seasonality of the species' fishery (Chapter 2). The model suggested that all target fishery groups would be virtually extinct by 2030. Local extinction might be possible i f socio-economic conditions worsened on Banggi, and desperation drove fishers to carry out full fledged resource destruction by overfishing and indiscriminate use of explosives or other destructive fishing methods. On the other hand, as a species' biomass declined more and became harder, and therefore more costly, to catch, fishers would tend to seek new fishing grounds or divert their effort to catching other species that were relatively more abundant and easy to catch. A s such, the biomass of Plectropomus and other targeted groups would likely level out once reaching a low level, as fishers switched to targeting other species. Having said that, it is likely that a switching strategy would not be sustainable in the long term when the ecosystem becomes so depleted that there are no species of nutritional or economic value to target anymore. Policy 1 reflected reality under the current management regime where there is no active regulation to limit effort of Banggi's artisanal fisheries. Population growth and lack of alternative employment are both factors that w i l l l ikely contribute to increasing fishing effort within the Maliangin area. In the long run, this w i l l be to the detriment of the reef fish populations as well as to the livelihood of the fishers. Policy 2 demonstrated the anticipated recovery in fish biomass upon creation of a marine reserve (e.g., Roberts, 1995; Russ, 2002). In the model, all target fish groups rebounded quickly upon cessation of fishing. This was to be expected; however, the quick and large recovery rate of biomass was a result of model behaviour, and the overly optimistic outcome should be interpreted with caution. 103 The rate of biomass recovery in a marine reserve has been subject to conflicting findings. Roberts (1995) found that the biomass of commercially fished families increased 60% over a two-year period in a small marine reserve in the Caribbean Sea. After reviewing 80 different marine reserves, Halpern and Warner (2002) suggested that increases in average levels of fish density and biomass occurred within one to three years of reserve establishment. A t the same time, however, the authors also pointed out that species with slow growing and late maturing life histories were likely to respond much more slowly to reserve protection. In fact, based on continuous and long-term monitoring of two coral reef marine reserves, Russ and Alcala (2004) argued that full recovery of target predatory fish populations might take in the order of 15 to 40 years. Similarly, McClanahan (2000) suggested that recovery of an important predator on Kenyan coral reefs might take more than 30 years. Therefore, although eventual increase in exploited fish biomass could be expected, this should not be expected to happen within the short time frame indicated by the present model. Large predatory reef fish in general have maximum potential lifespans of 15 to 45 years (Choat and Robertson, 2002), and maximum biomass gain rates may not occur until fish are two to four years of age (Russ and Alcala , 1996). Therefore the simulation result that Plectropomus and Lethrinidae biomass might increase to three times their 2005 biomass level within eight years of reserve establishment was not realistic. The observed behaviour probably arose from the underlying assumptions about the trophic dynamics (expressed through the vulnerability parameter), and P /B ratios of Plectropomus and Lethrinidae. This aspect could be addressed in future iterations of the model when localized studies about the life history of target fish groups are carried out.5 The model's prediction of no catch or economic value upon cessation of fishing was clearly unacceptable from a socio-economic point of view. Closing off the entire Maliangin fishing ground would be detrimental to the coastal community, and is not recommended as such. Instead, only a small portion of the current fishing ground should be placed aside for protection. This issue would benefit from further investigation using Ecospace. However, I 5 Several undergraduate and post-graduate students from the Borneo Marine Research Institute at the University of Malaysia, Sabah, conduct their field studies at Banggi as of 2005, and could be potential sources for this information. 104 did not attempt this in the present study as a much stronger underlying model would be needed to extend the analysis to the Ecospace environment, as was undertaken for a coral reef on the Great Barrier Reef (Tudman, 2001). There were indirect benefits from reserve establishment that were not reflected by the model. Firstly, the model did not capture the likelihood of a spillover effect, where there is a net emigration of adult fish from reserves to the surrounding area, thereby leading to increased fishery yields at fishing grounds adjacent to the reserve (Gell and Roberts, 2003). Although the occurrence of spillover is still controversial in the scientific literature (Russ, 2002), it has been suggested in several reserves, for example in Mombasa Marine Park, Kenya (McClanahan and Mangi , 2000) and at Apo Island (Russ and Alcala , 1996). A recent study by Abesamis and Russ (2005) strengthened the evidence for the occurrence of density-dependent export of a reef fish from a protected, no-take reserve to adjacent fished areas. The model also did not fully address the non-market values (e.g., intrinsic value) and values arising from non-consumptive uses of marine reserves (Sumaila et al, 2000). In terms of economic benefits, tourism development centred on marine reserves has become a popular choice for generating alternative sources of income and employment for formerly fishing dependent island communities (e.g., White et al, 2002). However, tourism has to be developed within a sustainable context, as all too often poorly planned projects have resulted in environmental degradation and negative social effects (Wong, 1998). Alternative livelihood options such as seaweed farming or cottage industries have been used in place of tourism in some communities. Whatever the choice of substitute income and employment may be, the important point is that cessation of fishing would have to occur in conjunction with viable (in terms of financial, socio-cultural, and environmental sustainability) alternative livelihood programmes for the local fishing communities. Policy 3 presented an improvement over policy 1 in terms of the status of target fish biomass, which in the long run would be maintained at a stable state, although at a level below the initial 2005 level. This implied that fishers would still be able to maintain catches of target fish, albeit at lower quantities than 2005 levels. However, market mechanisms would likely 105 push up the price of scarce fish (e.g., live fish species) and lower catches would therefore not necessarily translate to decreased economic benefits for the remaining fishers in the fishery. The rising price for valuable live fish such as Plectropomus throughout the 1990s (Daw et al., 2002) illustrates this point. However, this situation would likely not be sustainable, as economic incentives would merely encourage heightened targeting of the species, eventually leading to the local stock being fished out. This already occurred with other valuable species such as humphead wrasse (Cheilinus undulatus) and humpback grouper (Cromileptes altivelis) (Chapter 2), and the same outcome probably awaits Plectropomus i f the fishery continues unregulated. In Banggi, the feasibility of implementing effort controls is highly questionable given the socio-economic circumstances o f the island's coastal communities. A n y attempt to impose fishing effort restrictions when there is no alternative source o f food and income for the local villagers would only encourage poaching and illegal fishing. Moreover, the enormous amount of personnel and funding required to carry out effective enforcement of such restrictions is often not available to local enforcement/management agencies, as is the case in Sabah (Pilcher and Cabanban, 2000). Similar to Policy 3, the effort restrictions imposed in Pol icy 4 resulted in a stable state of target fish population. In this case, biomass of all target groups was either maintained or increased; subsequently, catches o f valuable species such as L C S and Plectropomus increased. This policy resulted in an ideal outcome in terms of generating fisheries as well as socio-economic conditions. However, as previously mentioned, the feasibility of actually implementing effort controls is very much in doubt given the current management regime and socio-economic environment. In summary, the varying simulation results indicated the need to make trade-offs when deciding on future management direction. Overall, Policy 2 and 4 showed the most potential as management strategies which could balance both ecological and socio-economic needs. 