UBC Faculty Research and Publications

Fisheries catch reconstructions: islands, part II Harper, Sarah; Zeller, Dirk 2011

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

Item Metadata


52383-FCRR_2011_19-4.pdf [ 1.87MB ]
JSON: 52383-1.0074737.json
JSON-LD: 52383-1.0074737-ld.json
RDF/XML (Pretty): 52383-1.0074737-rdf.xml
RDF/JSON: 52383-1.0074737-rdf.json
Turtle: 52383-1.0074737-turtle.txt
N-Triples: 52383-1.0074737-rdf-ntriples.txt
Original Record: 52383-1.0074737-source.json
Full Text

Full Text

ISSN 1198-6727  Fisheries Centre Research Reports  2011 Volume 19 Number 4  FISHERIES CATCH RECONSTRUCTIONS: ISLANDS, PART II  Fisheries Centre, University of British Columbia, Canada  FISHERIES CATCH RECONSTRUCTIONS: ISLANDS, PART II  Edited by Sarah Harper and Dirk Zeller  Fisheries Centre Research Reports 19(4) 143 pages © published 2011 by The Fisheries Centre, University of British Columbia 2202 Main Mall Vancouver, B.C., Canada, V6T 1Z4  ISSN 1198-6727  Fisheries Centre Research Reports 19(4) 2011 FISHERIES CATCH RECONSTRUCTIONS: ISLANDS, PART II Edited by Sarah Harper and Dirk Zeller  CONTENTS Director‘s foreword ............................................................................................................................................... 1 Preliminary estimate of total marine fisheries catches in Corsica, France (1950-2008) .................................3 A brief history of fishing in the Kerguelen Islands, France .............................................................................. 15 Reconstruction of total marine fisheries catches for Madagascar (1950-2008) ............................................. 21 Reconstruction of marine fisheries catches for Mauritius and its outer islands, 1950-2008 ....................... 39 Reconstruction of Nauru‘s fisheries catches: 1950-2008 ............................................................................... 63 Marine fisheries of Palau, 1950-2008: total reconstructed catch.................................................................... 73 Reconstruction of Sri Lanka‘s fisheries catches: 1950-2008 .......................................................................... 85 From local to global: a catch reconstruction of Taiwan‘s fisheries from 1950-2007 ...................................... 97 Reconstruction of Fisheries Catches for Tokelau (1950-2009) ..................................................................... 107 Reconstructing marine fisheries catches for the Kingdom of Tonga: 1950-2007 ........................................ 119 Reconstruction of marine fisheries catches for Tuvalu (1950-2009) ............................................................ 131  A Research Report from the Fisheries Centre at UBC  Fisheries Centre Research Reports 19(4) 143 pages © Fisheries Centre, University of British Columbia, 2011 FISHERIES CENTRE RESEARCH REPORTS ARE ABSTRACTED IN THE FAO AQUATIC SCIENCES AND FISHERIES ABSTRACTS (ASFA) ISSN 1198-6727  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  1  DIRECTOR‘S FOREWORD As marine fisheries resources around the world are increasingly threatened by pollution, climate change, and overfishing, it is more important than ever to know the amount and types of fish and invertebrates being extracted from the marine environment. Fisheries resources, particularly for island countries, provide a crucial source of food and income. However, the very fisheries which depend on these natural goods and services—notably small-scale fisheries—are being under-represented in fisheries statistics. In many of the countries highlighted in this report, the majority of seafood consumed is taken via subsistence fisheries. This non-commercial fishing sector is largely overlooked in statistical collection systems, particularly those of several decades ago, but continues to be under-represented today. In some places this is beginning to change, as the importance of small-scale fisheries to national food security is being recognized. In many developing countries, which lack the infrastructure and resources to fish their own waters for economic development through trade with external markets, foreign access fees are collected as a key source of revenue. In exchange for a modest fee, foreign fleets are allowed to fish their waters for high valued species. While this provides much needed income for the country, it also threatens the availability of these resources for domestic sustenance. While there is a range in the quality of fisheries reporting from one country to the next, in almost all countries Illegal, Unreported and Unregulated (IUU) fisheries exist. Fisheries landings statistics, as supplied to the United Nation‘s Food and Agriculture Organization (FAO), represent mainly the commercial and larger-scale fisheries. Artisanal, subsistence and recreational fisheries are mostly overlooked. Discarded bycatch and baitfish associated with certain fishing techniques are also rarely included in the official statistics. As a follow up to Fisheries catch reconstructions: Islands, Part I, this report continues to reconstruct total marine fisheries catches of island countries around the world from 1950 to present. This edition describes fisheries for island countries in the Pacific, Indian and Atlantic Oceans, highlighting the discrepancies that exist between reported landings and likely true catches. The reconstruction approach used here, as in the previous edition, aims to estimate all marine fisheries extractions as a baseline for monitoring and management purposes in the face of continued anthropogenic pressures. The future success of these countries relies, in part, on their ability to keep pace with an increasingly gl0bal economy while maintinaing a healthy supply of resources for domestic purposes. Ussif Rashid Sumaila, Director UBC Fisheries Centre August 2011  2  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  3  PRELIMINARY ESTIMATE OF TOTAL MARINE FISHERIES CATCHES IN CORSICA, FRANCE (1950-2008)1 Frédéric Le Manacha, Delphine Durab, Anthony Perecd, Jean-Jacques Riutorte, Pierre Lejeunec, Marie-Catherine Santonif, Jean-Michel Culiolif, and Daniel Paulyg aFaculty  of Science and Technology, University of Plymouth, Drake Circus, Plymouth PL4 8AA, United Kingdom; flm@livinblue.com bInstitut Polytechnique LaSalle Beauvais, 19 rue Pierre Waguet, BP 30313, Beauvais 60026 cedex, France; delphinedura@gmail.com cStation de Recherches Sous-marines et Océanographiques, Pointe Revellata, BP33, Calvi 20260, France; a.pere@stareso.com; pierre.lejeune@stareso.com dFaculté des Sciences et Techniques, Université de Corse Pascal Paoli, 7 avenue Jean Nicoli, BP 52, Corte 20250, France eBastia Offshore Fishing Association, 8 Parc Impérial, Furiani 20600, France; jjriutort@gmail.com fOffice de l’Environnement Corse, département "Parc Marin International des Bouches de Bonifacio", base technique de La Rondinara, BP 507, Ajaccio 20176, France; santoni@oec.fr, culioli@oec.fr gSea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver V6T 1Z4, Canada, d.pauly@fisheries.ubc.ca  ABSTRACT Corsica is an island in the Western Mediterranean belonging to France, located southeast of the French mainland and west of Italy. The island covers an area of about 8,700 km 2, is flanked by deep water along its west coast, and by a broad shelf along its east coast. Corsica has fisheries in its coastal lagoons, but its most commercially important fishery is the red spiny lobster (Palinurus elephas) fishery, followed by bottom trawling for finfish. Other smaller and poorly documented artisanal and recreational fisheries also occur, but overall fishing pressure appears to be low, and the number of full time fishers is declining. The total reconstructed catch from 1950 to 2008 was 118,700 tonnes - 5 times more than the 23,700 tonnes reported by France to FAO – of which 30% was unreported recreational catch by locals or tourists, 37% was bottom-trawl catch, 10% was associated bycatch (unreported, landed or discarded), and 23% was red spiny lobster and pelagic catches. The estimated mean annual catch in the 21 st century was 1,300 tonnes. Field investigations are needed to improve on these data, presented here as a first approximation of total extractions from the waters surrounding Corsica.  INTRODUCTION Corsica is the fourth largest island of the Mediterranean and a part of France. It is located southeast of the French mainland and north of Sardinia (Italy), and west of the Italian Peninsula (42° N and 9° E; Figure 1). Corsica is characterized by a mountainous landscape and a highly disparate underwater morphology, featuring a steep descent to depth along the western part of the island (down to 3,000 m, 10 km offshore). In contrast, wide expanses of shallow waters are present along the east coast, where a depth of only 150 m has been recorded 11 km offshore, and several lagoons important for the Island‘s marine fisheries are also found along the east coast (Riutort, 1994).  1  Cite as: Le Manach, F., Dura, D., Pere, A., Riutort, J.J., Lejeune, P., Santoni, M.C., Culioli, J.M., and Pauly, D. (2011) Preliminary estimate of total marine fisheries catches in Corsica, France (1950-2008). pp. 3-14. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727].  4  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Corsican waters host numerous fish and invertebrate species (de Caraffa, 1929; Miniconi, 1989, 2001) and valuable habitats (e.g., meadows of seagrass Posidonia oceanica), of which most are protected under European Commission directives or national legislations (Anon., 1975; 1979, 1981, 1987, 1992, 1994, 1999) (Figure 1). The Réserve Naturelle des Bouches de Bonifacio is the largest marine protected area (MPA) of metropolitan France, covering approximately 800 km² (Figure 1), including a 13 km² notake zone and 130 km² with restricted fisheries activities. Regulations and monitoring seem to be effective, as increasing catches have been reported in and around this MPA (Santoni and Culioli, unpub. data) Since the 1950s, tourists have been attracted to Corsica for its natural beauty and pristine habitats, and the tourist population currently reaches 3 million per year (Anon., 2010), with a resident Corsican population of less than 300,000. The tourism industry has a major impact on seafood consumption and hence on marine resources, as already highlighted in the 1960s (Maurin, 1965). Currently, numerous hotels and/or charter companies offer recreational fishing opportunities, separate or in combination with consumption of local seafood. Despite its potential attractiveness for Figure 1: Map of Corsica and its territorial waters (solid fishers, the waters around Corsica have never black line). Marine Protected Areas (blue areas and grey solid experienced heavy industrial fishing dots) are designated at a state level and aim to protect both pressure, and the history of Corsican habitats and wildlife by controlling, or even excluding human resource extraction was shaped more by activities (no-take zones). Natura2000 zones (blue stripes) land-based than maritime activities. are designated at a European level and aim to protect both Therefore, there is almost no export of habitats and wildlife, without excluding human activities. The seafood out of Corsica, and a substantial level of protection in the Natura2000 zones is lower than for fraction of the seafood consumed locally by MPAs. Source: www.affaires-maritimes.mediterranee.gov.fr Corsicans is imported from the French mainland or other Mediterranean countries. Currently, the number of professional fishers is declining, and Corsica likely experiences the lowest commercial fishing pressure in the Mediterranean Sea (Riutort, 1994; Relini et al., 1999). As a consequence, fisheries have generally not received much attention, and quantitative analyses of fisheries are scarce, except for the high-profile fishery for red spiny lobster (Palinurus elephas) (Pere et al., 2007; 2010) and for MPA fisheries (Rigo, 2000; Santoni, 2002; Mouillot et al., 2007; Rocklin et al., 2009). Corsica - via France - has only supplied fisheries statistics to the Food and Agriculture Organization of the United Nations FAO since 1970. This study therefore aims to reconstruct Corsican fisheries catches back to 1950, while ensuring that all extractions due to fishing are considered, following the catch reconstruction approach of Zeller and Pauly (2007). Like most countries in the Mediterranean, France has not declared a formal Exclusive Economic Zones for its Mediterranean coast (EEZ; Anon., 1976; Santoni, 2002; Cacaud, 2005). Hense, our estimates of historical catches for the period 1950 to 2008 are deemed to have come from Corsican EEZ-equivalent waters.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  5  MATERIALS AND METHODS Baseline data were extracted from the General Fisheries Commission for the Mediterranean (GFCM) section of the FAO FishStat database (FAO, 2009). As Corsica is remote from the French mainland, we assumed that all catches reported by France within the ‗Sardinia‘ FAO fishing area (Division 37.1.3) were Corsican. A bibliographical review of all Corsican fisheries was done to identify the ‗anchor points‘ required for inferences on historical catches back to 1950 (Zeller and Pauly, 2007). Data sources included peer-reviewed scientific articles, reports by local institutions, theses and other unpublished accounts, and local expert knowledge.  Total Corsican population Population statistics were extracted from the National Institute of Statistics and Economic Studies (INSEE, www.insee.fr/fr/themes/theme.asp?theme=2&sous_theme=1&nivgeo=6&type=3 [accessed: October 15, 2010]). Population data were used here to indirectly estimate total catches by local residents (see ‗recreational fisheries: residents and tourists‘ sub-section; Figure 2a).  Fishers and fishing vessels in Corsica The time-series of the population of fishers was obtained from Riutort (1994), and linear interpolations were applied between anchor points for years without data (Figure 2b). The number of fishers after the last anchor point (1993) was calculated by applying the trend in the number of fishers per vessel during the period 1950-1993 to the number of vessels for the period 1994-2008. It is worth noting that these  a)  b)  Figure 2. Basic statistics on Corsica: a) total resident Corsican population and b) trends in the number of fishers and vessels in Corsica. Anchor points are represented by closed circles.  vessels are usually smaller than 15 m, and operate close to shore (Miniconi, 1994; Riutort, 1994; Rigo, 2000; Santoni, 2002). The fishing industry in Corsica is therefore more artisanal than industrial, with small vessels (Riutort, 1994), short periods at sea, and a small supply chain (Riutort, 1989). The two timeseries in Figure 2b were used to estimate bottom-trawl catches for the 1950-1970 period (see ‗artisanal demersal fishery‘ sub-section).  Lobster fishery Red spiny lobster (Palinurus elephas) is mainly exploited along the west coast of Corsica, where its preferred hard-bottom habitats are found. The fishery for lobster is relatively small, and vessels stay close to the coast, fishing at depths not exceeding 200 m (Marin, 1987). The fishery was profitable very early on (de Caraffa, 1929), but it is not well documented. Thus, official catch data are deemed inaccurate, and is at best a refelction of trends (Marin, 1987). Here, we attempt to re-estimate lobster catches for the entire 1950-2008 period using various sources of information.  