106 4.4.3 Marine reserves The modelling exercise demonstrated the expected theoretical responses of fish populations when subject to varying degrees o f fishing pressure. Within the context of Banggi's artisanal fisheries, there is a clear need for a decrease or stabilization in fishing effort. Fishers also recognized that historical increase in fishing effort has contributed to the present state of declining fisheries (Chapter 3). Effort regulations such as gear controls and catch quotas have been applied extensively in fisheries all around the world. However, the collapse of the Canadian Atlantic cod fishery demonstrated that even in one of the world's most well funded and managed fisheries, effort controls can fail. Given the prevailing socio-economic and enforcement barriers in Banggi, the effectiveness of effort controls becomes extremely doubtful. On the other hand, marine reserves offer a potential solution to Banggi's declining fisheries. Marine reserves have been increasingly advocated as a tool for fisheries management and conservation (Russ, 2002), even though there is limited empirical evidence to support their theoretical benefits (Russ, 2002; Gel l and Roberts, 2003). Nevertheless, Sumilon and Apo Reserves in the central Philippines present perhaps the best evidence of the utility of marine reserves as management tools for small scale artisanal reef fisheries in a geographical setting comparable to Banggi. Long-term protection at both Sumilon and Apo resulted in significantly increased fish biomass in both reserves (Russ and Alcala , 1998), and increased yields to fishers at adjacent fishing areas at Apo (Russ et al, 2003), and possibly at Sumilon (Alcala and Russ, 1990). Marine reserves can potentially cause hardship to fishing communities (Hilborn et al, 2004). Christie (2004) found that while some marine reserves demonstrated biological success, social conflicts and economic distress were created in the process. Thus, in order to minimize economic hardship to fishers, the potential marine reserve in southern Banggi w i l l have to cover only a small portion of the Maliangin fishing ground. Furthermore, a community-based, bottom-up (rather than top-down) approach has been emphasized as the key to sustainable coral reef management (White and Vogt, 2000). Community-based management stresses the need for local education on management objectives, and full community participation in 107 planning, implementing, and maintaining the marine reserve. On top of this, local support for the reserve is vital (Sumaila et al, 2000). The results o f the present study, particularly interview results, can be used as a starting point from which to build local community buy-in for a marine reserve. A t Apo Island, full local support resulted in continuous protection of the reserve for more than 20 years. This not only substantially improved the fish abundance and fishery, but also enabled Apo Island to develop into a well known dive site, thereby bringing economic benefits to the island's communities (Alcala, 1998; White et al., 2002). In contrast, establishment of the original Sumilon reserve did not fully involve the community. The subsequent interplay of fragmented community support, politics, and false expectations eventually led to the reserve's collapse twice (Russ and Alcala , 1999). There are undoubtedly many challenges associated with initiating community-based management in Sabah, foremost being the centrally controlled management of Malaysian fisheries. Fishers, who have traditionally had minimal participation in the management process, are not familiar with the concept of self-management (Pomeroy and Viswanathan, 2003), and thus lack, the capability to do so. In addition, fishers exhibit an unwillingness to become stewards of their fishery resources and enforcer of laws and regulations. This same sentiment is prevalent among south Banggi's fishers, and w i l l require much education and motivation to redress. Nonetheless, the recent proposal for the Tun Mustapha Park, which includes Banggi, provides a unique opportunity for initiating a shift in management paradigm towards a community-based approach. Within this context I make the following recommendations: 1) Integrate field workers and researchers (social, ecological, or fisheries) into the community by living and working in the community; 2) These field staff w i l l then set up education and community outreach programmes with marine conservation and co-management themes. The aims of these programmes are to: i) motivate community members to take charge of their own reef resources; ii) educate them on the objectives and potential outcomes of marine reserve protection 108 so that there are no false expectations of immediate and observable benefits in terms of fishery yields and/or tourism development; A v o i d top-down control. Local fishers should be consulted about, and approve, the placement of one or several marine reserves. Marine reserve sites should ideally be: i) a small portion of the current actively fished area (e.g., Apo reserve is only 10% of the entire coral reef area). Small reserves minimize negative social and economic impacts on local fishing communities, and might also enhance fishery value more than large reserves (Sale et ai, 2005); ii) within view or easy reach o f a village to facilitate enforcement; i i i) placed where there is a mixture of good and degraded reef habitat; The local community should be given the responsibility for enforcing and maintaining the small marine reserve(s). Local enforcers should also be given the appropriate powers to effectively deal with violations; Management agencies should provide institutional and funding support to facilitate the process of establishing and protecting the marine reserves. Funding support could take the form of: a) Supplying and maintaining permanent and distinct marker buoys to clearly delineate the boundaries of the marine reserve; b) Funding the outreach workers needed to carry out the community work. c) Providing funds for the education programme and for local community leaders and fishermen representatives to travel to successful marine reserves (e.g. Apo Island) for a learning experience; d) Capital investment, such as a community patrol boat, or construction of a guard house for enforcement purposes; e) Investigating viable alternative income or livelihood options. For example, the Borneo Marine Research Institute runs two such programmes in Banggi -seaweed farming and small scale agriculture. 109 4.5 Future research and concluding remarks The lack of localized biological and population data for Banggi's reef fish populations resulted in an Ecopath model which was crude at best in its representation of the actual ecosystem. The utility of future models can be greatly enhanced by more research on various life history parameters of targeted food fish, such as growth and natural mortality rates, recruitment, spawning, migration, and diet compositions. At the same time, long-term monitoring of fisheries landings and fishing effort, combined with underwater visual census of target fishery groups, can improve our understanding of fishing effects and the effectiveness of the marine reserve. This will enable better assessments and decisions to be made regarding the sustainability of fishing practices in the Banggi area. Cost-benefit analysis to examine the trade-offs between different management strategies will be required. Furthermore, social research on understanding the perceptions of locals about the marine reserve is needed in order to foster the positive social dynamics that are key to coral reef management. Even though the two models built in this study were simple, they provided broad trends that were in line with our expectations, and were consistent with earlier work done in a similar physical and socio-economic setting as Banggi. (e.g., the 'fishing experiments' in Sumilon reserve (Russ and Alcala, 1998)). As such, these results can be informative for management agencies. However, when interpreting these results, we should keep in mind that catches and fishing effort in southern Banggi are influenced by a wide range of biological, environmental, and socio-economic drivers that are not accounted for in the model (See Chapter 3 for a discussion of these factors). The results from the 1980 model suggested that south Banggi's near-shore fisheries have progressed from lightly fished to one where target fishery groups are under high fishing pressure. The 2005 model alerted us to the probable depletion of targeted demersal reef fish if no fisheries management is initiated. The socio-economic environment in Banggi is such that imposition of effort controls is unlikely to succeed. It is therefore recommended that Banggi emulate the Apo Island example, in which establishment and continued protection of 110 the small no-take reserve for more than 20 years has been achieved through a community focused approach based upon strong local support for the reserve. If comparable buy-in and support is built and maintained by the Banggi fishing community, and facilitated/supported by local management agencies, there is the unique opportunity for southern Banggi communities to prevent irrevocable damage to their reef ecosystems, and create a sustainable future for their reef fisheries and socio-economic environment. 111 4.6 References Abesamis, R . A . , Russ, G.R. , 2005. Density-dependent spillover from a marine reserve: long term evidence. Ecol . App l . 15 (5), 1798-1812. Alcala , A . C . , Russ G.R. , 1990. A direct test of the effects of protective management on abundance and yield o f tropical marine resources. J. Cons. Int. Explor. Mer. 46, 40-47. Alcala , A . C . , 1998. Community-based coastal resource management in the Philippines: a case study. Ocean Coast. Manage. 38, 179-186. Al ino , P . M . , McManus L . T . , McManus, J.W., Nanola, C L . Jr., Fortes, M . D . , Trono, G . C Jr., Jacinton, G.S. , 1993. Initial Parameter Estimates of a Coral Reef Flat Ecosystem in Bolinao, Pangasinan, Northwestern Philippines. In: Christensen, V . , Pauly, D . (Eds.), Trophic models of aquatic ecosystems, I C L A R M , Manila , pp. 252-258. Arias-Gonzalez, J.E., Delesalle, B . , Salvat, B . , Galzin, R., 1997. Trophic functioning of the Tiahura reef sector, Moorea Island, French Polynesia. Coral Reefs 16, 231-246 Arias-Gonzalez, J.E., Nunez-Lara, E . , Gonzalez-Salas, C , Galzin, R., 2004. Trophic models for investigation of fishing effect on coral ref. ecosystems. Ecol . Model . 172, 197-212. Boutin, M . E . , 1990. The Bonggi. In: Lingenfelter, S.G. (Ed.), Social Organization of Sabah Societies. Sabah Museum and State Archives Department, Sabah, pp 91-109. Cabanban, A . S . , Biusing, R . E . , 1999. Coral reef fisheries and their contribution to marine fish production in Sabah, Malaysia. In: Cabanban, A . S . , Phillips, M . (Eds.), Proc. Workshop on Aquaculture of Coral Reef Fishes and Sustainable Reef Fisheries. Institute of Development Studies, Sabah. pp. 87-96. 112 Choat J .H. , Robertson, D.R. , 2002. Age based studies. In: Sale P.F. (Ed.), Coral reef fishes: dynamics and diversity in a complex ecosystem. Academic Press, San Diego, California, pp.57- 80. Christensen, V . , Walters, C.J . , Pauly, D . , 2004. Ecopath with Ecosim: a User's Guide. 154 pp. Available online at: www.ecopath.org Christie, P., 2004. Marine Protected Areas as Biological Successes and Social Failures in Southeast Asia . In: Shipley, J .B. (Ed.), Aquatic protected areas as fisheries management tools: design, use, and evaluation of these fully protected areas. American Fisheries Society, Bethesda, M D , pp. 155-164. Daw, T., Jeffrey, L . , B i n A l i , M . A . , 2002. Preliminary Assessment of the Live Reef Fish Trade in the Kudat Region. W W F Malaysia, Kota Kinabalu, 47 p. Department of the Environment and Heritage (DEH) Australia. 2004. Assessment of the Queensland East Coast Spanish Mackerel Fishery. 26 p. Available online at: http://www.deh.gov.au/coasts/fisheries/qld/east-coast-spanish-mackere1/pubs/east-coast-spanish-mackerel-assessment.pdf Garces, L .R . , Al ias , M . , A b u Talib, A . , Mohamad-Norizam, M . , Silvestre, G.T., 2003. A trophic model of the coastal fisheries ecosystem off the West Coast of Sabah and Sarawak, Malaysia, p. 333 - 352. In: Silvestre, G . , Garces, L . , Stobutzki, I., Ahmed, M . , Valmonte- Santos, R . A . , Luna, C , Lachica-Alino, L. ,Munro, P., Christensen, V . , Pauly, D . (Eds.), Assessment, Management and Future Directions for Coastal Fisheries in Asian Countries. WorldFish Center Conf. Proc. 67, pp. 333-352. Gel l , F.R. , Roberts, C M . , 2003. Benefits beyond boundaries: the fishery effects of marine reserves. T R E E 18 (9), 448-455. 113 Gribble, N . A . , 2001. A Model of the Ecosystem, and Associated Penaeid Prawn Community, in the Far Northern Great Barrier Reef. In: Wolanski, E . (Ed.), Oceanographic Processes of Coral Reefs: Physical and Biological Links in the Great Barrier Reef. C R C Press, pp. 189-207. Gribble, N . A . , 2003. GBR-prawn: modelling ecosystem impacts of changes in fisheries management of the commercial prawn (shrimp) trawl fishery in the far northern Great Barrier Reef. Fish. Res. 65, 493-506. Halpern, B.S . , Warner, R .R. , 2002. Marine reserves have rapid and lasting effects. Ecol . Lett. 5, 361-366. Harding, S., Lowery, C , Wesson, H . , Colmer, M . , Daw, T., 2001. The Pulau Banggi Project for Coral Reef Biodiversity 1st Annual Report (July 1999- September 2000). Greenforce, Kudat, 98 p. Hilborn, R. , Stokes, K . , Maguire, J.J., Smith, T., Botsford, L . W . , Mangel, M . , Orensanz, J., Parma, A . , Rice, J., Be l l , J., Cochrane, K . , Garcia, S., Ha l l , S., Kirkwood, G.P. , Sainsbury, K . , Stefansson, G . , Walters, C , 2004. When can marine reserves improve fisheries management? Ocean Coast. Manage. 47, 197-205. Jennings, S., Lock, J . M . , 1996. Population and ecosystem effects of reef fishing. In: Polunin, N . V . C . , Roberts, C M . (Eds.), Reef Fisheries. Chapman and Ha l l , London, U K pp. 193-218. Jennings, S., Polunin, N . V . C , 1996. Effects of fishing effort and catch rate upon the structure and biomass of Fijian reef fish communities. J. of App. Ecol . 33, 400-412 Koh , L . L . , Chou, L . M . , Tun, K .P .P . , 2002. The status of coral reefs of Pulau Banggi and its vicinity, Sabah, based on surveys in June 2002. R E S T Technical Report. National University of Singapore, Singapore. 36 p. 114 McClanahan, T.R., 2000. Recovery of a coral reef keystone predator, Balistapus undulates, in East African marine parks. B i o l . Conserv. 94, 191-198. McClanahan, T.R. , Mangi , S., 2000. Spillover of exploitable fishes from a marine park and its effect on the adjacent fishery. Ecol . A p p l . 