6  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Statistics were extracted from the Office de l’Environnement Corse (2010) for the 1950-1983 period. The early values (1950s) are in accordance with catches at the beginning of the 20 th century, i.e., around 300 t·year-1 (Doumenge, 1956). For the 1983-2008 period, values were extracted from studies by Riutort (1999), Marin (1987) and Pere et al. (2010). Changes in gears had an influence on both catch per unit of effort (CPUE) and discard rates (nonmarketable lobsters). In the early 1960s, the traps in common use were replaced by trammelnets (Miniconi, 1989), which had higher CPUE, but also generated a higher discarding rate. Discards for traps were estimated to be about 5% during the 1950-1964 period (Riutort, unpub. data). During the 1965-1980 period, fishers were using trammelnets for short trips, and a discard rate of 12.5% was therefore used (Riutort, unpub. data). For the 1981-1994 period, the mean value of 15.7% between the 1965-1980 period (12.5%) and the 1995-1999 period (20%) was used (Riutort, 1999; Pere et al., 2010). For the 2000-2003 period, the same value of 12.5% was used. For the 2004-2007 period, Pere et al. (2010) estimated a discarding rate of 11.4%, which was also used for the year 2008. It is worth noting here that two types of trammelnets are used in Corsica, to target either demersal fish (since before 1950) or red spiny lobster (since the early 1960s). Even when ‗lobster trammelnets‘ are used, a considerable amount of the bycatch is fish. Thus, 55% of total catch were fish species in 2008 (Riutort, 1989; 1994; Santoni and Culioli, unpub. data). This bycatch of fish is retained and landed, and was included in the next sub-section (artisanal demersal fisheries).  Artisanal demersal fisheries Demersal species are caught in Corsican waters with two types of gears: trammelnets and bottomtrawlers. Trammelnets have been in use for demersal fish for a long time (prior to 1950), while we assumed bottom-trawlers were introduced in the early 1950s (Riutort, 1994). Catches by trammelnets may represent 50% of total fish catches in the province of Bonifacio (Santoni and Culioli, unpub. data), and given that no other studies were available, we used this 50% ratio for the 1965-2008 time-period and the entire island. Thus, the remaining 50% of demersal fish catches were treated as caught by bottomtrawlers as of 1965. For 1950, we set bottom-trawl fish catches as zero, and interpolated linearly to 1965. FAO FishStat contains data on demersal species for the period 1970-1992 only. In the absence of any alternative, we considered these data to be realistic. Indeed, none of the documents available on Corsican fisheries allowed us to make an independent estimate of the bottom-trawl and trammelnet fisheries catches. Catches per fisher and catches per vessel (CPUE) for the period 1970-1992 were calculated by dividing catches of bottom-dwelling fish reported to FAO by the number of fishers or vessels (Figure 2b). CPUE for the 1950-1970 and 1993-2008 periods were then estimated by extrapolation of the trends of 1970-1992 CPUE time-series. The resulting CPUE data for the 1950-1970 and 1993-2008 periods were then multiplied by the number of fishers or vessels (Figure 2). Our estimate of total catch used the average values of these two catch time-series (one based on CPUE per fisher, one on CPUE per boat), which was then split evenly to create the bottom-trawl and trammelnets components. Trammelnet fishery The taxonomic breakdown of trammelnet commercial species in the Bonifacio MPA (Figure 1) was studied by Mouillot et al. (2007) from 2000 to 2006. Given that there were no other studies available that included a taxonomic breakdown, we assumed that the percentage of each species remained the same for the entire 1950-2008 time-period, and were similar for the entire island. A recent study concluded that trammelnet discards were representing approximately 10% of total catches, in the MPA of Bonifacio (Rocklin et al., 2009). These discards are composed of damaged, non-marketable fish. We used this study to estimate the taxonomic breakdown of these discards.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  7  Bottom-trawl fishery We assumed that the species composition of landed bottom-trawl catches were similar to the trammelnet fishery. However, we acknowledge that the species composition can vary significantly between the continental shelf and the slope. For higher depths (on the slope), many other species such as Nephrops norvegicus, Etmopterus spinax, Galeus melastomus, Merluccius merluccius or Trigla lyra can indeed account for a large part of the catch (Riutort, pers. obs.; Le Manach, pers. obs.). However, as it was not possible to estimate the percentage of each species or the importance of slope bottom-trawling, we did not take these observations into account. To estimate the bycatch by the artisanal bottom-trawl fishery, we used a bycatch rate of 40%, given by Machias et al. (2001) and Sanchez et al. (2004) for geographically close and similar fisheries. The MEDITS database (Bertrand et al., 1998) - 2009 update - was used to estimate the bycatch taxonomic breakdown. We assumed that non-commercial species occurring in this database were bycatch species, e.g., Spicara spp., Scyliorhinus spp., Raja spp., Micromesistius poutassou, Capros aper. Furthermore, as fishers land a portion of non-targeted bycatch, notably for their personal consumption and soupe de roche (‗rockfish soup‘), we conservatively assumed that 20% of the bycatch was landed, but unreported, and that the remainder (80%) was discarded.  Recreational fisheries: residents and tourists To estimate recreational catches by Corsicans, we used a ‗Fermi solution‘, i.e., an approach pioneered by the physicist Enrico Fermi, to estimate unknown quantities from limited data (von Baeyer, 1993; Pauly, 2010). Thus, based on local knowledge (Culioli, pers. obs.; Riutort, pers. obs.), we estimated three anchor points, for 1950, 1980 and 2008. For 1980, we assumed that 30% of the total population, i.e., 76,000 out of 255,000 inhabitants, was potentially recreational fishers. Of these potential fishers, we assumed that 15% were actually fishers, and that they were on average fishing once a month, with yields of 4 kg per trip. For 2008, we used the same assumption that 30% of the total population, i.e., 84,000 out of 280,000 inhabitants, were potentially recreational fishers, but that the proportion of actual fishers increased to 25%. As local residents report that there are less fish now than in the 1980s, we assumed that fishers currently fish on average only 10 times a year, with yields of 1 kg per trip. For 1950, we assumed a stable CPUE and fishing effort compared to 1980, and derived total catches from the total population size. Similarly, our estimate of recreational catches by tourists was based on the annual number of tourists, and assuming that sport fishing became more attractive in the 1990s. We conservatively assumed that 5% of tourists were catching on average 1 kg·year-1 for the 1950-1990 period, and that 8% of tourists were catching on average 1.5 kg·year-1 for the 1991-2008 period. Given that each tourist currently stays on average 10.3 days in Corsica, these assumptions seem reasonable (Anon., 2010).  Pelagic fisheries Three pelagic fisheries are taking place in Corsican waters. However, information is scarce and no studies enabled us to re-estimate their total catches. Therefore, we included data as provided to FAO in our total reconstruction (except for small pelagics – see below). Swordfish Swordfish (Xiphias gladius) started to be targeted by artisanal longliners in the 1980s (Regional Committee of Corsican Marine Fisheries, 2009; Riutort, unpub. data). However, as tonnages are likely small (15-20 t·year-1; Riutort, unpub. data), it is possible that these catches are accounted for in official FAO statistics as ‗marine fish nei‘ (FAO, 2009). Small pelagics ‗Blue fish‘ (i.e., sardines, anchovies and mackerels) are also fished along the Corsican coast. Several studies report substantial catches during the 1960s and 1970s in Corsica and along the French mainland (Maurin, 1965; Bonnet, 1973; Pichot and Aldebert, 1978). It is worth noting that FAO data include sardine statistics only for 1972-1976 and 2006; data for other years being either non-existent or unrealistically  8  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  low. However, older Corsican residents remember very abundant sardine and anchovy catches during the 1950-1960s, most of which were exported to the mainland (Riutort, unpub. data). For the period 19501971, we therefore used the average catches for the period 1972-1976, and kept the rest of the time-period unchanged. Tuna Maurin (1965) reports 100 tonnes of tuna caught in 1963 by Corsicans. However, he suggests that the tourism industry already accounted for a significant part of unreported catches, although he did not elaborate on this topic. Tuna may also be reported to FAO as ‗marine fish nei‘, and annual catches are likely very low or up to 15 tonnes (Riutort, unpub. data).  RESULTS AND DISCUSSION Lobster fishery As expected, our lobster catch reconstruction is very different from official statistics: our values are on average 16 times higher than data provided to FAO, and show a very different pattern over time (Figure 3). Lobster catches decreased from 300 t·year-1 in 1954 to 80 t·year-1 for 1959-1960. Then, catches increased again to 300 t·year-1 by 1962. At this time, a new (unspecified) crash occurred and catches dropped to 100 tonnes annually, staying at that level until the late 1970s. By Figure 3: Reconstructed catches of lobsters and associated discards for Corsica, 1950-2008. The dotted line represents 1984, catches increased to 250 t·year-1. Since official lobster landings data supplied to FAO. then, catches have been decreasing, reaching -1 80 t·year by the early 2000s. However, it is worth noting that catches currently seem to be increasing. Overall, the trend of the number of fishers and vessels (Figure 2b), and lobster catches (Figure 3) show a similar pattern, which confirms that this fishery is of great importance in Corsica and largely accounts for much of the fishing pressure. Fluctuations in lobster catches may be partly explained by new policies and gear modifications, along with biological features (e.g., larval migration; Pere et al., 2011). The first crash in the 1950s likely resulted from increasing fishing pressure, and the following increase in catches is likely the result of a gear change from traps to trammelnets, which increased the CPUE. In 1968, policymakers decided to close the lobster fishery between the 1st of October and 28th of February each year, which probably played a significant role in the stabilization of catches during the 1970s. Finally, new vessels were introduced in the early 1980s and were responsible for increased fishing effort. This increase is likely to have contributed to the increase in catches until 1984, then to the decrease in catches observed until the mid2000s.  Figure 4: Reconstructed catches and associated bycatch (landed and discarded) for the demersal finfish fishery by the two main gears (trammelnet and bottom-trawler) in Corsica, 1950-2008. The dotted line represents official landings data supplied to FAO.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  9  In the late 1970s, nine small marine protected areas (some of them no longer existing; Figure 1) were created, which also possibly led to the catch increases in the late 1970s early 1980s (Figure 3). The current decrease, which started around 1984, may be due to several factors, such as the decreasing number of fishers, or increasing fishing effort. It seems there is an increase in catches during the last few years. Such an increase could be due to biological parameters (e.g., larval migration), but no data were available to assess the validity of this assumption.  Artisanal demersal fishery Demersal catches totaled an estimated 56,500 tonnes, compared to only 18,800 tonnes reported to FAO (Figure 4). Catches fluctuated, but declined overall from approximately 1,300 t·year-1 in 1950 to 500 t·year-1 in the late 2000s. Bycatch followed a different trend, totaling 10,300 tonnes and peaking around 250 t·year-1 in the 1980s due to the increasing number of trawlers. Bycatch amounts in the 1950s and the 2000s are similar, slightly above 100 t·year-1 (Figure 4). Unlike our reconstruction for the lobster fishery (above), the reconstruction of artisanal bottom-trawl fishery catches was mainly based on official statistics. The main novelty in this result comes from the gaps in time-series originally supplied to FAO being filled in. Also, a significant part of total catches (18%) were previously unreported and are now reported as bycatch (either discarded or landed; Figure 4).  Figure 5: Reconstructed catches by recreational fishers in Corsica, 1950-2008. Anchor points are indicated by closed circles.  Recreational fisheries: residents and tourists Recreational fisheries were estimated to catch 35,150 tonnes, of which 80% was taken by local resident fishers, and 20% by tourists (Figure 5). These catches were previously not included in statistics provided to FAO.  Figure 6: Catches in Corsican waters, showing a) reconstructed total catches versus landings data as supplied to FAO; and b) taxonomic breakdown (top 10 species) of reconstructed total catches in Corsica, 1950-2008. The ‗others‘ group includes Scorpaena scofa, Phycis spp., Pagellus, spp., Labridae, Serranus spp., Seriola dumerili, Zeus faber, as well as other fish species of lower importance in term of percentage, and species of invertebrates.  10  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Based on our assumptions, we estimated that recreational catches by residents were the highest in 1950, with 612 t·year-1, and then declined to 210 t·year-1 by 2008 (Figure 5). On the other hand, recreational catches by tourists were estimated to have increased during the last two decades, increasing from 17 t·year-1 in 1950 to 360 t·year-1 by 2008 (Figure 5).  Overall reconstruction Reconstructed total Corsican fisheries catches total over 118,700 tonnes since 1950, compared to only 23,700 tonnes reported to FAO by local fisheries authorities. Overall, total catches appear to be steadily decreasing from approximately 2,800 t·year-1 in 1950 to 1,200 t·year-1 by 2008, interrupted by a peak catch of over 3,000 t·year-1 in 1975 (Figure 6a and Appendix Table A1). This decrease seems linked to the decline in both fishers and vessel numbers, but also to declines in fish abundance along the Corsican coast. Official statistics likely accounted for commercial (artisanal) fisheries only, that is, red spiny lobster and bottom-trawl fisheries. Recreational fisheries by Corsicans, or by tourists, were not considered by official authorities. Finally, we highlighted the existence of discards (for red lobster and bottom-trawl fisheries), which are generally not included in reported statistics (Zeller et al., 2011). This improved accounting of total catches (versus reported commercial landings) is also evident in the improved taxonomic accounting provided by our study (Figure 6b). Data reported by FAO on behalf of Corsica, besides being of poor quality, also had a poor taxonomic breakdown. Species present in these official data were reported as arbitrary, according to the local fisheries literature. In contrast, we have been able to assign catches to over 30 taxa, of which each had catches allocated in accordance to the literature (Figure 6b and Appendix Table A2). This study provides an estimate of total fisheries catches in Corsican waters since 1950, and although some sectors such as the pelagic fisheries have not been dealt with in detail, two major conclusions emerge from our work: (1) historical events, changes in gear and emergence of new fisheries illustrate that, despite being assumed to be one of the areas of the Mediterranean with the lowest fishing pressure (Riutort, 1994; Relini et al., 1999), Corsican waters may be exposed to higher fishing pressure than previsously assumed; and (2) our results suggest that Corsicans seem to be much more involved in marine resource exploitation than it appears in the literature and in official statistics.  ACKNOWLEDGEMENT We thank the Sea Around Us project, a scientific collaboration between the University of British Columbia and the Pew Environment Group. We also thank Jacques Bertrand (Ifremer) for sharing information and knowledge.  