10 (6), 1792-1805. McManus, J.W., Menez, L . A . B . , Kesner-Reyes, K . N . , Vergara, S.G., Ablan, S .M.C . , 2000. Coral reef fishing and coral-algal phase shifts: implications for global reef status. ICES J. Mar. Sci . 57, 572-578. Opitz, S., 1993. A Quantitative Model of the Trophic Interactions in a Caribbean Coral Reef Ecosystem. In: Christensen, V . , Pauly, D . (Eds.), Trophic models of aquatic Ecosystems. I C L A R M , Manila, pp. 259-267. Pauly, D . , Christensen, V . , Walters, C , 2000. Ecopath, Ecosim, and Ecospace as tools for evaluating ecosystem impact of fisheries. ICES J. of Mar. Sci . 57 (3), 697-706. Pilcher, N . , Cabanban, A . S . , 2000. The Status of Coral Reefs in Eastern Malaysia. Global Coral Reef Monitoring Network ( G C R M N ) Report. Australia Institute of Marine Science, Townsville. 58 p. Pomeroy, R.S.,Viswanathan, K . K . , 2003. Experiences with fisheries co-management in Southeast As i a and Bangladesh. In: Wilson, D . U . , Nielsen, J.R., Degnbol, P. (Eds.), The Fisheries Co-management Experience Accomplishment, Challenges and Prospects. Kluwer Academic Publishers, Boston, M A , pp. 99-115. Roberts, C M . , 1995. Rapid Build-up of Fish Biomass in a Caribbean Marine Reserve. Cons. B io . 9 (4), 815-826. 115 Russ, G.R. , 2002. Yet another review of marine reserves as reef fisheries management tools. In: Sale, P.F. (Ed.), Coral reef fishes: dynamics and diversity in a complex ecosystem. Academic Press, San Diego, C a l i f , pp. 4 2 1 - 443. Russ, G.R. , Alcala , A . C . , 1989. Effects o f intense fishing pressure on an assemblage of coral reef fishes. Mar. Ecol . Progr. Ser. 56, 13-27. Russ, G.R. , Alcala , A . C . , 1996. Marine reserves: rates and patterns of recovery and decline in abundance of large predatory fish. Ecol . App l . 6, 947-961. Russ, G.R. , Alcala , A . C . , 1998. Natural fishing experiments in marine reserves 1983-1993: community and trophic responses. Coral Reefs 17, 383-397. Russ, G.R. , Alcala , A . C . , 1999. Management histories of Sumilon and Apo Marine Reserves, Philippines, and their influence on national marine resource policy. Coral Reefs 18, 307-319. Russ, G.R. , Alcala , A . C . , Maypa, A . P . , 2003. Spillover from marine reserves: the case of Naso vlamingii at Apo Island, the Philippines. Mar. Ecol . Prog. Ser. 264 (15), 15-20. Russ, G.R. , Alcala , A . , 2004. Marine reserves: long-term protection is required for full recovery of predatory fish populations. Oecologia 138, 622-627. Sale, P.F. (Ed.), 1991.The Ecology'of Fishes on Coral Reefs. Academic Press, San Diego, C A . Sale, P.F. (Ed.), 2002. Coral reef fishes: dynamics and diversity in a complex ecosystem. Academic Press, San Diego, C A . 116 Sale, P.F., Robert, C . K . , Danilowicz, B .S . , Jones, G.P. , Kritzer, J.P., Lindeman, K . C . , Planes, S., Polunin, N . V . C . , Russ, G.R. , Sadovy, Y . J . , Steneck, R.S. , 2005. Critical science gaps impede use of no-take fishery reserves. T R E E 20 (2), 74-80. Sumaila, U . R . , Guenette, S., Alder, J. , Chuenpagdee, R., 2000. Addressing ecosystem effects of fishing using marine protected areas. ICES J. of Mar. Sci . 57, 752-760. Teh, L . , Cabanban, A . S . , Sumaila, U . R . , 2005. The reef fisheries of Pulau Banggi, Sabah: A preliminary profile and assessment of ecological and socio-economic sustainability. Fish. Res. 76, 359-367. Tudman, P .D. , 2001. Modell ing the trophic effects of fishing on a mid-shelf coral reef of the central Great Barrier Reef. BSc . Honours thesis, James Cook University, Townsville. Walters, C , Christensen, V . , Pauly, D . , 1997. Structuring dynamic models of exploited ecosystems from trophic mass balance assessments. Rev. Fish. B i o l . Fish. 7, 139-172. White, A . T . , Vogt, H.P. , 2000. Philippine Coral Reefs Under Threat: Lessons Learned After 25 Years of Community-Based Reef Conservation. Mar . Pollut. B u l l . 40 (6), 537-550. White, A . T . , Courtney, C . A . , Salamanca, A . , 2002. Experience with Marine Protected Area Planning and Management in the Philippines. Coast. Manage. 30, 1-26. Wong, P. P., 1998. Coastal tourism development in Southeast As ia : relevance and lessons for coastal zone management. Ocean Coast. Manage. 38 (2), 89-109. Zeller, D . C . 1997. Home range and activity patterns of the coral trout Plectropomus leopardus (Serranidae). Mar. Ecol . Prog. Ser. 154, 65-11. 117 C H A P T E R 5 C O N C L U S I O N 5.1 Summary of main findings The objective of this study was to conduct a preliminary analysis of southern Banggi's reef fisheries. The main findings are: 1) The reef fisheries appear to be fairly productive when compared to other similar Southeast Asian reefs known to be overfished. However, signs of unsustainability, such as the continued practice of blast fishing, and intense targeting of coral groupers, are also accumulating, suggesting that fisheries resources are under threat, and that mitigation actions are needed; 2) The catch rates of artisanal fishers appear to have declined substantially over the past 20 years, and particularly in the last five years. Fish catches and fishing effort are highly variable, and influenced by seasonal changes and variation in biological, ecological, biophysical, and socio-economic conditions. However, fishers suggest that the decline in catch rates has been primarily driven by socio-economic factors (e.g., human population growth, market demands for valuable fish species, commercial purse seine operations); 3) Management strategy exploration using Ecopath with Ecosim confirmed the obvious need to reduce fishing effort for the long-term sustainability of the reef fisheries. In terms of marine reserve design, a network of small reserves is suggested as the best option for fisheries management (Hastings and Botsford, 2003). Moreover, a small reserve minimizes the social and economic hardship imposed on local fishing communities. Thus, the establishment of a small scale marine reserve, operating under a community based approach, and supported with alternative income or livelihood programmes for the coastal communities, is considered the most suitable first step that needs to be taken for the sustainable management of Banggi's reef fisheries. 118 5.2 Research approach This research took a multidisciplinary approach to understanding Banggi's reef fisheries. This was reflected in the research methods, which combined quantitative methods (fishery data collection and ecosystem modelling) with interviews and personal observations and participation. Using this approach enabled me to gain an overall view of the fisheries, and also the ability to cross verify information, which might not have been possible by using either one method alone. A distinct feature of this research was that it was able to make use of a relatively sparse knowledge base to undertake exploratory analysis and generate recommendations for future management direction. In doing so, it overcame the data poor barrier that so often impedes research for decision-making on reef fisheries management. 5.2.1 Potential sources of error Surveys and estimates Although I attempted to sample catches in two different fishing seasons, unexpected weather conditions during the second field season meant that fish catch levels probably did not reach their usual peak levels. The reef fisheries of Banggi would therefore benefit from year-long monitoring of landings, which would give a much more accurate representation of catches and effort. This is especially true for the case of Spanish mackerel, whose catches peak during the end of the year, and as such was not captured in any of the sampling seasons. A s I chose to carry out the creel surveys at fish traders' stalls, I did not capture the subsistence catches, and also catches from the spear and trap fisheries, most of which were not sold. Estimations of reef fisheries yields vary depending on how one defines the reef area being fished, and the fish species to be included in the reef fisheries (Dalzell, 1996). In this study, the fished reef area included only the actively fished spots within the Maliangin fishing grounds. This would therefore have underestimated the total area of actively growing coral reef in the fishing ground, leading to higher estimated yield per area. Moreover, the area can be considered a coarse approximation as it was estimated from a map. In terms of fish species, reef-associated fish such as scombrids and carangids constituted a substantial portion 119 of the catches in Banggi. Again, this led to an estimated yield which would be on the high side compared to an estimate which included strictly demersal reef fish only. Interviews The participation of fishers in interviews was a key source o f information in this research. However, it is recognized that fishers' knowledge is subject to memory loss (Neis et al, 1999), and this might have affected the reliability o f information pertaining to temporal trends in catch rates. In addition, although I am proficient in Malay, I am not a native speaker, much less of Ubian, the local language. Therefore, I might have misunderstood or misinterpreted some parts of a response. Model Models are limited by the quality of data used as the basis for the input parameters (Whipple et al, 2000). This research attempted to use Ecopath with Ecosim as a tool for exploring ecosystem effects of various management strategies. Ecopath was first developed to model the coral reefs o f the French Frigate Shoals, thus, its applicability to reef ecosystems is not disputed. However, in general, information on parameters pertaining to coral reef food webs is still lacking (Munro, 1996), and this is particularly so for the reef ecosystem of Banggi. A s a result, the Ecopath models presented in this research were weak due to the absence of local biological parameters. In addition, the time series data requirements of Ecosim presented a barrier for reef fisheries that are poorly, i f at all , monitored. More comprehensive data for these two areas are needed before a more refined model can be built. 5.3 Significance of research This study is significant as it has provided insights into the probable state of south Banggi's reef fisheries, which has not been investigated before. In doing so it has raised awareness about the need for management initiatives to mitigate the current declining trend of the island's reef fisheries. This in itself can be considered an achievement, given the general marginalization of small scale fisheries by decision makers (Pauly, 1997). The recent approval of the Tun Mustapha Park presents a unique opportunity to shape the direction of 120 future management decisions. Moreover, this study constitutes an ex ante assessment of south Banggi's reef fisheries. Baseline information gathered now w i l l be valuable in providing a benchmark against which to judge the effectiveness of future management decisions. Collection of pre-protection data is normally hard to come by (Gell and Roberts, 2003), making these research results particularly valuable. The data poor condition of coral reef fisheries in general has often been attributed to the lack of effective management, hence overexploitation. The lack of data stems from a lack of funding (Sadovy, 2005), due to the perception among policy makers that reef fisheries are not economically valuable enough to warrant spending money on (Russ, 1991). However, this study demonstrated that sufficient amount of baseline fishery and socio-economic data can be gathered, analyzed, and used for high level decisions, even with limited funds and time. This challenges the conventional perception among policy makers that reef fisheries are too costly in terms o f personnel and funding to warrant monitoring (Russ, 1991). 5.4 Data gaps and further research This research has enabled the identification of several key areas where knowledge relevant to Banggi's reef fisheries management is either lacking or incomplete. These include: i) biological status of targeted fish; ii) long-term data on fish landings and fishing effort; i i i) socio-economic status of fishers; and iv) physical studies; e.g., coral reef habitat. In addition, investigation of many more facets of reef fisheries remained beyond the scope of this research. With the recent approval of Tun Mustapha Park, one of the immediate research needs w i l l be to examine how to design an effective marine reserve which w i l l maintain or improve fisheries productivity, while minimizing economic and social hardship on existing fishers. This is a challenging task, given the considerable knowledge gaps in the ecological aspects of marine reserves (Sale et al, 2005). Once a marine reserve is established, monitoring studies related to the ecological, social-economic and policy effects of marine reserves w i l l be a priority. Briefly, potential future research focuses for Banggi might include 121 1) Fishery effects of marine reserves, such as those carried out by G . Russ and A . Alcala in Sumilon and Apo Islands, which combined long term underwater visual censuses with fishery catch and effort monitoring; 2) Biological studies on target species (e.g., life history traits, movement, recruitment and settlement rates, and larvae dispersal) to assess the status of target fish populations. These studies w i l l also contribute valuable input data for future ecosystem modelling; 3) The spatial allocation of fishing effort and underlying factors driving this spatial pattern (e.g., Pet-Soede et al, 2001). The use of spatial analysis using Geographic Information Systems (GIS) could be applied here; 4) Cost-benefit analysis and economic valuation of coral reef resources to investigate the trade-offs between various management strategies; 5) Social studies related to community management, as well as investigating the perceptions of fishers regarding the status of their fish stocks, and of management, etc.; 6) The application of Ecospace, the spatial component of E w E , to investigate the impacts related to placement of marine reserves. 5.5 Outlook Coral reef management is inherently multidisciplinary in nature, as the state of coral reef fisheries resources is linked to its wider social, economic, and biophysical environments (McManus, 1997; Sadovy, 2005). One increasing point of focus is the importance of community participation and co-operation in reef resource management (White and Vogt, 2000). Thus, I see future management of Banggi's reef fisheries entailing a process which integrates the physical coral reef environment with its associated social and economic systems. This expands the focus from merely managing fishing effort to include the implementation of marine conservation education programmes, alternative livelihoods, and community building programmes, to name a few. The amalgamation of these areas can perhaps be best addressed under the broader framework of integrated coastal zone 122 management ( ICZM) . Sabah, being one of the few Malaysian states to have a completed I C Z M plan, has the institutional framework from which to pursue this approach. 5.6 Concluding remark Banggi's reef fisheries are at a point where they can be considered to be in relatively good shape compared to reef fisheries elsewhere; yet signs of unsustainability are also increasingly obvious, and continued exploitation at current rates may likely lead to depleted fisheries resources in the future. Therefore action should be taken to alleviate drivers of overexploitation as soon as possible in order to prevent irrevocable damage. While marine reserves are by no means a foolproof solution for declining fisheries (Gell and Roberts, 2003; Sale et al, 2005), continuing with the status quo (no management) is clearly not an option. Thus, the closure of a small portion o f the Maliangin fishing ground for the establishment o f a no-take marine reserve is recommended. I f done under a community-based and holistic approach, this presents possibly the best option for managing southern Banggi's reef fisheries while maintaining socio-economic benefits to the community. 