REFERENCES Anon. (1975) Décret n° 75-1128 du 9 décembre 1975 portant création de la réserve naturelle de Scandola (Corse). Journal Officiel de la République Française: 12612. Anon. (1976) Loi n°76-655 du 16 juillet 1976 relative à la zone économique au large des côtes du territoire de la république. Journal Officiel de la République Française: 4299. Anon. (1979) Council Directive 79/409/EEC of 2 April 1979 on the conservation of wild birds. Official Journal L 103: 001-0018. Anon. (1981) Décret n° 81-205 du 3 mars 1981 portant création, délimitation, réglementation et gestion (par le préfet) de la réserve naturelle des iles Cerbicale (Corse du Sud). Journal Officiel de la République Française: 704. Anon. (1987) Décret n° 87-494 du 29 juin 1987 portant création de la réserve naturelle des iles Finocchiarola (Haute Corse). Journal Officiel de la République Française: 7364. Anon. (1992) Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora. Official Journal L 206: 7.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  11  Anon. (1994) Décret n° 94-688 du 9 août 1994 portant création de la réserve naturelle de l'étang de Biguglia (Haute Corse). Journal Officiel de la République Française: 11770. Anon. (1999) Décret du 23 septembre 1999 portant création de la réserve naturelle des Bouches de Bonifacio (Corse du Sud). Journal Officiel de la République Française: 14243. Anon. (2010) Résultats et enseignements de l‘année touristique 2009. Agence du tourisme Corse. Bertrand, J.A., Gil de Sola, L., Papaconstantinou, C., Relini, G. and Souplet, A. (1998) An internation bottom trawl survey in the Mediterranean: the MEDITS programme. Actes de colloques IFREMER 26: 76-93. Bonnet, M. (1973) Les pêches maritimes sur le côtes françaises de la Méditerranée – actualités, perspectives. Science et Pêche 222: 18. Cacaud, P. (2005) Fisheries laws and regulations in the Mediterranean: a comparative study. Studies and reviews 75. General Fisheries Commission for the Mediterranean. United Nations Food and Agriculture Organization (FAO), Rome (Italy), 40 p. de Caraffa, T. (1929) poissons de mer et la pêche sur les côtes de la Corse. Laffitte Reprints (1980), Marseille, 336p. Doumenge, F. (1956) Problèmes d‘équipement de la pêche en Corse. Bulletin de la Société Languedocienne de Géographie, 28(4). FAO (2009) FishStat Plus - Universal software for fishery statistical time series. v2.3. United Nations Food and Agriculture Organization (FAO), Rome (Italy). Machias, A., Vassilopoulou, V., Vatsos, D., Bekas, P., Kallianiotis, A., Papaconstantinou, C. and Tsimenides, N. (2001) Bottom trawl discards in the northeastern Mediterranean Sea. Fisheries Research 53(2): 181-195. Marin, J. (1987) Exploitation, biologie et dynamique du stock de langouste rouge de Corse, Palinurus elephas Fabricus. Thèse de doctorat d‘état. Université d‘Aix Marseille II. Faculté des sciences de Luminy, 327p. Maurin, C. (1965) La pêche française de 1964 en Méditerrannée. Science et Pêche 134: 1-4. Miniconi, R. (1989) Les poissons et la pêche en Corse. Thèse de doctorat d‘état. Université d‘Aix Marseille II. Faculté des sciences de Luminy, 504 p. Miniconi, R. (1994) Les poissons et la pêche en Méditerranée : la Corse. Piazzola and La Marge, Ajaccio, France. Miniconi, R. (2001) Les poissons de Corse - Biologie, pêche, appellations. 262 p. Mouillot, D., Tomasini, J.A., Culioli, J.M. and Do Chi, T. (2007) Développement durable de la pêche artisanale sur le site de la Réserve Naturelle des Bouches de Bonifacio (Corse du sud). Programme MEDD LITEAU 2 : Gestion intégrée des zones côtières. Rapport final, 66 p. Pauly, D. (2010) Five easy pieces: how fishing impacts marine ecosystems. Island Press, Washington, DC, 193 p. Pere, A., Lejeune, P. and Pelaprat, C. (2010) Suivi scientifique de la pêche langoustière corse - Rapport final - Années 2004-2009. Contrat Office de l‘Environnement de la Corse - Stareso, 106 p. Pere, A., Pelaprat, C. and Lejeune, P. (2011) Overview of the corsican fishery and presentation of series of available data. 1st LanConnect Workshop. 21-23 March 2011. Palma de Mallorca (Spain). Pere, A., Pelaprat, C., Pergent-Martini, C., Livrelli, J.N. and Lejeune, P. (2007) The spiny lobster fishery in Corsica. 8th International Conference and Workshop on Lobster Biology and Management. September 23-28 2007. Charlottetown (Canada). Pichot, P. and Aldebert, Y. (1978) La pêche de la sardine en Méditerranée française. Science et Pêche 277: 16. Regional Committee of Corsican Marine Fisheries (2009) Corse - l'interdiction de pêcher l'espadon suscite des vagues. Corse Matin. Relini, G., Bertrand, J.A. and Zamboni, A. (1999) Synthesis of the knowledge on Bottom fishery resources in central Mediterranean (Italy and Corsica). Biologia Marina Mediterranea 6 (suppl. 1). 868 p. Rigo, D. (2000) La pêche professionnelle dans la Réserve Naturelle des Bouches de Bonifacio, effort et productions. Rapport de D.E.S.S. Université de Corse, 53 p. Riutort, J.J. (1989) Premières estimation des captures et de l'effort de pêche déployé par les « petits métiers » sur le littoral nord-ouest de la Corse. Etude de la biologie des principales espèces cibles. Rapport Région de Corse, 151 p. Riutort, J.J. (1994) La pêche en Corse : le rouget de roche, espèce cible. Project Stareso/EEC XIV - 1/MED/91/006, 167 p. Riutort, J.J. (1999) Pêche de la langouste rouge (Palinurus elephas), en Corse (production, sélectivité des engins de pêche, effet réserve des cantonnements à crustacés). Rapport final. 117 p. Rocklin, D., Santoni, M.C., Culioli, J.M., Tomasini, J.A., Pelletier, D. and Mouillot, D. (2009) Changes in the catch composition of artisanal fisheries attributable to dolphin depredation in a Mediterranean marine reserve. ICES Journal of Marine Science 66(4): 699-707.  12  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Sanchez, P., Demestre, M. and Martin, P. (2004) Characterisation of the discards generated by bottom trawling in the northwestern Mediterranean. Fisheries Research 67(1): 71-80. Santoni, M.C. (2002) Evolution de l‘effort et des productions de la pêche artisanale sur le site de la Réserve Naturelle des Bouches de Bonifacio. Office de l‘Environnement de la Corse : Réserve Naturelle des Bouches de Bonifacio. Rapport de D.E.S.S. Université de Corse, 64 p. von Baeyer, H.C. (1993) The Fermi solution: essays on science. Random House, New York, 176 p. Zeller, D. and Pauly, D., eds. (2007) Reconstruction of marine fisheries catches for key countries and regions (19502005). Fisheries Centre Research Reports 15(2). Fisheries Centre, University of British Columbia, Vancouver, 163 p. Zeller, D., Rossing, P., Harper, S., Persson, L., Booth, S. and Pauly, D. (2011) The Baltic Sea: Estimates of total fisheries removals 1950-2007. Fisheries Research 108(2-3): 356-363.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  13  Appendix Table A1: Annual catches by Corsican fisheries. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  Domestic fisheries (t) Data reported to FAO 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 945 784 1280 966 1719 2128 1632 625 759 607 1012 899 1100 986 967 1042 895 811 1093 604 574 584 502 220 137 99 96 85 74 59 28 26 22 21 23 15 281 0 0  Reconstructed catches 2759 2751 2744 2736 2728 2666 2603 2540 2475 2478 2484 2634 2740 2698 2657 2615 2506 2427 2422 2421 2373 2204 2206 1757 2772 3195 2647 1457 1662 1450 1968 1836 2109 1956 1930 1934 1746 1672 2094 1473 1462 1469 1358 1540 1482 1468 1448 1433 1418 1404 1370 1342 1314 1287 1259 1243 1482 1194 1226  14  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Appendix Table A2: Six most important taxa caught by domestic fisheries in Corsica‘s EEZ, 1950-2008. Year  Palinurus elephas  Sardina pilchardus  Diplodus spp.  Dentex dentex  Scorpaena scrofa  Phycis spp.  Othersa  1950 375 387 189 171 165 106 1951 373 387 187 169 163 104 1952 372 387 186 167 160 102 1953 370 387 185 165 157 101 1954 369 387 184 164 154 99 1955 325 387 183 161 150 96 1956 281 387 182 159 147 94 1957 237 387 181 157 143 91 1958 193 387 180 154 139 89 1959 153 387 179 156 143 91 1960 153 387 178 155 142 91 1961 259 387 177 156 145 93 1962 363 387 176 154 143 92 1963 324 387 176 153 141 90 1964 285 387 176 153 139 89 1965 246 387 176 152 137 88 1966 204 387 176 148 130 83 1967 163 387 176 146 126 81 1968 163 387 176 146 125 80 1969 163 387 178 147 125 80 1970 160 387 179 145 119 76 1971 150 387 182 137 100 64 1972 146 420 184 136 95 61 1973 133 265 186 125 70 45 1974 184 348 189 170 164 105 1975 192 548 191 178 178 114 1976 179 328 188 165 154 99 1977 132 70 184 124 68 44 1978 150 2 181 137 102 65 1979 154 3 177 125 78 50 1980 199 15 174 147 131 84 1981 209 3 172 140 118 75 1982 241 0 169 152 146 93 1983 248 0 167 143 129 82 1984 263 6 165 140 124 79 1985 191 6 162 143 134 86 1986 178 13 160 133 114 73 1987 169 13 165 131 105 67 1988 190 13 171 154 148 95 1989 150 7 179 126 80 51 1990 145 6 188 129 76 49 1991 142 7 187 129 77 50 1992 133 7 186 124 66 42 1993 141 7 185 133 86 55 1994 135 6 184 130 81 52 1995 128 8 183 129 80 51 1996 129 4 182 128 79 51 1997 130 4 181 127 78 50 1998 132 4 180 125 76 49 1999 133 4 180 124 75 48 2000 125 0 179 123 73 47 2001 118 0 178 122 71 45 2002 110 0 177 120 69 44 2003 103 0 176 119 67 43 2004 95 0 175 118 65 42 2005 98 0 174 116 63 41 2006 83 192 173 115 62 39 2007 88 0 172 114 60 38 2008 135 0 171 112 58 37 ‘Others’ comprises Pagellus, spp., Labridae, Serranus spp., Spicara spp., Raja spp., other clupeiformes, Mullus spp., Scorpaena, Maja squinado, Sepia spp., Homarus gammarus, Lophius spp., Capros aper, Micromesistius poutassou, Scyliorhinus spp., other miscellaneous marine fish, cephalopods and crustaceans.  1366 1368 1369 1371 1371 1363 1354 1343 1332 1368 1378 1416 1424 1426 1428 1429 1377 1347 1343 1342 1306 1184 1164 933 1612 1794 1534 835 1025 862 1217 1120 1307 1187 1154 1212 1076 1022 1325 879 869 876 799 932 894 888 875 864 852 841 824 809 794 780 765 751 819 723 712 other  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  15  A BRIEF HISTORY OF FISHING IN THE KERGUELEN ISLANDS, FRANCE1 M.L.D. Palomares and D. Pauly  Sea Around Us Project, Fisheries Centre, University of British Columbia, 2204 Main Mall, Vancouver, V6T1Z4, Canada m.palomares@fisheries.ubc.ca; d.pauly@fisheries.ubc.ca  ABSTRACT Catch statistics from around the (uninhabited) Kerguelen Islands, which are part of the French Antarctic and sub-Antarctic Territories, and where distant-water fisheries began in 1970, were obtained from the CCAMLR (Commission for the Conservation of Antarctic Marine Living Resources) Statistical Bulletin (Area 58.5.1) and complemented by statistics reported through the French KERPECHE program. Catches originally expressed by fishing seasons were re-expressed as calendar years, which results in a slight between-season smoothing. These catches show a general decline over a 30 year-period and an expansion of the longline fishery to deeper waters in the last 10 years.  INTRODUCTION The Kerguelen Islands The Kerguelen Islands (49°30‘S, 69°30‘E) are part of the French Antarctic and sub-Antarctic Territories, which also include the islands of Crozet, Amsterdam and St. Paul, and the Antarctic district of Terre Adélie (www.taaf.fr). They consist of a main island called ‗La Grande Terre‘ (6,700 km2) and a number of smaller surrounding islets. Kerguelen Island sits in the middle of the combined shelf of the Kerguelen and Heart Islands (Australia), known as the Kerguelen Plateau, which covers an area of 100,500 km2 above 500 m depth (Pruvost et al., 2005: see Figure 1). The islands are uninhabited both because of their isolated locations and the extreme climate prevailing in the area.  Figure 1. Map of Kerguelen Islands, CCAMLR areas 58.5.1, showing French (Kerguelen Islands) and Australian (Heart Island) Exclusive Economic Zones, as well as 500 m and 1000 m depth contour.  Fisheries and their resource species Fishery prospecting cruises (mostly by the USSR, i.e., the Ukraine; Zeller and Rizzo 2007) in the 1960s led to the development of a modern fishery in the Kerguelen Islands starting in 1970 with about 10 Ukrainian bottom trawlers operating during 6-month fishing seasons without management or control. They targeted marbled rockcod (Notothenia rossii), mackerel icefish (Champsocephalus gunnari) and gray rockcod (Lepidonotothen squamifrons), and also caught unspecified by-catch species of the plateau, at 200-500 m depths in what is now known as CCAMLR area 58.5.1 (Pruvost et al., 2005). This unmanaged fishery continued until France declared an Exclusive Economic Zone (EEZ) around the islands as well as the 1  Cite as: Palomares, M. L. D. and Pauly, D. (2011) A brief history of fishing in the Kerguelen Islands, France. pp. 15-20. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN1198-6727]  16  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  other ‗Terres Australes et Antarctiques Françaises‘ in 1978 (TAAF; see Duhamel, 1995). Since the implementation of the French management scheme in 1980, foreign fleets could access the Kerguelen Islands‘ EEZ only through access agreements for which the French government granted quotas, limiting to 7 the number of trawlers operating at any one time (Pruvost et al., 2005; Duhamel, 1995). The discovery of a large stock of Patagonian toothfish, Dissostichus eleginoides, by USSR bottom trawlers in the 1984-1985 fishing season on the slopes of the Kerguelen shelf led to the development of this highvalue trawl fishery. In 1996, the former USSR stopped trawling in Kerguelen waters and only 2 Ukrainian longliners and 2 French bottom trawlers remained (Pruvost et al., 2005). In the same year, a joint French and Japanese prospecting cruise aboard the M/V Anyo Maru established that longlining was an effective method for catching Patagonian toothfish (Duhamel and Hautecoeur, 2009), which led to the development of this fishery, completely replacing the bottom trawl fishery in the 2000-2001 fishing season (Lord et al., 2006). The high initial abundance of this stock encouraged a rapid expansion of the longline fisheries and the subsequent proliferation of non-licensed longliners from non CCAMLR member states (Kock, 2001). The illegal fishery catch peaked between 1996 and 2004, with catches reaching four times that of the regulated catch in 1997 (Agnew, 2000). In 2005, illegal fishing was curtailed and the fishery was limited to 7 French longliners (Pruvost et al., 2005). Overall, both trawl and longline fisheries in the Kerguelen Islands increased their effort throughout the period considered here, i.e., 1970-2005. Their catch per unit of effort has consequently strongly diminished, in spite of the expansion of the longline fishery from an average fishing depth of 500 m to 1,000 m (Lord et al., 2006). This suggests massive declines in the target fish biomass; the mean individual size of Patagonian toothfish has also declined (Duhamel and Hautecoeur, 2009).  MATERIAL AND METHODS The Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR) catch statistics for CCAMLR area 58.5.1 were used as a basis of this catch reconstruction for the period 1970-2010. These statistics were missing data for the period 1979-1987. Lord (2005) provided statistics for: (i) catches of Ukrainian and French trawlers for Patagonian toothfish, marbled rockcod, mackerel icefish and grey rockcod fisheries for the period 1979-2001; (ii) longline catches for the Patagonian toothfish fishery and its by-catch (mainly rays and grenadiers) for the period 1990-2003 using the French KERPECHE database (see Lord et al., 2006 and Pruvost et al., 2005); and (iii) catch estimates from illegal fishing operations based on recorded arrests for the period 1996-2003. In those cases where the CCAMLR and KERPECHE statistics overlapped, the KERPECHE catches ranged from 73% (marbled rockcod trawl fishery) to 204% (rays as by-catch of the longline fishery) of the CCAMLR statistics. We used the KERPECHE statistics in lieu of the CCAMLR statistics for: (a) trawlers for the period 1979-1989 for mackerel icefish, marbled and gray rockcod, and 1979-1991 for Patagonian toothfish; and (b) longliners, notably during the period of expansion for Patagonian toothfish, in order to be able to include some of the unreported catches. Catches reported by the USSR to the CCAMLR were all assigned to the Ukraine because this part of the world‘s ocean was exploited by USSR vessels from the Ukrainian SSR during the Soviet era (see Romanov, 2003; Zeller and Rizzo 2007). Also, it is only Ukrainian vessels which exploited the Kerguelen following the breakup of the USSR (see Pruvost et al., 2005).  RESULTS AND DISCUSSION Appendix Table A1 and A2 present a summary of the catch statistics available from the Kerguelen Islands extracted from the CCAMLR (2010) and completed with data from Lord (2005, Annex 3). The catches originally presented by CCAMLR ‗season‘, from the 1 st of July of a particular year to the 30th of June of the next year, were converted to calendar years by assuming that the catch in the first half of the season (in a given year) is equal to that of the second half of the season (in the next year). This does not affect cumulative catches and, in fact, corresponds to a slight between-season smoothing.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  30.0  a) Notothenia rossii  Catch (t x 103 )  25.0  Notothenia squamifrons  20.0 15.0  Dissostichus eleginoides  10.0  Champsocephalus gunnari  5.0 0.0 7.01970  b)  6.0  Catch (t x 103 )  This brief account of the fisheries in the Kerguelen Islands is meant to present the Kerguelen Island fisheries statistics in such a way that they can be included in the Sea Around Us project‘s (www.seaaroundus.org) catch mapping procedure (see Watson et al., 2004). This is the reason why the catch is reported by calendar years and not as done in the original literature, which account for fishing ‗seasons‘. Moreover, we include estimates of illegal catches, which although highly tentative, are likely to be more correct than the statistically very precise but inaccurate estimate of zero commonly used as a replacement for difficult to estimate quantities such as illegal catches (see Zeller et al., 2011).  17  5.0  4.0  1980  Macrourus 1990 2000 carinatus 2010 other Macrourus spp Year Rajiformes  Notothenia squamifrons Notothenia rossii  3.0 2.0  Dissostichus eleginoides  Catch (t x 103 )  Champsocephalus 1.0 gunnari The resulting catch statistics for the Ukraine (see Appendix Table A2 and 0.0 Figure 2a), i.e., the fishery which 35.01970c) 1980 1990 2000 2010 heavily exploited mackerel icefish over three decades, show peaks and 30.0 Year troughs similar to patterns reported Ukraine 25.0 for South Georgia, South Orkney Islands, Elephant Island and South 20.0 Shetland Islands (Kock, 1991). Illegal France 15.0 Heavy fishing pressure on the strong 1973-1974 year classes may have 10.0 reduced the stock size to a level that 5.0 prevented adequate recruitment and thus recovery (Anon., 2001). The 0.0 trend of the peaks shows a steady 1970 1980 1990 2000 2010 decline in the catch, and Kock and Everson (2003) concluded that this Year decline is the result of a combination Figure 2. Reconstructed fisheries catches (in metric tonnes) for of factors, including heavy fishing the Kerguelen Islands (CCAMLR Area 58.5.1) with statistics pressure, changes in the abundance adapted from CCAMLR (2010) and complemented with data from of icefish predators (Antarctic fur Lord (2005) for ;(a) Ukrainian (i.e., former USSR and Russian seals and penguins) and prey (krill), Federation statistics); (b) French trawlers and long-liners; and (c) and warming in the northern parts total catch by country including some estimates of illegal fishing of the distributional range of icefish. during the period 1996-2003 from Lord (2005). Commercial fishing for mackerel icefish was banned at the end of the 1980s (Kock, 1991) resulting in the tapering off of statistics reported by the Ukraine. The increasing French catch trend (see Appendix Table A1 and Figure 2b), on the other hand, reflects an exploratory fishery tending towards expansion to deeper waters. Duhamel et al. (1997) speculated that the level of longline bycatch (mainly of rays and rattails or grenadiers, Family Macrouridae) have the potential to replace the Patagonian toothfish fishery. Although smaller and subjected to management and monitoring, this expanding fishery has effectively ‗counterbalanced‘ its decreasing catch per unit of effort.  18  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Overall, it is masking the fact that no new fishing grounds have been found since all accessible shallow shelf stocks have been over exploited (Lord et al., 2006). The illegal catch estimates shown in Figure (2c) that were reported both by the CCAMLR (2010) and by Lord (2005) may well be underestimates, i.e., catches may be twice (or more) of those reported to the CCAMLR.  ACKNOWLEDGEMENTS The Sea Around Us Project is a collaboration between the University of British Columbia and the Pew Environment Group.  REFERENCES Agnew, D.J. (2000) The illegal and unregulated fishery for toothfish in the Southern Ocean, and the CCAMLR catch documentation scheme. Marine Policy 24: 361-374. Anon. (2001). Report of the Workshop on Approaches to Management of Icefish. In CCAMLR of the Twentieth Meeting of the Scientific Committee, Annex 5, Appendix D, pp. 467-515. CCAMLR, Hobart, Australia. CCAMLR, (2010). CCAMLR Statistical Bulletin 2010. Volume 22 (Database Version; www.ccamlr.org). CCAMLR, Hobart, Australia. Duhamel, G. (1995). Gestion des pêches aux îles Kerguelen. Recherche Marine 13: 16-17. Duhamel, G., Hautecoeur, M. (2009). Biomass, abundance and distribution of fish in the Kerguelen Islands EZ (CCAMLR Statistical Division 58.5.1). CCAMLR Science 16: 1-32. Duhamel, G., Pruvost, P., Capdeville, D. (1997) By-catch of fish in longline catches off the Kerguelen Islands (Division 58.5.1) during the 1995/1996 season. CCAMLR Science 4: 175-193. Pruvost, P., Duhamel, G., Palomares, M.L.D. (2005) An ecosystem model of the Kerguelen Islands‘ EEZ. p. 40-64 In: M.L.D. Palomares, P. Pruvost, T.J. Pitcher and D. Pauly, D. (eds.), Modeling Antarctic Marine Ecosystems. Fisheries Centre Research Reports 13(7). Fisheries Centre, University of British Columbia, Vancouver, Canada. Kock, K.-H. (1991) The state of exploited fish stocks in the Southern Ocean – a review. Archiv für Fischereiwissenschaften 41: 1-66. Kock, K.-H. (2001) The direct influence of fishing and fishery-related activities on non-target species in the Southern Ocean with particular emphasis on longline fishing and its impact on albatrosses and petrels – a review. Reviews in Fish Biology and Fisheries 11: 31-56. Kock, K.-H., Everson, I. (2003) Shedding new light on the life cycle of mackerel icefish in the Southern Ocean. Journal of Fish Biology 63: 1-21. Lord, C. (2005) Étude d‘une population exploitée de légines (Dissosticus eleginoides) aux Îles Kerguelen. Mémoire de Master. Université Pierre et Marie Curie, Paris VI. 32 p. Lord, C., Duhamel, G., Pruvost, P. (2006) The Patagonian toothfish (Dissostichus eleginoides) fishery in the Kerguelen Islands (Indian Ocean Sector of the Southern Ocean). CCAMLR Science 13: 1-25. Romanov, E.V. (Editor) (2003). Summary and Review of Soviet and Ukrainian Scientific and Commercial Fishing Operations on the Deepwater Ridges of the Southern Indian Ocean. FAO Fisheries Circular No. 991. FAO/YugNIRO, Rome, Italy. 84 p. Watson, R., Kitchingman, A., Gelchu, A., Pauly, D. (2004) Mapping global fisheries: sharpening our focus. Fish and Fisheries 5: 168-177. Zeller, D. and Rizzo, Y. (2007) Country disaggregation of catches of the former Soviet Union (USSR). p. 157-163 In: D. Zeller, and D. Pauly (eds.) Reconstruction of marine fisheries catches by countries and regions (1950-2005). Fisheries Centre Research Reports 15(2). Fisheries Centre, University of British Columbia. Zeller, D., Rossing, P., Harper, S., Persson, L., Booth, S. and Pauly, D. (2011) The Baltic Sea: estimates of total fisheries removals 1950-2007. Fisheries Research 108: 356-363.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Appendix Table A1: Kerguelen Islands, CCAMLR area 58.5.1, fisheries catch statistics (t) by French trawlers and longliners from 1970-2010 adjusted from fishing season to calendar year (see text and Figure 2). France Ridge Marbled Gray scaled Rattailsa Raysb rockcod rockcod rattail 1970-1978 – – – – – – – – – 1979 – 75 – 9 – – 65 247 – 1980 – 136 – 11 – – 506 597 – 1981 – 952 – 9 – – 984 574 – 1982 – 2328 – 14 – – 645 326 – 1983 – 1832 – 15 – – 143 312 – 1984 – 855 – 379 – – 136 622 – 1985 – 1407 – 396 – – 139 548 – 1986 – 1093 – 201 – – 71 228 – 1987 – 158 – 207 – – 28 93 – 1988 – 1292 – 118 – – 16 104 – 1989 – 1295 – 152 – – 24 163 – 1990 – 188 – 208 – – 170 67 – 1991 – 8 – 1199 – – 148 5 0 1992 – 7 – 1611 – – 0 0 0 1993 – 6 – 1582 – – 0 0 0 1994 – 6 – 2960 – – 0 0 1 1995 – 42 1 4178 – – 0 1 1 1996 – 45 1 3742 – – 0 1 0 1997 – 3 2 3744 – 0 0 0 6 1998 – 0 3 3919 – 6 0 0 11 1999 – 0 1 3984 – 7 1 0 87 2000 – 0 0 5139 – 87 1 0 189 2001 – – – 5443 – 132 0 0 298 2002 – – – 4450 – 201 – – 623 2003 – – – 4722 – 472 – – 815 2004 – – – 5231 – 805 – – 383 2005 7 – – 5123 – 489 – – 477 2006 21 – – 5115 – 476 – – 428 2007 15 – – 5179 – 537 – – 351 2008 35 – – 5026 409 276 – – 186 2009 58 – – 5045 896 – – – 313 2010 23 – – 2620 488 – – – 136 Totals 157 11,826 6 81,727 1,792 3,487 3,077 3,443 4,302 a Macrourus spp.;b Raja spp. and unidentified Rajiformes, most probably Bathyraja eatonii and Bathyraja irrasa (Lord, 2005; Lord et al., 2006). Year  Blue Mackerel antimora icefish  Unicorn icefish  Patagonian toothfish  19  20  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Appendix Table A2: Kerguelen Islands, CCAMLR area 58.5.1, fisheries catch statistics  (t) for the Ukrainian (i.e., former USSR and Russian Federation statistics) from 19702010 adjusted from fishing season to calendar year and estimates of illegal fishing during the period 1996-2003 (see text and Figure 2). Ukraine  Illegal Mackerel Patagonian Marbled Gray Other b Fisheries icefish toothfish rockcod rockcod finfisha 1970 3 – – – – – 1971 8982 – – – – – 1972 17995 – – – – – 1973 9018 – – – – – 1974 410 – – – – – 1975 900 – – – – – 1976 6275 – – – – – 1977 25110 – – – – – 1978 24847 – – – – – 1979 6118 71 522 1978 – – 1980 1485 90 4045 4772 – – 1981 8019 74 7875 4595 – – 1982 18622 113 5162 2607 – – 1983 14657 123 1141 2493 – – 1984 6841 3031 1088 4979 – – 1985 11253 3170 1113 4388 – – 1986 8747 1609 571 1824 – – 1987 1261 1655 225 748 – – 1988 10338 946 126 829 – – 1989 11585 1212 172 777 13 – 1990 4494 1354 143 1397 13 – 1991 6699 3902 150 674 – – 1992 6647 3772 144 51 5 – 1993 16 4096 1 0 12 – 1994 614 1530 1 0 7 – 1995 1926 1311 – 9 31 – 1996 1312 1279 – 8 31 1000 1997 – 1008 – – – 6913 1998 – 966 – – – 8275 1999 – 739 – – – 5163 2000 – 297 – – – 5410 2001 – – – – – 5760 2002 – – – – – 5213 2003 – – – – – 2063 2004 – – – – – – 2005 – – – – – – 2006 – – – – – – 2007 – – – – – – 2008 – – – – – – 2009 – – – – – – 2010 – – – – – – Totals 214,171 32,345 22,478 28,567 111 39,795 a Osteichthyes reported in the CCAMLR statistics, most probably including incidental bycatch of southern lantern shark (Etmopterus granulosus), porbeagle (Lamna nasus), Greenland shark (Somniosus microcephalus), moray cod (Muraenolepis marmoratus) and gray rockcod (Notothenia squamifrons) (see Lord et al., 2006); bConsisting only of longlines targeting Patagonian toothfish; their bycatch consisting mainly of rays and rattails. Note that the blue antimora (Antimora rostrata), unicorn icefish (Channichthys rhinoceratus), ridge scaled rattail (Macrourus carinatus) and other rattail species and rays are also bycatch of the longline fishery which are sometimes legally reported. Year  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  21  RECONSTRUCTION OF TOTAL MARINE FISHERIES CATCHES FOR MADAGASCAR (1950-2008)1 Frédéric Le Manacha, Charlotte Goughb, Frances Humberb, Sarah Harperc, and Dirk Zellerc aFaculty  of Science and Technology, University of Plymouth, Drake Circus, Plymouth PL4 8AA, United Kingdom; flm@livinblue.com bBlue Ventures Conservation, Aberdeen Centre, London, N5 2EA, UK; charlie@blueventures.org; fran@blueventures.org cSea Around Us Project, Fisheries Centre, University of British Columbia 2202 Main Mall, Vancouver, V6T 1Z4, Canada ; s.harper@fisheries.ubc.ca; d.zeller@fisheries.ubc.ca  ABSTRACT Fisheries statistics supplied by countries to the Food and Agriculture Organization (FAO) of the United Nations have been shown in almost all cases to under-report actual fisheries catches. This is due to national reporting systems failing to account for Illegal, Unreported and Unregulated (IUU) catches, including the non-commercial component of small-scale fisheries, which are often substantial in developing countries. Fisheries legislation, management plans and foreign fishing access agreements are often influenced by these incomplete data, resulting in poorly assessed catches and leading to serious over-estimations of resource availability. In this study, Madagascar‘s total catches by all fisheries sectors were estimated back to 1950 using a catch reconstruction approach. Our results show that while the Malagasy rely heavily on the ocean for their protein needs, much of this extraction of animal protein is missing in the official statistics. Over the 1950-2008 period, the reconstruction adds more than 200% to reported data, dropping from 590% in the 1950s to 40% in the 2000s. This discrepancy has profound management implications as well as consequences for current understanding of Madagascar‘s fisheries economy and communities‘ reliance on wild fish for food security.  INTRODUCTION Madagascar is located in the western Indian Ocean, and separated from Africa by the Mozambique Channel (Figure 1). With a land area of approximately 587,000 km², it is the fourth largest island in the World and an African biodiversity hotspot, with around 80% of its terrestrial species being indigenous, and its endemic biodiversity threatened by habitat loss (Brooks et al., 2006; Anon., 2008a). Given its great size, spanning 14 degrees of latitude, Madagascar exhibits a range of geological, oceanic and climatic environments, for example, the east of the country is mountainous with a narrow continental shelf facing the prevailing trade winds and oncoming east equatorial current, while the west side is characterised by large plains in a rain shadow, with the coast fringed by a wide continental shelf (Cooke et al., 2003). The southern region is subject to more arid conditions (Jury, 2003), restricting its agricultural potential. These environmental differences have also shaped marine ecosystems: mangroves are almost exclusively present on the west coast (Giri and Muhlhausen, 2008), whereas coral reefs span the southwest, west and northeast coasts, and include one of the largest coral reef systems in the Indian Ocean, totalling approximately 2,230 km² (Spalding et al., 2001). These geographical differences have also resulted in spatial divergence in the distribution of the island‘s human population with the eastern part of the island having the highest density, while the west coast is home to the majority of fishers and therefore experiences the highest fishing pressure (Guidicelli, 1984; Bellemans, 1989; Rafalimanana, 1991; Laroche et al., 1997).  1  Cite as: Le Manach, F., Gough, C., Humber, F., Harper, S. and Zeller, D. (2011) Reconstruction of total marine fisheries catches for Madagascar (1950-2008). pp. 21-37. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN1198-6727].  