123 5.7 References Dalzell , P., 1996. Catch rates, selectivity and yields of reef fishing. In: Polunin, N . V . C . , Roberts, C M . (Eds), Reef Fisheries. Chapman & Ha l l , London, pp.161-192. Gel l , F.R. , Roberts, C M . , 2003. Benefits beyond boundaries: the fishery effects of marine reserves. T R E E 18 (9), 448-455. Hastings, A . , Botsford, L . W . , 2003. Comparing designs of marine reserves for fisheries and for biodiversity. Ecol . App l . 13, S65-S70. McManus, J.W., 1997. Tropical marine fisheries and the future of coral reefs: a brief review with emphasis on Southeast Asia . Coral Reefs 16, Suppl: S121-S127. Munro, J.L., 1996. The scope of tropical reef fisheries and their management. In: Polunin, N . V . C , Roberts, C M . (Eds.), Reef Fisheries. Chapman & Ha l l , London, pp.1-14. Neis, B . , Schneider, D . C , Felt, L . , Haedrich, R . C , Fischer, J., Hutchings, J .A. , 1999. Fisheries assessment: What can be learned from interviewing resource users? Can. J. Fish. Aquat. Sci . 56, 1949-1963. Pauly, D . , 1997. Small-Scale Fisheries in the Tropics: Marginality, Marginalization and Some Implications for Fisheries Management. In: Pikitch, E . K . , Huppert, D .D. , Sissenwine, M . P . (Eds), Global Trends: Fisheries Management. American Fisheries Society Symposium 20, Bethesda, Maryland, pp. 40-49. Pet-Soede, C , van Densen, W . L . T . , Hiddink, J .G., K u y l , S., Machiels, M . A . M . , 2001. Can fishermen allocate their fishing effort in space and time on the basis of their catch rates? A n example from Spermonde Archipelago, S W Sulawesi, Indonesia. Fish. Manag. Ecol . 8, 15-36. 124 Russ, G.R. , 1991. Coral Reef Fisheries: Effects and Yields. In: Sale, P.F. (Ed..), The Ecology of Fishes on Coral Reefs. Academic Press, California, pp. 601-635. Sadovy, Y . , 2005. Trouble on the reef: the imperative for managing vulnerable and valuable fisheries. Fish Fish 6, 167-185. Sale, P.F., Cowen, R . K . , Danilowicz, B .S . , Jones, G.P. , Kritzer, J.P., Lindeman, K . C . , Planes, S., Polunin, N . V . C . , Russ, G.R. , Sadovy, Y . J . , Steneck, R.S. , 2005 Critical science gaps impede use of no-take fishery reserves. T R E E 20 (2), 74-80. Whipple, S.J., Link, J.S., Garrison, L .P . , Fogarty, M . J . , 2000. Models of predation and fishing mortality in aquatic ecosystems. Fish Fish 1:22-40. White, A . T . , Vogt, H.P. , 2000. Philippine Coral Reefs Under Threat: Lessons Learned After 25 Years of Community-Based Reef Conservation. Mar. Pollut. B u l l . 40 (6), 537-550. 125 APPENDIX I: INTERVIEW QUESTIONNAIRE 1 Name: Age: Date: Gender: Education: Primary/Secondary/None Time: Village: 1. Personal 1. How many years have you lived here (Karakit area)? 2. Have you always lived here? Y / N 3. If NO, When did you move here? 4. Where did you come from? 5. How many people were living here when you first arrived? 6. How many people are there in your household? 2. Fishing I. General 1. What gears do you use to fish? Hook & Line Gillnet Spear (compressor/free dive) Trap Main gear: 2. Where do you fish? (point out on map) 3. How deep are the places where you fish? 4. Did you always fish here? 5. When did you start fishing in this area? (How many years ago) 6. Before that where did you fish? 7. Why did you start fishing in this area? 8. When you started fishing how many fishermen fished in the same area? 9. Now how many other fishermen fish in the same area? 10. Does the number of fishermen vary during the year? Y / N ' Prior approval from the U B C Behavioural Research Ethics Board was received to conduct these interviews. 126 If Yes M i n M a x Average High season L o w season 11. Can you tell me how many fishermen were fishing in this area 5, 10.. .20 years ago? # years ago M a x M i n Average 5 10 15 20 12. Were all these fishermen locals (lived in Banggi)? 13. Do you think the number of fishermen fishing in Maliangin area w i l l increase in the next 5, 10, 15, 20 years? 14. Do you ever fish anywhere else? 15. Where? 16. Why do you go fish elsewhere? 17. When do you normally go elsewhere to fish? 18. Do other fishermen do the same thing? 19. How many hours do you fish a day? 20. Last time did you spend more or less time fishing per day? 21. How many days do you fish a week? H r s fishing per day Day fishing per week Max M i n Average M a x M i n Average High season L o w season 127 II . F o r day time fishers: 22. Do you fish at night? Y / N (If Y go to 23, N go to III.) 23. How often? 24. What gear do you use? 25. Is there a certain time of month/year when you fish at night more often? 26. Where do you go to for night fishing? 27. How much (kg) do you catch at night? 28. What species do you catch at night? 29. Why do you fish at night? III. F o r hook & line fishers: 30. Do you ever use gillnet? Y / N (If Y go to 31, N go to 34) 31. How often do you use gillnet? 32. When do you use gillnet? 33. Do you fish in the same area with gillnet? 34. Why don't you use gillnet? 35. I f you had the opportunity would you want to fish with gillnet? 36. Why? I V . F o r gillnet fishers: 37. Do you ever use hook and line? Y / N (If Y go to 38, N go to 41) 38. How often do you use hook and line? 39. When do you use hook and line? 40. Do you fish in the same area with hook and line? 41. Why don't you use hook and line? 42. If you had the opportunity would you want to fish with hook and line? 43. Why? V . Traps 44. Do you use traps? Y / N ( if Y go to 45, N go to 55) 45. How many times a week do you set your trap? 46. How many days do you leave your trap in the water? 47. Where do you set your trap? 48. Do you use traps the whole year? 49. Is there a time of year when you set trap more often? 50. When is this? 51. What type of fish do you normally catch with a trap? 52. How many kilos off ish on average can you catch with a trap? 53. What is the max amount you can catch with a trap each time? 54. What is the min amount? 55. W h y don't you use traps? VI. Catch quantity & amount kept for food Catch per day (kg) Quanti ty kept for food (kg) Max M i n Average Max M i n Average High season L o w season 1. What species do you normally catch? 2. Which are the more valuable species? a. Have they always been more valuable? Y / N (If Y go to b, N go to c) b. Why? c. When did they start to become valuable? 3. Out of a catch of 10 kilos, how many kilos would be valuable fish? 4. How much (kg or # of fish) fish do you eat per day? 5. What types of fish do you use for bait? 6. How do you get your bait fish? Buy (go to 8) Catch own bait (go to 9) 7. Where/from whom do you buy bait? a. How much do you buy at one time? b. How many days does this bait last? c. How much does it cost? 129 d. Can you get it throughout the year? 8. Where do you go to catch bait? a. What gear do you use to catch bait? b. Can you find bait fish throughout the year? VII. L i v e fish (leopard coral grouper, Plectropomus spp.) 9. How many times a week can you catch a sunnoh? 10. Is there a season when you can catch sunnoh more often? Y / N 11. When is this? (What months?) 12. Why do you think the fish are easier to catch during this time? 13. Where do you go to catch sunnoh? 14. Does this location change during the year? Season Location # per week Size (kg or cm) Species M a x M i n Average M a x M i n Average High L o w 15. When (in what year) did you start targeting sunnoh? 16. Has the price of sunnoh changed since you started selling them live? Y / N 17. B y how much? VIII. Temporal T rend 1. Has the amount of fish you catch each day changed from the past? Y / N 2. If you can catch 10 kg a day now, how much could you catch Weight of catch (kg) Size (kg or cm) 5 years ago 10 years ago 15 years ago 20 years ago 130 3. If it takes you 5 hours to catch 10kg of fish now, how long did it take to catch 10kg 5....20 years ago? # of years ago Time to catch 10 kg (hours) 5 10 15 20 4. Compared to last time, has the species that you catch changed? 5. Which species has decreased/increased in abundance the most? 6. Compared to last time, has the size of each species changed? Species Change in abundance Change in size 7. What has caused the decrease in catch? 8. What can be done to stop the decrease? 9. Do you think there w i l l be a time when there is very little/no fish left to catch? 10. Why or why not? 11. Would you agree to a no fishing zone? a. Why/Why not? 12. Where would the no fishing zone have to be in order for you to obey it? 131 3. Other fishing activity I. Outsiders 1. How often do you see outsiders fish in Maliangin area? 2. Is there a period/season when they come to Maliangin more than other times of the year? 3. When is this? 4. Why? 5. How many outsider boats normally fish at one time in Maliangin? 6. What gear do they use? 7. How many kilos do you think they catch a day? Season Catch per day (kg) High Low II. Invertebrates 1. Do you collect sea cucumber / scallop / abalone / giant clam? 2. Where do you go? 3. How do you collect it? 4. When do you do it? 5. Is there a season / time of month when it's easier to find? 6. How often do you do it? 7. How many people collect inverts in the Maliangin area? 8. Who do you sell the invertebrates to? 9. Which are the most expensive invertebrates? 10. What price do you get for it? 11. How many years have you been collecting invertebrates? 12. Have you noticed that it's getting easier/harder to find the invertebrates? 132 4. Fishing Costs 1. Do you own your own pump boat? 2. How old is your pump boat? 3. How many years does a pump boat last? 4. How much did your pumpboat and engine cost? 5. How much does fuel cost? How much does fuel cost for each fishing trip? 6. Gear costs: Cost factor RM/uni t Replacement frequency Hook & line: -- Hooks - Line -Reel -Bait Gillnet: - net - floats Trap 5. Fishing Income 1. Is fishing your only source of income? 2. If no what other source of income do you have? 3. What is your daily/monthly income from fishing? R M / d a y RM/month Max Min Avg Max Min Avg High season Low season 4. Do you have another job? Y/N (if Y go to 8) 5. What is your other job? 133 a. Where is it located? b. Do you do it year round? Y / N (if Y go to 9, N go to 10) 6. What time of the year do you do your other job? a. For how many months? 7. H o w much income (daily/monthly) do you get from that job? 8. Is the income you earn from fishing enough to live on? 9. Who do you sell your fish to? 10. If there were other jobs available on Banggi that offer same income as fishing, would you switch jobs or continue to fish? a. Why? 11. Do you think the number of fishermen w i l l decrease or increase in the future? a. Why? 6. Management Scenarios Imagine that I find out your fishing grounds are being overfished, and that the reefs need protection. Under these circumstances what do you think would have to be done? First, let interviewee come up with own suggestions, 1) 2) 3) If they do not give any suggestions, I will then go through the following questions: 1. Do you think government could restrict the type of gear you use to fish? Why/Why not? 2. Do you think they could restrict where you fish? 3. Do you think they could restrict when you fish? 4. Do you think they could ban fishing? Why/why not? 5. Could they limit fuel? Why/Why not? 6. Who should be managing the reefs? 134 A P P E N D I X I I : L I S T O F C O M M O N L Y C A U G H T F I S H S P E C I E S Table 1 List of commonly caught fish by hook and line. Local names are mostly in Ubian, a commonly spoken local language in south Banggi.* Family Local Name Scientific Name English Common Name Serranidae tengal Epinephelus fasciatus Blacktip grouper kutkut Epinephelus quoyanus Longfin grouper kerapu Epinephelus coioides Orange-spotted grouper bagahab Epinephelus fuscoguttatus Brown-marbled grouper kuhao Epinephelus aerolatus Areolate grouper bata Epinephelus corallicola Coral rockcod kuang kuang Cephalopholis miniata Coral hind sunnoh merah Plectropomus leopardus Leopard coral grouper sunnoh taising Plectropomus maculatus Spotted coralgrouper Lethrinidae ketambak Lethrinus lentjan Pink ear emperor syokong Lethrinus ornatus Ornate emperor laiparpar Lethrinus miniatus Trumpet emperor Lethrinus obsoletus Orange-striped emperor lahosuk Lethrinus olivaceus Longface emperor Lethrinus microdon Smalltooth emperor Lutjanidae sepak bakal/ bungamohok Lutjanus decussatus Checkered snapper komamohok Lutjanus carponotatus Spanish flag snapper bahawak Lutjanus fulviflamma Dory snapper pisang pisang Lutjanus vitta Brownstripe red snapper dapak Lutjanus gibbus Humpback red snapper Labridae bukaan Choerodon anchorago Orange-dotted tuskfish lampet Cheilinus fasciatus Redbreast wrasse syoke unidentified labrid lamun lamun Choerodon schoenleinii Blackspot tuskfish Nemipteridae bakit bakit Pentapodus spp. Whiptails tong tong Pentapodus emeiyii Double whiptail tong tong lemok Scolopsis temporalis Bald-spot monocle bream kerisi Nemipterus spp. Threadfin breams Scombridae bakulan sobat Euthynnus affinis Kawakawa tengirri Scomberomorus commerson Narrow-barred Spanish mackerel Carangidae silai silai Atule mate Yellowtail scad * Scientific and English common names were verified using FishBase (www.fishbase.org). 135 Family Local Name Scientific Name English Common Name Carangidae tulai Selar boops Ox-eye scad Caesionidae sulit Caesionidae spp. Fusiliers anjang anjang Pterocaesio tessellata One-stripe fusilier anduhao Pterocaesio marri Marr's fusilier Balistidae pegot Triacanthus biaculeatus Short-nosed tripodfish Balistapus undulatus Orange-lined triggerfish Acanthuridae komais Naso unicornis Brown unicornfish * Scientific and English common names were verified using FishBase (www.fishbase.org). Table 2 List of commonly caught fish by j gillnet. Local names are mostly in Ubian, a commonly spoken local language in south Bangg i * Family Ubian Name Scientific Name English Common Name Carangidae karian Carangoides orthogrammus Island trevally (trevallies) matak Caranx ignobilis Giant trevally tongap Ulua mentalis Longrakered trevally ingatan Carangoides gymnostethus Bludger lundiwak/lundiuk Carangoides fulvoguttatus Yellowspotted trevally butitikan / biru biruan/ Caranx melampygus Bluefin trevally langar languan Carangidae kubal kubal / gurungunj I Megalaspis cordyla Torpedo scad (other) termanung Atule mate Yellowtail scad Lethrinidae ketambak Lethrinus lentjan Pink ear emperor syokong Lethrinus ornatus Ornate emperor laiparpar L. miniatus Trumpet emperor L. obsoletus Orange-striped emperor kuambal Monotaxis grandoculis Humpnose big-eye bream Lutjanidae komamohok Lutjanus carponotatus Spanish flag snapper bahawak Lutjanus fulviflamma Dory snapper pisang pisang Lutjanus vitta Brownstripe red snapper Haemulidae lepeh Plectorhinchus picus Painted sweetlips Piectorhinchus pictus Trout sweetlips Scombridae rumahan Rastrelliger kanagurta Indian mackerel talang Scomberoides spp. Queenfish Epiphididae bunak Platax batavianus Humpback batfish Dasyatidae kyampau Taeniura lymma Bluespotted ribbontail ray * Scientific and English common names were verified using FishBase (www.fishbase.org). 136 A P P E N D I X III: A S S U M P T I O N S F O R E S T I M A T I N G F I S H I N G E F F O R T (F ISHING D A Y S P E R F I S H E R P E R Y E A R ) F O R H O O K A N D L I N E A N D G I L L N E T F I S H E R I E S # of fishing days per week Hook and Line Gillnet Season Duration (weeks) M i n M a x A v g M i n Max A v g Windy 22 3 4 3.5 4 5 4.5 Calm 13 6 7 6.5 6 7 6.5 Intermediate 18 4 5 4.5 5 6 5.5 Total per year - 216 269 243 256 309 283 137 APPENDIX IV: ASSUMPTIONS FOR PROFITABILITY AND R E V E N U E ESTIMATES Table 1 Assumptions for calculation of hook and line profitability Fishing trips per month June-Aug 2004 & Mar-Apr 2005 (NE & SW monsoon, windy) Apr-May 2005 (SW monsoon, calm) Ox-eye scad (full/ illuminated moon) Fixed cost Pumpboat 16 24 8 Estimated cost R M 16.25 per month Variable fishing costs Fuel Hooks, lines, lure, weight, spool** R M 10 per day R M 0.20 per day Repair and maintenance R M 0.55 per day Total variable costs per month Notes Cost of pump boat is assumed to be R M 1950, and is depreciated on a straight line basis over 10 years.