22  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Historically, Madagascar has had several political regimes (Schraeder, 1995). After the Berlin Convention in 1885, which decided the fate of most of the African continent during colonisation, Madagascar was invaded by France in 1896, turning Madagascar into a French colony and finally Overseas Territory in 1946. Although the colonial power invested in national infrastructure such as trains and schools, this period was characterised by protracted political violence, with around 700,000 out of 3 million inhabitants being killed within a few decades (Rousse, 2010). Giving increasing power to national institutions, the French government withdrew step-by-step and in 1960 the First Republic was proclaimed. However, the first Malagasy President was unpopular with the country‘s people, mainly due to the continuing strong economic and political ties with France. In 1975, the Second Republic aligned itself with the USSR; key sectors of the economy were nationalized and the country experienced a radical socialist and authoritarian political regime. Ten years later, heavy opposition to this regime developed, Figure 1: Madagascar and its Exclusive Economic Zone (solid line). and in 1992 the Third Republic was proclaimed. Political instability continues to the present day, following a military-backed coup in 2009. Madagascar‘s current unelected regime faces ongoing economic sanctions and is not recognised by the international community, including the European Union (EU) or the Southern African Development Community (SADC). Economically, Madagascar is one of the poorest countries in the world. Per capita GDP has declined steadily since Independence, having never exceeded $410, and currently is at less than $300 (year 2000 USD). Approximately 70% of the population currently lives below the poverty threshold, and over half of the country‘s population is dependent on the exploitation of natural resources for their livelihoods (World Bank, 2010; Horning, 2008). Subsistence fisheries are of prime importance for coastal communities, especially in the south and west of the country where agriculture is virtually impossible due to aridity. In developed countries, scientific fisheries assessments such as stock assessments can provide robust data on which to base fisheries management decisions. However, these approaches are expensive, technically complex and often challenged (Murawski, 2010). Developing countries such as Madagascar rarely have adequate scientific capacity or resources to undertake stock assessments. Consequently, poor or incomplete catch data often serve as the only basis for policy and decision making in such countries. In the absence of effective regulations, catch statistics are thought to approximate fluctuations in fish stocks and are therefore viewed as an acceptable proxy for stock assessments. However, catch statistics generally do not account for Illegal, Unreported and Unregulated (IUU) catches, which are widely recognised as a major barrier towards sustainable fisheries management (Sumaila et al., 2006; Hosch et al., 2011). Such IUU catches often result in serious under-estimates of extracted resources and over-estimate of their sustainability (Jacquet et al., 2010). FAO FishStat (FAO, 2009) provides time-series data on marine  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  23  fisheries landings starting in 1950. These data are based on statistics provided to FAO by member countries. However, it has been shown for many countries that the statistics submitted to the FAO are often incomplete, particularly with regards to artisanal and subsistence fisheries (Zeller et al., 2007; Jacquet et al., 2010; Wielgus et al., 2010). This is likely the case with Madagascar as well, where the importance of seafood for domestic food security has rarely been recognised by the various governments. Here, we re-estimated total marine catches by Madagascar within its EEZ (or EEZ-equivalent waters) for the period 1950 to 2008, with the aim of providing a more accurate baseline for use in policy decisions. This included a review of all fisheries sectors in the country, which allowed us to highlight missing or under-reported components. The present work is also published in Le Manach et al. (2012). Note that Le Manach et al. (2012) contains a mislabelled Figure (3b), which was corrected through a subsequent errata, and is correctly presented here as Figure (6).  MATERIALS AND METHODS Human population data Human population data were obtained from PopulStat (www.populstat.info) and various other sources (Central Intelligence Agency, 2010; Globalis, 2010) and used to derive the number of fishers for the whole time-series (1950-2008). Linear interpolations were made between years of known data (Figure 2a). A fishers‘ census conducted by the FAO documented the percentage of artisanal fishers among the population in 1988 for each district (Bellemans, 1989). In the absence of more recent estimates, we assumed that these ratios remained stable between 1988 and 2008. It is likely conservative, as the declining per capita GDP during this period would suggest a growing reliance on small-scale artisanal fishing for livelihood and food security. For the 1950-1988 period, the proportion of artisanal fishers among the total population was assumed to have doubled, increasing from approximately 2% to 4%. Indeed, Bellemans (1989) reported that the number of fishers approximately doubled during the two decades preceding the census (Figure 2b). Billé and Mermet (2002) have also Figure 2: a) Human population of Madagascar with anchor indicated a two-fold increase in the number points indicated by solid circles, and b) number of artisanal of fishers between the early 1980s and the fishers with the 1987-88 census data foundation (Bellemans, early 2000s. Based on this, we estimated a 1989) indicated by dots. fisher population of 100,000 individuals in 2005 (G. Hosch, pers. comm., Fisheries Planning and Management; Gough and Humber, unpub. data). For this study, the coast was divided in two areas: (1) the southwest, comprising the district of Toliara (Figure 1), where the fishing pressure is known to be the highest (Laroche et al., 1997), and (2) the remaining coastal districts, where fishing pressure is consequently thought to be somewhat lower.  Fisheries sub-sectors The officially reported landings data which served as the baseline for the study were extracted from the FAO FishStat database (FAO, 2009), and a thorough bibliographic review of fisheries activities in  24  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Madagascar allowed us to determine which sectors were either missing or being under-reported. Data sources included peer-reviewed publications, reports by non-governmental organizations (NGOs), and other grey literature. Expert and local knowledge was collected for each sector in order to formulate the best assumptions possible. For each of these fisheries sectors, catches were then derived for the entire 1950-2008 period, based on anchor points found in the literature and informed knowledge-based assumptions.  Shrimp fishery The shrimp fishery represents around 4% of Malagasy reported landings for the last decade. However, its market value is significant, reaching almost 70% of the officially recorded marine resource contribution to the GDP, mainly due to important export to Europe, Asia and North America (Soumy, 2006; Anon., 2008b; Kasprzyck, 2008). The industrial shrimp fishery officially started in 1967, after several years of exploratory trawling (Fourmanoir, 1952a, b; Crosnier, 1965). The number of vessels steadily increased from 11 in 1967 to a maximum of 71 vessels between 1993 and 2003, when the number of licenses started to be controlled. Since then, the number of vessels has decreased, with 47 licensed vessels operating in 2008. Although linked to the economic recession of the 2000s, the increasing price of fuel and the international market being flooded by cheaper Chinese shrimp products, this drop was also due to declines in landed catch and increasing competition with artisanal fishers (Kasprzyck, 2008). Conflict between artisanal and industrial fishers continues to be a serious concern for west coast fishing communities (Cripps, 2009). Despite the intensive exploitation by industrial operators in the past, the decline in the economic viability of the fishery is causing owners to reduce their fleets (Razafindrainibe, 2010; McNeish, 2011). The Société de Pêche de Morondava (SoPeMo), a shrimp fishing company based on the west coast, stopped commercial trawling in the region in 2008 (C. Gough, pers. comm.). Unima, the biggest shrimp company in Madagascar is accusing artisanal fishers to have caused the decline in catches (McNeish, 2011). Artisanal fishers have targeted shrimp since the 1970s. Before then, shrimp was considered incidental bycatch and was used as a complementary food, consumed as an overcooked paste. Since then, artisanal fishers have been attracted by this high-value resource and sell a large portion of their catches to local markets and processors. Due to its rapid expansion, we assumed a yearly growth rate for the artisanal shrimp fishery of 4.5% from 1970 to 2005 (Kasprzyck, 2008; C. Chaboud, pers. comm., Institut de Recherche pour le Développement). The 2005 landings of 3,500 tonnes were used as a basis to carry backward this 4.5% yearly growth rate to 1970. Bycatch from shrimp fisheries is largely associated with industrial trawl vessels, and is known to be particularly high for tropical shrimp fisheries. Typically, discarded bycatch is not accounted for in reported landings, and we assumed this also applied to Madagascar. One of the earliest estimates for bycatch in the Malagasy shrimp fishery was an amount between 18,000-20,000 tonnes per year in the 1980s (Roullot, 1989). Kelleher (2005) proposed a 1:4.1 bycatch ratio and a discard rate of 72%, which gives tonnage values for the late 1980s similar to Roullot (1989), suggesting that the bycatch was almost entirely discarded. However, it is worth noting that a part of this bycatch is traded to local fishers who fill up their pirogues in exchange for some tobacco or a small amount of money and the local fishers land it and sell it for consumption in the local markets (A. Harris, pers. comm., Blue Ventures Conservation). Here, we applied a discard rate of 72% from 1990 to 2000, and a 90% discard rate for the 1967-1990 period (Table 1), based on Roullot (1989). Lower bycatch ratios and discard rates have been suggested for the last few years (late 2000s). Randriarilala et al. (2008) reported a bycatch ratio of 1:2.5 for 2003-2005, which suggests an annual amount of between 8,500 tonnes and 12,700 tonnes per year. This decrease, even stronger since 2005, seems to have been influenced by two developments in Malagasy regulations (Razafindrainibe, 2010). The first development was the introduction of legislation in the 1990s (decree 1999/2000) requiring industrial vessels to retain at least 50% of bycatch to supply fish to local markets. However, the effectiveness of this regulation has been questioned by Randriarilala et al. (2008), who assessed a discard rate of 62% for the period 2003-2005. The second development was the introduction of Turtle Excluder Devices (TEDs) and Bycatch Reduction Devices (BRDs) in the Malagasy shrimp fishery in 2003 (Anon, 2003; Razafindrainibe, 2010). Since 2005, all vessels are mandated to be equipped with  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  25  such BRDs (decree 2003-1101), which considerably reduce the amount of bycatch (Kasprzyck, 2008; Table 1). This reduction is estimated to be approximately 32%, depending on the year, fishing conditions and area (Fennessy and Isaksen, 2007). Therefore, we considered a bycatch ratio of 1:1.7 from 2005 onward, while discard rates were set at 62% since 2001 (Table 1). However, only approximately 30% of licensed boats carry enforcement personnel from the national fisheries surveillance authority, and it is not clear whether crews use TEDs and BRDs when not under surveillance. Indeed, without surveillance, crews have little incentive to follow legislation, since the economic conditions under which crews work create a major incentive to maximise bycatch for private sale at sea. Shrimpers work seven days a week and are paid per tonne of shrimp landed, with base salaries as little as $25 per month. The bycatch is dominated by fin fish (e.g., Otolithes argenteus, Johnius dussumieri, Trichiurus lepturus and Pomadasys maculatus), and to a lesser extent, invertebrates such as sea urchins and jellyfish, which can represent an important component of total bycatch. Chen and Chow (2001) estimated a discard survival rate of 8% for fin fish, 35% for cephalopods and 60% for crustaceans in a tropical shrimp fishery in Asia. We assumed here that all the discarded bycatch had a similar mortality, and thus applied mortality rates of 8%, 35% and 60% to the respective bycatch amounts of fin fish, cephalopods and other crustaceans. The commercial shrimp fleet also comprises two other sectors of much lower capacity: -  -  A small fleet of mini-trawlers with engine power less than 50 horse power. This fleet was intended to introduce more efficient gears to the artisanal fisheries. However, they were taken up by industrial fisheries societies to allow them to fish in certain areas otherwise not accessible by their large boats (Kasprzyck, 2008; Direction des Pêches in Morondava, pers. comm.). Landings reported to FAO increased from 45 tonnes to 700 tonnes per year from 1975 to 2008 (peak at 750 t in 2003), and were considered reliable and were included without modification in the ‗industrial grouping‘ described above. A deep-sea shrimp fishery was initiated in 1992 and ended in 2005 due to technical issues, mainly due to the nature of the sea floor. No information was available concerning associated bycatch. Catches fluctuated between 100 and 150 t∙year-1 and were also included without modification in the ‗industrial grouping‘.  Table 1: Summary of data, assumptions and sources used to reconstruct total catches by shrimp fishing fleets in Madagascar. Time period  Associated bycatch and discards Shrimp catches (t·year-1) Bycatch Discards ratio (%) 300 - 13,300a -  Sources  Comment  Domalain et al. (2000); Goedefroit et al. (2002); Razafindrakoto (2008), Rokotodratsimba et al. (2008) ; FAO (2009) 1967-1989 1:4.1 90 Roullot (1989) 1990-2000 1:4.1 72 Kelleher (2005) 2001-2004 1:2.5b 62 Randriarilala et al. (2008) Decree 1999/2000 2005-2008 1:1.7b 62 BRD introduction a Values reported to FAO were kept forb the years 1966, 1994, 2000-2003, as they were deemed more representative than those reconstructed. Based on a 32% reduction of bycatch due to BRDs (Fennessy and Isaksen, 2007). 1967-2008  Shark fishery Although consumption of sharks is common, Madagascar only reports landings of less than 10 tonnes per year for the 2001-2008 period, according to FAO. However, the FAO trade database documents shark exports of up to 85 tonnes per year since 1992, and an independent report suggests that shark meat and fins have been exported since the early 20th century (Petit, 1930). In fact, artisanal fishers target sharks for the export market of fins, but carcasses are rarely discarded, and the meat is either consumed locally or to a lesser extent, sold to Comoros. A number of endangered benthic species, such as the critically  26  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  endangered sawfish (Pristis pectinata), were once commonly caught by artisanal fishers along the mangrove-fringed coast of western Madagascar, but are now considered extremely rare throughout the region (A. Harris, pers. comm., Blue Ventures Conservation). Sharks are also caught as bycatch in other Malagasy commercial fisheries, such as the shrimp fishery, in which sharks have been reported as representing 1% of the bycatch (C. Chaboud, pers. comm., Institut de Recherche pour le Développement; Table 2). Due to the high price of shark fins, we assumed that all sharks are finned, and that all of the carcasses are retained for local consumption or exported to Comoros (C. Chaboud, pers. comm., Institut de Recherche pour le Développement; A. Harris, pers. comm., Blue Ventures Conservation). Finally, shark liver oil has also been traditionally used for cooking and to waterproof wooden boats (Cooke, 1997). The quantity of oil used for this purpose is substantial in the Maldives, as Anon. (2001) reports a use of between 54 and 58 kg of oil per year, for each boat. However, as it was impossible to assess the number of sharks required to treat the whole Malagasy artisanal fleet, this component was not considered in this study. Shark liver oil is also a valuable commodity on international markets, with Madagascar‘s sharks targeted by illegal, unreported and unregulated (IUU) boats for this purpose as well as for the fins. A number of known IUU vessels, which previously targeted Patagonian toothfish in the southern Ocean (see Palomares and Pauly, this volume), are reported to have recently converted to shark fishing in southern and western Madagascar by substituting bottom trawl nets with bottom-set gillnet gear to target nurse sharks for liver oil (mainly Nebrius ferrugineus and the vulnerable Pseudoginglymostoma brevicaudatum) (SADC, 2008; Anon. 2010a and b). Table 2: Summary of data, parameters, assumptions and sources used for the reconstruction of shark fisheries catches in Madagascar. Sector Targeted  Timeperiod 1950-1979  dry fina (t) -  1980-1985  -  1986-1995 1996-2008 1967-2008  34.5 - 64.7 -  Sources Petit (1930)  Catch (t) 160 - 570b 600 - 3050  Cooke et al. (2001)  3,430 - 6,440 5,400 - 3,760c Up to 385 t (1998)d  Comment Exports at least since 1930s. Backward extension of 1980 per fisher catch rate Backward extension of 19861988 trend in derived catch Decrease of 3%∙year-1  Shrimp bycatch a Hong Kong and Singapore imports; b 1950 value of 60 t derived through keeping the 1980 catch per fisher fixed. c Values for this period were based on interpolations from the 1994 fin trade data, and an assumed 3% per year decrease in catches (McVean, 2006; Cooke et al., 2001; Y. Sadovy, pers. comm. University of Hong Kong). d Values were based on reconstructed industrial shrimp catches and an assumed 1% of total shrimp bycatch composed of sharks (C. Chaboud, pers. comm., Institut de Recherche pour le Développement).  Cooke (1997) and Cooke et al. (unpub. data) review the shark fishery in Madagascar, focusing on exports of fins to the Hong Kong and Singapore markets. Given the high market price of fins, we assumed that all captured sharks were finned, and therefore, trade in fins was the best proxy available to assess the minimum quantity of sharks caught each year. Three approaches were used to reconstruct total shark catches by Malagasy fishers (Table 2). Data on the trade of shark fins were used to conservatively estimate the likely minimum catches of sharks that occurred in Madagascar‘s waters during the period 1970-1994: dried fins imported between 1986 and 1995 by Honk-Kong and Singapore from Madagascar were converted to whole body, wet weight using a conversion factor of 98.5% (Cortes and Neer, 2006; Jacquet et al., 2008; Y. Sadovy, pers comm., University of Hong Kong). We assumed that the market started to greatly expand in 1980, and therefore linearly extended the 1986-1988 trend backwards to 1980. For the 1950-1979 period, we assumed that the 1980 per fisher catch rate remained constant back to 1950, and expanded it to total catches using fisher population data. For the 1996-2008 period, we conservatively assumed that the 1994 per fisher catch rate decreased by 3%∙year-1, based on literature and local knowledge (McVean et al., 2006; Y. Sadovy, pers. comm., University of Hong Kong). Currently, catches are reported to decrease, and fishers catch fewer and smaller sharks, most of the time farther from shore than before (Cooke, 1997 and 2003; McVean, 2006; Cooke et al., unpub. data). The high market demand for shark fin as a lucrative yet diminishing fisheries resource is a key factor driving Madagascar‘s nomadic  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  27  Vezo fishers further afield during their annual migration, with shark fishers exploiting remoter regions of the west coast of Madagascar, further offshore and in larger numbers, than ever before (Cripps, 2010). As an example of this escalation in fishing effort, the recent introduction of new intensive fishing techniques in the offshore Barren Isles archipelago and around Morondava, involves teams of artisanal fishers deploying weighted ‗barrage‘ nets several kilometres in length, targeting sharks and guitarfish (Cripps, 2010). Based on this information, we applied a 3% per year decrease in catches since 1994 (Table 2).  Non-shrimp invertebrate fisheries The remaining landings data reported to FAO have been aggregated into a miscellaneous invertebrate grouping, which includes cephalopods and other molluscs, crabs and lobsters, shells and sea cucumbers. These species are heavily targeted by men, women and children for both subsistence and commercial purposes (Rasolofonirina and Conand, 1998; Frontier Madagascar, 2003; Anderson et al., 2008; Barnes and Rawlinson, 2009; Cripps, 2009; Gough et al., 2009; Tucker et al., 2010). The under-reporting of invertebrate fisheries is visible in the statistics reported to the FAO, since reported landings are very similar to exported fisheries products (r²=0.75; not shown). Also, there are no reported invertebrate catches before 1962. However, coastal populations rely heavily on reef gleaning for invertebrates for their daily protein needs, although a significant amount is sold (Cripps, 2009; Gough et al., 2009). Indeed, invertebrate landings account for a major component of fisheries-derived income for artisanal fishers in many parts of Madagascar. Beside holothurian fisheries, octopus is the dominant commercial fishery in much of the southwest and northeast of the country, and lobster plays a crucial role in coastal livelihoods from the rocky shores of the southeast. A thriving trade in marine curios, predominantly molluscs, is also present in most coastal towns. In all these cases, catches are sold to collectors by local fishers for international export. Therefore, we assumed that invertebrate extraction by the local population was happening prior to 1962 and that this sector is missing from the official data. In order to re-estimate the total extraction of invertebrates, product weight as it appears in the trade data was converted to (whole body) wet weight, using FAO conversion factors (Anon., 2000). A highly conservative export rate of 80% (for sea cucumber, cephalopods, crabs and lobsters) or 20% (for the other products) was then applied for the entire time period for which exports were thought to have occurred (1970s-2008) in order to calculate the domestic subsistence component. Finally, the average subsistence catch rates for the first three years of exports were applied to the number of inhabitants prior to the first year of export, in order to estimate the domestic subsistence component of invertebrate catches back to 1950. Table 3: Summary of parameters used for the estimation of small-scale catches. Coastal Areaa South-West  The remainder  Time period 1950 1988 1991 1950-1988 2008 1988-2008 1950 1988 1950-1988 2002 2008 1988-2008  Number of fishers a2,900 16,000 18,100 40,100 a5,300 28,500 55,300 71,600 -  CPUE 6.7 5.8 5.0* 2.1 6.1 5.7 4.3* 3.8 -  Fishing daysb 260 260 260 260  Catches (t) a5,155 24,110 23,574 21,853 a8,300 42,347 62,178 71,344 -  Comment  Source  Laroche (1997) Stable CPUE -5% CPUE∙year-1 Stable CPUE Doukakis (2007) -2% CPUE∙year-1  *Anchor points; a See Figure 2 for area definitions; bAssumed average number of fishing days in Madagascar (McVean, 2006; Gough and Humber, unpub. data; G. Hosch, pers. comm. Fisheries Planning and Management).  Small-scale fisheries: subsistence and artisanal catches Between 1950 and 2008, landings of tuna-like species, narrow-barred Spanish mackerel (Scomberomorus commersoni) and miscellaneous marine fish represented the bulk of the data reported to FAO, accounting  28  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  on average for 82% of total seafood landings in Madagascar. They are exclusively caught by artisanal fishers, as there are no industrial fisheries targeting these species.Unfortunately, estuarine catches were reported by Madagascar as inland catches until 1989, after which they were accounted for in the marine landings (Stamatopoulos and Rafalimanana, 1991). The same authors also report these estuarine catches to be 30,000 tonnes per year in the late 1980s. Finally, the official records of fish caught in the 1950s would signify a highly unrealistic local consumption of approximately 0.7 kg∙person-1∙year-1 (based on a coastal population representing 90% of the total human population). Considering the change in reporting protocol and this unrealistic consumption rate, we replaced the data supplied to FAO (tuna-like species, mackerel and miscellaneous marine fish) and re-estimated small-scale catches based on CPUE and fishing effort data from independent studies (Laroche and Ramananarivo, 1995; Laroche et al., 1997; Doukakis et al., 2007), in combination with the number of fishers for the two regions as defined in Figure 2. Based on local knowledge and reports, which suggest decreases in CPUE over time (Bellemans, 1989; Laroche et al., 1997; Frontier Madagascar, 2003; Langley, 2006; Doukakis et al., 2007; Gough et al., 2009), we applied different CPUE estimates for areas and time periods as shown in Table 3.  Other fisheries A commercial joint venture fisheries operation between Japan and Madagascar was established in the 1970s under the name of Compagnie Malgache Nippone de Pêcherie (COMANIP) for the exploitation of Madagascar‘s skipjack tunas, Katsuwonus pelamis (Marsac and Stequert, 1984; Gilbert and Rabenomanana, 1996). Independent catch data were not available; therefore we considered the data reported to FAO as reliable and included them in the final result without modification. Thus, we assumed no underreporting of skipjack tuna catches during this period. This is likely conservative, given the known occurrence of substantial and widespread underreporting of tuna catches. An exploratory deep-sea fishery in the Malagasy EEZ started in 2001. There is only one value provided by the FAO of 4157 tonnes for the year 2002. This sector is described by Soumi (2004) to have increased without any further indication; therefore, we assumed a growth rate of 5% per year between 2002 and 2008. In the last two decades, fishing tourism has rapidly expanded in Madagascar (Jain, 1995). Most of the catch, dominated by large pelagic species such as marlin and tuna, is not catch-and-release, and is therefore killed and retained. However, none of the 60 people contacted in this study (employees or managers of sport-fishing charter companies) were willing to share information with us. As a result we were unable to quantify extractions made by this sector. In terms of overall tonnage, this sector is likely to be small, but may have effects on the population structure of these species, especially the billfishes.  Foreign fishing in Madagascar’s waters Since 1986, a fishing agreement has been in place with the European Commission, allowing EU purseseine vessels to catch tuna in Malagasy waters (Gilbert and Rabenomanana, 1996). Catches of 10,000 tonnes have been declared each year since 1986. However, given that licence fees are based on this tonnage, it is highly likely that catches are largely under-reported, and may actually be around 18,00020,000 tonnes per year (Anon., 2002). It is also interesting to note that a substantial Asian long-line fleet has been fishing in Malagasy waters most of the time illegally without access agreements, with entirely un-reported catches of tuna. Anon. (1995) reported legal catches of 6,000-8,000 tonnes. However, some estimates are up to 50,000 t∙year-1 (Fowler, 2005). Indeed, Malagasy authorities do not possess the resources to patrol their own EEZ and therefore cannot address the problem of illegal fishing for such high-value species (Jain, 1995; Cooke, 1997; A. Harris, pers. comm., Blue Ventures Conservation; G. Hosch, pers. comm., Fisheries Planning and Management). Given that no formal access agreements exist between Madagascar and these countries, these catches are illegal under international law. For this longline fleet, 7.5% of the bycatch is composed of sharks (Cooke, 1997; Fowler, 2005), of which only fins are retained. Finally, a longline fleet, whose catches are uncertain (Fowler, 2005), is operating from La Réunion (René et al., 1998) and is targeting tuna and other large pelagic species (e.g., swordfish, marlin). For our purposes, we assumed this longline fleet started fishing in Malagasy waters around 1990 and catches increased linearly to 5,000 t∙year-1 in 2008. Cases of this sort of unreported fishing are sometimes  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  29  covered by media journals (Anon., 2010b) but have remained largely unaddressed by Malagasy authorities or Regional Fisheries Management Organizations (Cullis-Suzuki and Pauly, 2010). These catches, important in an ecosystem sense, are estimated here and listed in Appendix Table A1, but are not included in our reconstruction of Malagasy fisheries catches, since they are made by foreign countries.  RESULTS Reconstructed catches by sector, taxa and year are presented in Appendixes Tables (A1, A2).  Shrimp fishery Total shrimp catches, which were very low prior to 1967, increased until the late 1990searly 2000s, with peak catches of 12,850 t∙year-1 in 2003 (Figure 3a). Industrial landings stabilised in the 1990s at around 8,500 t∙year-1, with a peak of 9,850 t∙year-1 in 1998. They have declined substantially since 2002, with only 5,200 t caught in 2008. Similarly, small-scale catches increased from approximately 300 t∙year-1 during the 1950s, to 750 t∙year-1 in 1970 and to 3,500 t∙year-1 in 2005. Thereafter, they underwent a gradual decline to around 2,700 t∙year-1 in 2008 (Figure 3a). Bycatch also followed a similar trend, and reached a maximum of 38,800 t∙year-1 in 1998, of which 25,500 t were discarded (72%). Since then, bycatch has decreased to around 8,000 t∙year-1 in 2008, of which 4,400 tonnes were discarded (62%; Figure 3a). Over the 1950-2008 time-period, the artisanal component represents 27% and the industrial sector 73% of total shrimp catches.  Figure 3: a) Shrimp catches by the industrial and artisanal sectors, and the associated bycatch (landed and discarded); b) total shark catches by small-scale fishers and c) catches for non-shrimp invertebrates, separated by exports or commercial (light grey), and subsistence (dark grey) catches (dotted line represents the data supplied to FAO).  Shark fishery The reconstructed data for sharks conservatively suggests low catches until 1980, averaging 350 t∙year-1, followed by a rapid and substantial increase, from approximately 500 tonnes in 1980 to a peak of almost 7,000 t∙year-1 in 1992 (Figure 3b). Since then shark catches have decreased to 3,800 t ∙year-1 in 2008 (Figure 3b).  30  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Non-shrimp invertebrate fishery Total invertebrate catches increased from 1,500 t∙year-1 in 1950 to 4,000 t∙year-1 in 1975 and were deemed to be exclusively for subsistence purposes during that time (Figure 3c). Since 1975, total catches have comprised both a subsistence and a commercial (export) component and, although fluctuating over time, have steadily increased to approximately 16,000 t∙year-1 by 2008 (Figure 3c). Over the 1950-2008 time-period, invertebrate catches have totalled 193,800 tonnes, of which the subsistence component represents 45% and commercial exports 55%.  Small-scale finfish fisheries: subsistence and artisanal catches  Figure 4: Total reconstructed catches of the small-scale finfish fishery, showing the taxonomic breakdown (based on Laroche et al., 1997). Approximately half of these catches are for subsistence, and the other half for sale on the local market (C. Gough and F. Humber, pers. obs.).  Reconstructed subsistence and artisanal catches of finfish by small-scale fishers have increased steadily over the 1950-2008 time-period (Figure 4). Total subsistence and artisanal catches in 1950 were around 13,500 t∙year-1 and have increased to around 93,000 t∙year-1 by 2008 (Figure 4).  Figure 5: a) Total reconstructed catches versus reported landings as supplied to FAO by Madagascar; and b) Taxonomic composition of the overall reconstructed catches. Fenneropenaeus indicus is a shrimp species; all the other groups (except the mixed-group ‗others‘) are from the small-scale, finfish fishery.  Overall reconstruction Over the whole 1950-2008 time-period, total catches taken by Malagasy fishers in Madagascar‘s EEZ are estimated at 4.7 million tonnes. Thisreconstructed total is twice as high as the data supplied to the FAO by the government of Madagascar (Figure 5). Significantly, the re-estimation added over 550% for the earlier time-period (1950s), but adjusted the reported data by only 39% for the 2000-2008 period (Figure 5a). The taxonomic composition of reconstructed total catches shows a constant attern over time (Figure 5b).  