* Hooks cost R M 2 for 10; fabric for lures R M 3; spool RM3; sinker R M 1; nylon line R M 3; metal eye R M 1. Baited hooks and lured hooks are usually used for about 5 days before changing. Nylon lines last about a week to 10 days, while metal weight and spool lasts substantially longer. This represents 5% of total variable costs, and is estimated based on the repair and maintenance costs for a small-scale handline fishery in Peninsular Malaysia, which constituted 2% of total operating costs.*** During Jun-Aug 2004 During Mar-Apr 2005 During Apr-May 2005 Ox-eye scad fishery R M 171.80 R M 214.70 R M 257.70 R M 43.80 * Pump boats last for about 10 years, and are depreciated over their useful lifes. ** Bait fish is not included as a fishing related cost because the majority of observed fishermen catch their own bait, since bait fish is not readily available for sale in Karakit. The fishermen who do get a regular supply of bait fish are those who join piskadul operations which travel to reefs beyond southern Banggi. In these cases the vessel owner provides them with bait for R M 1.20 per kg. If bait is included as a variable cost for regular hook and line fishing (assuming RM2 worth of bait is purchased each day), profitability will drop between 7 to 12% from estimates presented in Table 6, depending on season. *** Source: Fredericks, L.J. , Nair, S., Yahaya, J., 1985. Cost structure and profitability of small-scale fisheries in Peninsular Malaysia. In: Panayotou, T. (Ed.), Small-sclae fisheries in Asia: socioeconomic analysis and policy. IDRC, Ottawa, pp. 176-183. 138 Table 2 Assumptions for calculation of gillnet profitability Fishing trips per month Jun-Aug 2004 Nov 04-Feb 05, Mar-May 2005 Fixed costs Pumpboat Nets Variable fishing costs Fuel - S. Banggi fishing grounds - to Sibogo Ice (to Sibogo only) Repair and maintenance - June-August 2004 - Nov 04 to Feb 05 and Mar-May 2005 20 24 Estimated cost R M 16.25 per month R M 8.33 per month R M 8 per day R M 30 per trip R M 6 per trip R M 48 per month R M 54 per month 16 trips within southern Banggi, 4 trips (once a week) to Sibogo. 20 trips within southern Banggi, 4 trips to Sibogo. Notes Same as for hook and line Each net costs about R M 100. This consists of several individual nets, each costing R M 30, which are usually joined together to make longer nets at least 150-200 m in length. Prices for plastic floats and rope assumed to be RM10. Each net is assumed to last one year. RM320 This represents 15% of total variable costs. The estimated 15% is an average based on the repair and maintenance costs of two studies: 1) the repair and maintenance costs for two small-scale drift net fisheries in Peninsular Malaysia, which constituted 7 and 8% of total operating costs*; 2) the maintenance cost for a small-scale drift net fishery (24% of operational cost) in the west coast of Pensinsular Malaysia** During Jun-Aug 2004 Total variable costs per month R M 304 During Nov 04-Feb 05, Mar-May 2005 *Source: Fredericks, L.J. , Nair, S., Yahaya, J., 1985. Cost structure and profitability of small-scale fisheries in Peninsular Malaysia. In: Panayotou, T. (Ed.), Small-sclae fisheries in Asia: socioeconomic analysis and policy. IDRC, Ottawa, pp. 176-183. **Source: Ahmad, A .T . , Salim, K., Chee, P.E., Isa, M . M . , Lim, C.F., 2003. An Overview of the Socioeconoic Status of Fisheries in Malaysia. In: Silvestre, G., Garces, L , Stobutzki, I., Ahmed, M . , Valmonte-Santos, R.A., Luna, C , Lachica-Alino, L . , Munro, P., Christensen, V., Pauly, D. (Eds.), Assessment, Management and Future Directions for Coastal Fisheries in Asian Countries. WorldFish Center Conference Proceedings, 67, pp. 517-540. 139 T a b l e 3 Estimated monthly revenue from the sale of coral grouper during coral grouper season for hook and line fishers Number of fishing trips per month Frequency of coral grouper capture Average revenue from sale of coral grouper Total coral grouper revenue per month in Apr-May 2005 24 30% R M 44.46 R M 320.11 (=24*0.3 *RM44.46)* Based on frequency from 2004 (see chapter 1) Based on frequency from 2004 (see chapter 1) * If R M 320 is added to gross monthly revenue of R M 719.52 in Apr-May 2005 (Table 3-7), profitability increases to 74%. 140 A P P E N D I X V : F I S H L E N G T H S R E C O R D E D D U R I N G L A N D I N G S S U R V E Y S ( M A R - A P R 2005) Table 1 Average total lengths (cm) for commonly caught hook and line species. Species Average Total No. of Length (cm) samples Epinehphelus quoyanus 23.9 4 Lethrinus ornatus 23.6 12 Lutjanus decussates 22.6 4 Epinephelus fasciatus 22.0 10 Atule mate 24.6 9 Selar boops 19.9 20 Euthynnus affinnis 28.8 6 Scomberomorus commerson 85.0 3 Average all species, excluding 23.6 -scomberomorus commerson Table 2 Average total lengths (cm) for commonly cauj species. *ht gillnet Species Average Total Length (cm) No. of samples Atule mate 25.2 32 Caragoides orthogrammus 39.0 16 Ulua mentelis 26.7 31 Lethrinus lentjen 26.6 58 Lethrinus miniatus 26.4 28 Plectorhinchus picus 36.0 35 Lutjanus vitta 26.7 14 Lutjanus fulviflamma 23.2 14 Lutjanus carponotatus 25.1 20 Lutjanus lemniscatus 35.8 11 Alectis indicus 36.3 14 Carangoides fulvoguttatus 33.4 28 Rastrelliger kanagurta 26.2 35 Euthynnus affinnis 35.0 34 Caranx ignobilis 35.4 13 Megalaspis cordyla 31.8 14 Nemipterus spp. 24.5 24 Scomberomorus commerson 60.1 23 Average all species, excluding 30.0 -Scomberomorus commerson 141 APPENDIX VII: DIET MATRIX FOR 2005 ECOPATH MODEL Prey \ Predator 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1 Sharks 0.010 2 Large carangids and scombrids 0.145 3 medium carangids and scombrids 0.019 0.303 0.022 0.023 0.021 0.010 4 small pelagics 0.070 0.138 0.451 0.005 0.036 0.045 0.002 5 Benthopelagic feeders 0.029 0.001 0.005 0.001 6 Dermersal zoobe 0.001 0.003 0.004 0.010 0.018 0.005 0.010 7 Labridae 0.019 0.005 0.031 0.160 0.380 0.035 0.050 0.023 8 Lutjanidae 0.029 0.010 0.005 0.005 0.013 0.023 0.006 9 Plectropomus spp. 0.002 0.001 0.001 10 Other Serranidae 0.029 0.001 0.001 0.001 11 Caesionidae 0.019 0.053 0.050 0.005 0.060 0.066 0.053 0.021 0.008 12 Chaetodontidae 0.029 0.001 0.001 0.005 0.006 0.005 0.006 0.023 13 Pomacentridae 0.029 0.004 0.020 0.166 0.103 0.064 0.050 0.008 14 Pomacanthidae 0.010 0.021 0.005 0.022 0.023 0.050 0.023 15 Mullidae 0.005 0.041 0.001 0.005 0.012 16 Lethrinidae 0.029 0.002 0.005 0.00.1 0.001 0.000 0.001 17 Haemulidae 0.001 0.000 0.001 0.001 18 Scaridae 0.010 0.048 0.020 0.041 0.295 0.043 0.011 19 Siganidae 0.038 0.003 0.001 0.001 0.023 0.006 20 Acanthuridae 0.010 0.002 0.001 0.008 0.005 0.011 0.015 21 Balistidae 0.010 0.002 0.001 0.011 22 Nemipteridae 0.019 0.004 0.005 0.001 0.004 0.002 0.005 23 Large benthic inveterbrates 0.155 0.041 0.050 0.020 0.588 0.374 0.608 0.378 0.013 0.439 0.238 0.348 0.762 24 Mollusks/worms 0.260 0.080 0.074 0.002 0.031 0.203 0.050 0.568 0.248 0.109 25 Cephalopods 0.002 0.135 0.001 0.001 0.043 0.005 0.011 26 sea cucumbers and sea urchins 0.010 0.001 27 other echinoderms 0.010 0.001 0.035 0.112 28 Zoobenthos 0.100 0.089 0.026 0.208 0.129 0.253 0.092 0.038 0.100 0.300 0.160 29 Zooplankton 0.041 0.064 0.738 0.186 0.260 0.154 0.908 0.149 0.150 0.034 30 Corals 0.029 0.052 0.695 31 Benthic autotrophs 0.147 0.004 0.700 0.700 32 Phytoplankton 0.080 0.079 0.008 33 Detritus 0.138 0.020 0.062 0.106 18 19 20 21 22 23 24 25 26 27 28 29 30 0.176 0.001 0.010 0.058 0.999 0.800 0.198 0.100 0.942 0.100 0.297 0.353 0.303 0.100 0.149 0.176 0.303 0.023 0.101 0.053 0.100 0.100 0.149 0.001 0.198 0.294 0.218 0.078 0.100 0.006 0.108 0.444 0.316 0.100 0.150 0.020 0.010 0.200 0.203 0.316 0.150 0.012 0.095 0.101 0.050 0.050 0.737 0.442 0.483 0.051 0.316 0.100 0.800 0.263 0.300 0.700 Import Sum 143