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  31  Foreign fishing in Madagascar’s waters Tuna catches taken by foreign vessels are thought to have increased substantially since the mid-1980s. Current catches are somewhat uncertain due to obvious unreported and illegal catches. A minimum estimate is over 70,000 t∙year-1 (Figure 6). Catches are likely dominated by Asian longline fleets operating illegally, with catches having increased to 50,000 t∙year-1 by 2008. The EU purseseine fleet, although operating legally through access agreement, is known to substantially under-report by at least 100%, with official catches reported as around 10,000 t∙year-1, while estimated actual catches are around 18,000 t∙year-1 (Figure 6). Note that the publication by Le Manach et al. (2012) contains a mislabelled version of Figure (6). The present version is labelled correctly.  Figure 6: Estimated time-series of tuna catches in Madagascar‘s EEZ.  DISCUSSION Overall, reported data show a steady increase in landings, due to the expansion of industrial fleets in the early 1980s. According to our analysis, Malagasy fisheries had been underreported by over 500% in the early time period, and seem to be underreported at present by at least 40%. The reporting is therefore improving, but current values, likely under-estimates, are still very substantial. Our reconstructed catch time-series shows a levelling off of catches over the last two decades (Figure 5). It is worth noting that this levelling off of total catches is partly related to the improvement of bycatch handling by the shrimp industry causing decreased bycatch/discards since the 1990s. However, it is also certainly due to decreases in catches seen in various invertebrates (including shrimp) and shark fisheries, which suggest that overfishing is likely to be taking place. The official data also fail to account for a large part of the subsistence fishery, which represents 75% of the total reconstructed catch over the whole period, and 83% for the period 1950-1980. Such marginalization of small-scale fisheries is common (Pauly, 1997), although inclusion in official statistics is crucial. As a consequence of this marginalization, total Malagasy catches may be approaching or even exceeding sustainable yields for coastal stocks, estimated at 180,000 t∙year-1 (Anon., 2008b), with it remaining undetected. Another issue that has been dealt with in this report is the poor taxonomic information included in official statistics supplied to FAO. The major group in these official statistics represents over 80% of total catches, and is only described as ‗marine fish nowhere else included‘. Although we made a taxonomic breakdown of total catches, information related to species composition remains extremely poor. This fact justifies the importance of the implementation by FAO of taxonomic census every three to five years in order to create more reliable species composition times-series. Although not included in the total reconstruction, we also reviewed foreign fisheries in Madagascar‘s EEZ. Current catches of tuna, billfishes and sharks are likely to be over 70,000 t∙year-1, most of which are made illegally. This situation raises serious legal questions, and also points to the issue of inappropriate low fishing access fees paid by developed countries (Kaczynski and Fluharty, 2002; Hanish and Tsamenyi,  32  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  2009) and poor to non-existing monitoring and enforcement of such agreements (e.g., Jain, 1995). Indeed, the monitoring and enforcement system for the entirety of Madagascar is only composed of 3 monitoring vessels, 8 speedboats, 18 inspectors and 22 observers (R. Fanazava, pers. comm., Centre de Surveillance des Pêches). This lack of monitoring and enforcement capability has led to increasing illegal pirate fishing in the waters of Madagascar, as evidenced here, which likely contributes significantly to unsustainable fishing practices in the Western Indian Ocean. For the large-scale commercial shrimp industry, no real discrepancies exist between landings reported to FAO and our re-estimated landings. However, the overall CPUE is decreasing, possibly because catches have been significantly higher (by up to 5,000 t∙year-1) than the estimated maximum sustainable yield of 8,700 tonnes∙year-1 (Kasprzyck, 2008). It is worth noting that no values were reported to the FAO before 1964, when the first exploratory trawls were conducted. Prior to this date, local people were nevertheless fishing and consuming shrimp, and our study has filled this gap by assigning a subsistence component to this sector, although this was negligible compared to total catches. Also, the significant bycatch produced by shrimp trawlers is missing from official data, and this issue needed to be addressed since this bycatch is often made unavailable to the local population, when not collected by artisanal fishers. Our study highlights the importance of such bycatch for Madagascar since the beginning of this fishery. The reconstructed time-series of shark catches gives a very different picture to the official data. The former considers that an artisanal fishery has existed since at least 1950, while the latter show little indication that a shark fishery exists despite this being fairly well documented in the independent literature (see Petit, 1930; Cooke, 1997). These values are based on the fin trade using strong assumptions and are considered highly conservative given that Hong Kong and Singapore do not account for 100% of the fin market. However, they are likely to be closer to actual catches than the previous assumption in which a lack of data has been incorrectly interpreted as no catch. Concerning the small-scale finfish fishery, our reconstruction provides very similar estimates to those provided to the FAO for the 1989-2008 period. For 1950-1989, our results however differ greatly, as they fill the gap made by estuarine catches accounted for in reported inland catches before 1989. This misreporting of estuarine catches is documented but has never been incorporated into official statistics. Importantly, there are increasing concerns about the rate of growth in small-scale catches slowing, eventually leading to declining catches by the small-scale, artisanal and subsistence fisheries. This finfish fishery supplies the bulk of local seafood consumption demand, as most of the catches are sold and consumed locally, and declines in catches could have significant impacts on the food security of coastal communities. Throughout this report, historical events and changes in reporting protocols illustrate the importance of linking historical information and fisheries data to current management plans, especially in maritime developing countries such as Madagascar, where fisheries are of fundamental importance for the food security of the people. The consequences of diminishing fisheries are likely to be particularly severe in an island nation in which over 50% of children under five years of age suffer delayed development due to a chronically inadequate diet, and where chronic food insecurity affects over 65% of the population (BackMichaud et al., 2009; Anon., 2010c).  ACKNOWLEDGMENTS We thank of the Sea Around Us Project, a scientific collaboration between the University of British Columbia and the Pew Environment Group. We also thank Alasdair Harris, Gilles Hosch, Phaedra Doukakis, Rojo Rambonimanana, Christian Chaboud and the Wildlife Conservation Society for information and knowledge sharing.  REFERENCES Anderson, S.C., Mills Flemming, J., Watson, R. and Lotze, H.K. (2010) Serial exploitation of global sea cucumber fisheries. Fish and Fisheries, 23 p.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  33  Anon. (1995) IOTC proceeding of the 6th Expert Consultation on Indian Ocean Tunas, Colombo (Sri Lanka), 25-29 September 1995. [in French]. 65p. Anon. (2000) Conversion factors - landed weight to live weight. FAO Fisheries Circular 847, Revision 1. FIDI/C847. United Nations Food and Agriculture Organization, Rome. 176 p. Anon. (2001) Report on the Maldivian fishing industry - 1960. In: Waheed, Z. (ed.) Maldives Marine Research Bulletin, 5. 101p. Anon. (2002) Madagascar environmental threats and opportunities assessment. United States Agency for International Development (USAID. 118p. Anon. (2003) Décret N°2003/1101 modifiant certaines dispositions du décret n°71-238 du 12 mai 1971, réglementant l'exercice de la pêche par chalutage. In: la mer territoriale malgache. [in French]. Anon. (2008a) Madagascar environmental threats and opportunities assessment - an update. United States Agency for International Development (USAID), 61 pp. Anon. (2008b) Madagascar - National Fishery Sector Overview. United Nations Food and Agriculture Organization, Rome. Anon. (2010a) Pêche illégale: Un bateau chinois arraisonné à Toliara. La Gazette, September 8th, 2010. Available at : www.lagazette-dgi.com/index.php?option=com_content&view=article&id=5927:peche-illegale-un-bateauchinois-arraisonne-a-toliara&catid=45:newsflash&Itemid=58. [Date accessed: February 25th 2011]. [in French]. Anon. (2010b) Une pêche exceptionnelle pour l‘Atsantsa. Le journal de l‘ile, 18 septembre 2010. Anon. (2010c) Enquête démographique et de santé de Madagascar 2008-2009. INSTAT et ICF Macro, Antananarivo. [in French]. Barnes, D.K. and Rawlinson, K.A. (2009) Traditional coastal invertebrate fisheries in south-western Madagascar. Journal of the Marine Biological Association of the United Kingdom, 89(08): 1589-1596. Bellemans, M.S. (1989) Estimations premières de la production des pêcheries traditionnelles côtières malgaches. UNDP/FAO Project MAG/85/04. United Nations Food and Agriculture Organization. 95p. [in French]. Billé, R. and Mermet, L. (2002) Sectoralization of an integrated coastal management programme: A case study in Madagascar. Journal of environmental planning and management, 45(6): 913-926. Black-Michaud, A., de Meulder, F., Randrianasolo, H. and Conan, C. (2009). Rapport d‘évaluation de l‘intervention prolongée de secours et de redressement à Madagascar (Ipsr 10442.0) – Aide pour faire face aux catastrophes naturelles et à l‘insécurité alimentaire saisonnière (Juillet 2006 – Juin 2009). Rome: Programme Alimentaire Mondial. [in French]. Brooks, T.M., Mittermeier, R.A., da Fonseca, G.A.B., Gerlach, J., Hoffmann, M., Lamoreux, J.F., Mittermeier, C.G., Pilgrim, J.D. and Rodrigues, A.S.L. (2006) Global biodiversity conservation priorities. Science, 313: 58–61. Central Intelligence Agency (2010) The world factbook – Madagascar. Available at: www.cia.gov/library/publications/the-world-factbook/geos/ma.html. [accessed: October 12th 2010]. Chen, C. and Chow, Y. (2001) Examining on the bycatch and discard catch of the shrimp beam trawl fishery in the coastal waters off Chiayi, southwest Taiwan. Journal of the Fisheries Society of Taiwan, 28(3): 229-248. Cooke, A., Jonahson, M., Doukakis, P., Smale, M. and du Feu, T. (2001) Sharks, shark fisheries and shark fin trade in Madagascar – review and analysis, with suggestions for action. Unpublished. 9p. Cooke, A., Lutjeharms, J.R.E. and Vasseur, P. (2003) Marine and coastal ecosystems. p. 179-208. In: Goodman, S.M. and Benstead, J.P., (eds.) The natural history of Madagascar, University of Chicago Press, Chicago. Cortes, E. and Neer, J.A. (2006) Preliminary reassessment of the validity of the 5% fin to carcass weight ratio for sharks. Collective Volume of Scientific Papers ICCAT, 59(3): 1025-1036. Cripps, G. (2009) Understanding migration amongst small-scale fishers in Madagascar. Blue Ventures Conservation Report for ReCoMaP. 165 p. Cripps, G. (2010) Feasibility study on the protection and management of the Barren Isles ecosystem, Madagascar. Blue Ventures Conservation Report (2009), for WWF and the "Reseau interdisciplinaire pour une gestion durable de la biodiversite marine: diagnostic environnemental et social autour des tortues marines dans le sud-ouest de l‘Océan Indien ". 272 p. Crosnier, A. (1965) Les crevettes Pénaeidés du plateau continental Malgache, état de nos connaissances sur leur biologie et leur pêche en septembre 1964. Cahiers ORSTOM, série Océanographie (Suppl.), 3(3): 1-158. [in French]. Cullis-Suzuki, S. and Pauly, D. (2010) Failing the high seas: A global evaluation of regional fisheries management organizations. Marine Policy 34: 1036-1042. Doukakis, P., Jonahson, M., Ramahery, V., Randriamanantsoa, B.J.D. and Harding, S. (2007) Traditional fisheries of Antongil Bay, Madagascar. Western Indian Ocean Journal of Marine Science, 6(2): 175-181. Fennessy, S.T. and Isaksen, B. (2007) Can bycatch reduction devices be implemented successfully on prawn trawlers in the Western Indian Ocean? African Journal of Marine Science, 29(3): p.453-463. Food and Agriculture Organization (2009) Fishstat Plus: Universal software for fishery statistics time series, version 2.3. Fourmanoir, P. (1952a) Compte rendu des nouvelles expériences de pêche du « Gabriel II » effectuées du 24 juillet au 15 novembre 1951. Bulletin de Madagascar, 52: 17-21. [in French].  34  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Fourmanoir, P. (1952b) Les crevettes d'intérêt économique à Madagascar. Le Naturaliste Malgache, 4(2): 163-168. [in French]. Fowler, S.L. (2005) Sharks, rays and chimaeras: the status of the Chondrichthyan fishes: status survey. World Conservation Union. 476p. Frontier-Madagascar (2003) Fin-fish resource use: artisanal fisheries of Beheloka. Woods-Ballard, A.J., Chiaroni, L.D., and Fanning, E., (eds.) Frontier-Madagascar Environmental Research Report 11. Society for Environmental Exploration, UK and the Institute of Marine Sciences, University of Toliara, Madagascar. 24 p. Gilbert, F. and Rabenomanana, L. (1996) Aperçu sur la pêche thonière à Madagascar. Appendix of the IOTC proceeding of the 6th Expert Consultation on Indian Ocean Tunas, Colombo (Sri Lanka), 25-29 September 1995. [in French]. 3p. Giri and Muhlhausen (2008) Mangrove forest distributions and dynamics in Madagascar (1975–2005). Sensors, 8: 2104-2117. Globalis (2010) Madagascar. Available at : http://globalis.gvu.unu.edu/country.cfm?country=MG. [Date accessed: October 12th 2010]. Gough, C., Thomas, T., Humber, F., Harris, A., Cripps, G., and Peabody, S. (2009) Vezo Fishing: An Introduction to the Methods Used by Fishers in Andavadoaka Southwest Madagascar. Blue Ventures Conservation Report. 37 p. Guidicelli M. (1984). Les Pêcheries maritimes malgaches: leurs principaux potentiels et leurs besoins pour le développement. SWIOP/Document OISO, RAF/79/065/WP/17/84, Seychelles. 112 p. [in French] Hanich, Q. and Tsamenyi, M. (2009) Managing fisheries and corruption in the Pacific Islands region. Marine Policy, 33(2): 386-392. Horning, N.R. (2008) Strong support for weak performance: donor competition in Madagascar. African Affairs, 107: 405-431. Hosch, G., Ferraro, G. and Failler, P. (2011) The 1995 FAO code of conduct for responsible fisheries: adopting, implementing or scoring results? Marine Policy, 35: 189-200. Jacquet, J., Alava, J.J., Ganapathiraju, P., Henderson, S. and Zeller, D. (2008) In hot soup: sharks captured in Ecuador‘s waters. Environmental Sciences; 5(4): 269-283. Jacquet, J., Fox, H., Motta, H., Ngusaru, A. and Zeller, D. (2010) Few data but many fish: marine small scale fisheries catches for Mozambique and Tanzania. African Journal of Marine Science, 32(2): 197-206. Jain, M. (1995) Inventory of marine and coastal activities - a strategy proposal for Madagascar. knowledge and effective policies for environmental management (KEPEM). Report no. 20. 103 p. Jury, M.R. (2003) Climate: the climate of Madagascar. p. 75-87. In: Goodman, S.M. and Benstead, J.P., (eds.) The natural history of Madagascar. University of Chicago Press, Chicago. Kaczynski, V.M. and Fluharty, D.L. (2002) European policies in West Africa: who benefits from fisheries agreements? Marine Policy, 26: 75-93. Kasprzyck, Z. (2008) Shrimp fishing in Madagascar. p. 223-234. In: Gillett, R., (ed.) Global study of shrimp fisheries. FAO Fisheries Technical Paper 475. United Nations Food and Agriculture Organization, Rome. Kelleher, K. (2005) Discards in the world's marine fisheries: an update. FAO Fisheries Technical Report 470. United Nations Food and Agriculture Organization, Rome. 154 p. Langley, J. (2006). Vezo knowledge: traditional ecological knowledge in Andavadoaka, southwest Madagascar. 73 p. Laroche, J. and Ramananarivo, N. (1995) A preliminary survey of the artisanal fishery on coral reefs of the Tulear Region (southwest Madagascar). Coral Reefs, 14(4): 193-200. Laroche, J., Razanoelisoa, J., Fauroux, E. and Rabenevanana, M.W. (1997) The reef fisheries surrounding the southwest coastal cities of Madagascar. Fisheries Management and Ecology, 4: 285-299. Le Manach, F., Gough, C., Harris, A., Humber, F., Harper, S. and Zeller, D. (2012) Unreported fishing, hungry people and political turmoil: the recipe for a food security crisis in Madagascar? Marine Policy 36: 218-225. Marsac, F. and Stequert, B. (1984) Bilan des prospections sur les ressources thonières de surface effectuées depuis 1971 dans l'Océan Indien Occidental. La pêche maritime, 1271: 83-94 McNeish, H. (2011) Madagascar‘s UNIMA to close shrimp factory in March. Bloomberg. Available at: http://www.businessweek.com/news/2011-01-26/madagascar-s-unima-to-close-shrimp-factory-inmarch.html. [Date accessed: February 25th 2011]. McVean, A.R., Walker, R.C. and Fanning, E. (2006) The traditional shark fisheries of southwest Madagascar: a study in the Toliara region. Fisheries Research, 82(1-3): 280-289. Murawski, S.A. (2010) Rebuilding depleted fish stocks: the good, the bad, and, mostly, the ugly. ICES Journal of Marine Science, 67(9): 1830-1840. Pauly, D. (1997) Small-scale fisheries in the tropics: marginality, marginalization, and some implications for fisheries management. In: Pikitich, E.K., Huppert, D.D. and Sissenwine, M., (eds.) Global Trends: Fisheries Management. American Fisheries Society, 20: 40-49. Petit, C. (1930) L‘industrie des pêches à Madagascar. Société d'Editions Géographiques, Maritimes et Coloniales, Paris. 392 p. [in French].  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  35  Rafalimanana, T. (1991) Estimation des captures de la pêche traditionnelle malgache en 1990. Ministère de la production animale, eaux et forêts. Direction de la pêche et de l‘aquaculture. UNDP Project MAG/85/014. Field report, 27. 84 p. [in French]. Randriarilala, F., Rafalimanana, T. and Caverivière, A. (2008) Les captures accessoires des crevettiers industriels et artisanaux. p. 237-258. In : Caverivière, A. Chaboud, C. and Rafalimanana, T., (eds.) Les crevettes côtières de Madagascar : biologie, exploitation, gestion. IRD Editions, Marseille. [in French]. Rasolofonirina R. and Conand C. (1998) L‘exploitation des holothuries dans le sud-ouest de Madagascar, région de Toliara La bêche-de-mer. Bulletin de la CPS, 10:10-13. Razafindrainibe, H. (2010) Baseline study of the shrimp trawl fishery in Madagascar and strategies for bycatch management, Project TCP/MAG/3201 - REBYC2. United Nations Food and Agriculture Organization, Rome. 111 p. René, F., Poisson, F. and Tessier, E. (1998) Evolution de la pêcherie palangrière ciblant l'espadon (Xiphias gladius) à partir de La Réunion. In: Cayré P, Le Gall JY (eds.). Le thon : enjeux et stratégies pour l'océan Indien. ORSTOM, Paris. [in French]. Roullot, J. (1989) Utilization of shrimp fisheries by-catch in Madagascar. Project RAF/87/008. United Nations Food and Agriculture Organization, Rome. 37 p. Rousse, E. (2010) La Grande île, de 1642 à 1960. Le quotidien de la Réunion et de l'océan Indien. Available at: http://www.lequotidien.re/actualites/ocean-indien/120683-la-grande-ile-de-1642-1960.html [Date accessed: January 5th, 2011]. [in French]. Schraeder, P.J. (1995) Madagascar. In: Chapin Metz, H. (ed.) Indian Ocean: Five island countries, 3rd ed., Federal Research Division, Library of Congress, Washington D.C. Soumy, M. (2006) Country review - Madagascar. In: De Young, C. (ed.) Review of the state of world marine capture fisheries management: Indian Ocean. United Nations Food and Agriculture Organization, Rome. 458 p. Southern African Development Community (2008) Study and analysis of the status of IUU fishing in the SADC region and an estimate of the economic, social and biological impacts. Volume 2 - Main report. 72 p. Spalding, M.D., Ravilious, C. and Green, E.P. (2001) World atlas of coral reefs. UNEP World Conservation Monitoring Centre, University of California Press, Berkeley, 416 p. Stamatopoulos, C. and Rafalimanana, T. (1991) Observations sur le système statistique de la pêche traditionnelle (maritime et continentale) à Madagascar. Project MAG/85/014. United Nations Food and Agriculture Organization, Rome. 12 p. [in French]. Sumaila, U.R., Alder, J. and Keith, H. (2006) Global scope and economics of illegal fishing, Marine Policy, 30(6): 696-703. Tucker, B., Tsimitamby, M., Humber, F., Benbow, S. and Iida, T. (2010) Foraging for development: a comparison of food insecurity, production, and risk among farmers, forest foragers, and marine foragers in southwestern Madagascar. Human Organization, 69(4): 375-386. Wielgus, J., Zeller, D., Caicedo-Herrera, D. and Sumaila, U.R. (2010) Estimation of fisheries removals and primary economic impact of the small-scale and industrial marine fisheries in Colombia. Marine Policy, 34: 506-513. World Bank (2010) Madagascar country profile. Available at: http://data.worldbank.org/country/madagascar. [Date accessed: October 12th, 2010]. Zeller, D., Booth, S., Davis, G. and Pauly, D. (2007) Re-estimation of small-scale fishery catches for U.S. flagassociated island areas in the western Pacific: the last 50 years. Fisheries Bulletin, 105: 266-277.  36  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Appendix Table A1: Annual catches by domestic and foreign fisheries in Madagascar‘s EEZ, 1950-2008. Domestic fisheries (t) Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  Data reported to FAO 2400 2500 2500 2600 2600 2600 2600 3000 3000 3500 4000 4000 3500 4000 5501 7801 17500 16600 16900 13400 15100 15800 16200 22401 28701 19501 18451 19760 18160 17260 17373 16875 20455 21195 35038 35112 44353 52488 61141 67731 73515 71438 77021 84317 86618 85840 84644 86547 84405 87638 90167 93615 99326 99671 103416 99986 100943 115148 87834  Reconstructed catch 14295 15170 16070 16995 17946 18923 19925 20942 21447 22811 24205 25243 26304 27387 28491 29617 30883 36787 40937 44617 54319 62029 66801 72408 77014 69410 68934 74354 75462 74619 79804 83133 87439 91293 96501 101705 107033 114869 111675 112374 115101 121887 118912 126392 137783 128033 132157 138809 139285 129942 137120 140240 140689 140276 138493 128148 133449 134454 131771  Foreign fisheries (t) Purse-seine Longline Longline (EU) (Asia) (France) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 520 1040 1550 8125 7400 8000 11250 14500 14889 15278 15667 16056 16444 16833 17222 17611 18000 18000 18000 18000 18000 18000 18000  0 847 1695 2542 3390 4237 5085 5932 6780 7627 8475 9322 10169 11017 11864 12712 13559 14407 15254 16102 16949 17797 18644 19492 20339 21186 22034 22881 23729 24576 25424 26271 27119 27966 28814 29661 30508 31356 32203 33051 33898 34746 35593 36441 37288 38136 38983 39831 40678 41525 42373 43220 44068 44915 45763 46610 47458 48305 50000  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 263 526 789 1053 1316 1579 1842 2105 2368 2632 2895 3158 3421 3684 3947 4211 4474 4737 5000  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  37  Appendix Table A2: Six most important taxa caught by domestic fisheries in Madagascar‘s EEZ, 1950-2008. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  Hemiramphidae  Scombridae  3043 3232 3426 3626 3831 4042 4259 4481 4587 4882 5184 5409 5639 5874 6113 6357 6606 7002 7407 7823 8367 8926 9065 9203 9341 9478 9818 10164 10515 11050 11595 12152 12721 13300 13891 14494 14653 14803 15022 15232 15432 15530 15608 15681 16095 16494 16878 17248 17605 17948 18278 18743 19193 19620 20031 20427 20807 20902 20991  1246 1323 1403 1484 1568 1655 1744 1835 1878 1999 2123 2215 2309 2405 2503 2603 2705 2867 3033 3203 3426 3655 3711 3768 3824 3880 4020 4161 4305 4524 4747 4975 5208 5445 5687 5934 5999 6061 6151 6236 6318 6358 6391 6420 6590 6753 6910 7062 7208 7348 7483 7674 7858 8033 8201 8363 8519 8558 8594  Fenneropenaeus indicus 274 291 308 325 343 361 380 389 414 440 459 479 499 519 540 561 689 1436 1856 2183 3522 4489 5271 5309 5068 5242 5107 5603 5466 4823 5296 5340 5503 5581 5846 6078 6539 7216 6427 6438 6415 6973 6419 7477 7884 7003 7211 8200 8561 7537 8195 7990 8942 9098 8363 7189 7233 6554 5717  Haemulidae  Lethrinidae  Elasmobranchii  Others  677 719 762 806 852 899 947 996 1020 1086 1153 1203 1254 1306 1359 1414 1469 1964 2284 2554 3388 4022 4460 4491 4369 4450 4475 4857 4885 4647 5074 5249 5502 5708 6043 6380 6823 7366 6905 6952 6283 6713 6363 6991 7042 6764 6890 7421 7744 7047 7405 7011 6172 6203 6144 5415 5558 5463 5323  1049 1114 1181 1249 1320 1393 1468 1544 1581 1682 1787 1864 1943 2024 2107 2191 2277 2413 2553 2696 2884 3076 3124 3172 3219 3266 3383 3503 3624 3808 3996 4188 4384 4584 4787 4995 5050 5102 5177 5249 5318 5352 5379 5404 5547 5684 5817 5944 6067 6185 6299 6459 6614 6762 6903 7040 7170 7203 7234  605 642 681 721 761 803 846 890 912 970 1030 1075 1121 1167 1215 1263 1313 1395 1477 1561 1675 1790 1821 1849 1875 1902 1970 2041 2110 2214 2324 2621 3003 3381 3758 4141 4394 4860 5013 4834 5156 5411 5969 5824 6258 5257 5829 5809 5788 5756 5736 5735 5728 5727 5725 5715 5714 5665 5615  7402 7850 8310 8784 9270 9770 10282 10806 11054 11752 12469 12998 13539 14091 14654 15228 15824 19711 22328 24598 31057 36071 39348 44618 49318 41190 40162 44025 44556 43554 46771 48608 51119 53294 56489 59683 63574 69461 66980 67432 70178 75549 72781 78596 88369 80078 82622 87125 86312 78121 83725 86627 86182 84834 83124 73998 78449 80109 78298  38  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  39  RECONSTRUCTION OF MARINE FISHERIES CATCHES FOR MAURITIUS AND ITS OUTER ISLANDS, 1950-20081 Lea Boistol, Sarah Harper, Shawn Booth and Dirk Zeller  Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, V6T 1Z4, Canada l.boistol@fisheries.ubc.ca; s.harper@fisheries.ubc.ca; s.booth@fisheries.ubc.ca; d.zeller@fisheries.ubc.ca  ABSTRACT Total marine fisheries catches by Mauritius and its outer dependencies were estimated from 1950 to 2008, and include unreported catches from the small-scale fisheries carried out around the islands of Mauritius, Rodrigues, Agalega and St. Brandon, recreational marine catches, estimates of catches taken by the Mauritian fleets along the Mascarene Ridge, and discards of the tuna purse-seine fishery. Summed for 1950-2008, total marine fisheries catches for Mauritius and its dependencies were estimated to be 682,392 t, which is 42 percent larger than currently reported landings of 478,305 t presented by FAO on behalf of Mauritius. This discrepancy was largely due to better accounting of small-scale catches carried out around Mauritius and Rodrigues islands by part-time fishers. This study illustrates the need for improved reporting of catches including all fisheries sectors in Mauritius, especially for the small-scale sector, which provides food security and a source of income for a large portion of the local population.  INTRODUCTION Mauritius is an island of volcanic origin located between 20°10′S and 57°31′E about 850 km east of Madagascar (Figure 1). It covers a land area of approximately 1,860 km² and shelters a population of around 1.2 million people. Mauritius is an island state including several dependencies in the Southwest Indian Ocean, namely the island of Rodrigues, the St. Brandon (or Cargados Carajos) group of islands and islets, and the twin islands of Agalega. Mauritius was uninhabited when it was first colonized by the Dutch in 1638, and it has later been under French (1715-1810) and British (1810-1968) rule. The colonization period coincided with large-scale deforestation of the islands for sugar cane farming and the introduction of alien species which have severely damaged the islands‘ ecosystems and indigenous species (Paul, 1987; Sobhee, 2004; Turner and Klaus, 2005). At the time of independence in 1968, the economy was dominated by the sugar industry, and it has later undergone rapid growth and diversification with the development of the textile manufacturing industry and tourism. Mauritius includes a large Exclusive Economic Zone (EEZ) that is approximately 1.7 million km2 (Figure 1). The fishing sector includes small-scale fisheries in the lagoon and non-lagoon areas around Mauritius, Rodrigues, Agalega and St. Brandon islands, offshore semi-industrial fishing on the oceanic banks along the Mascarene Ridge stretching from St. Brandon to Saya de Malha and around the Chagos Archipelago, and on the high seas targeting migratory tuna stocks. Of late, a semi-industrial fishery targeting pelagic swordfish resources has been active since 1999, and two local vessels recently operated a deep sea demersal trawl fishery in the Southwest Indian Ocean from 2000 to 2006 (Jehangeer, 2006; Anon., 2007a) In terms of its contribution to GDP and employment (about 10,000 people), fisheries are of limited economic importance to the national economy. However, the lagoon and inshore fisheries are an important source of employment and food security to many coastal communities of Mauritius and on the island of Rodrigues (Hollup, 2000; Vogt, 2001; Sobhee, 2004; Anon., 2007b; Hardman et al., 2007).  1  Cite as: Boistol, L., Harper, S., Booth, S. and Zeller, D. (2011) Reconstruction of marine fisheries catches for Mauritius and its outer islands, 1950-2008. pp. 39-61. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727].  40  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Tourists, partly through their recreational fishing activities, also add to the fishing pressure on marine resources (Paul, 1987; Sobhee, 2006). However, catches of the recreational and small-scale fishing sectors are often underreported in the official statistics and especially for the Western Indian Region (Van der Elst, 2005). The FAO FishStat database, which currently offers time series data on marine fisheries landings from 1950 to the present, is based on national statistical data supplied by its member countries. Therefore, the quality of the FAO data depends on efficiency of statistical collection within these countries. FAO data have been the basis for many influential fisheries studies (Pauly et al., 1998), but they are, in fact incomplete (Zeller et al., 2006). On Mauritius and Rodrigues Islands, the lagoon and nonlagoon areas are exploited by many people from different sectors, who fish for commercial, subsistence and recreational purposes. Most are not professional but amateurs, people in Figure 1. Map of the Mauritius Exclusive Economic Zone which search of a meal or to supplement what includes Mauritius, Rodrigues, Agalega, and St. Brandon is generally a meager income. In Islands. addition, both professional and amateur fishers use illegal and destructive fishing techniques, such as fine meshed nets, illegal spearguns, dynamites and chemical agents (Ardill, 1979; Paul, 1987; Hollup, 2000; Sobhee, 2004, 2006). Although such catches are not ignored and have been mentioned in recent and past studies (Paul, 1987; Pearson, 1988), they have never been estimated over a long time period, even though long time series of fisheries catches are necessary to evaluate the ecological effect of fisheries on the marine ecosystems. In this context, the purpose of the present study is to reconstruct the likely total catches of marine resources for the 1950-2008 time period following Zeller et al. (2007), to serve as a scientific baseline in the face of climate change and potential threats to food security.  METHODS The existing reported catches were first examined. Such data were extracted from the FAO FishStat database, which currently offers time series data on marine fisheries landings from 1950 to 2008, and from national documents or from the Ministry on behalf of the Albion Fisheries Research Centre. For comparison of pelagic and non-pelagic species catches, we grouped Albacore (Thunnus alalunga), Bigeye tuna (Thunnus obesus), Black marlin (Makaira indica), Indo-Pacific sailfish (Istiophorus platypterus), ‗Marlins, sailfishes, etc. nei‘, Striped marlin (Tetrapturus audax), Swordfish (Xiphias gladius), ‗Tuna-like fishes nei‘, Skipjack tuna (Katsuwonus pelamis) and Yellowfin tuna (Thunnus albacares) as ‗pelagic species‘ and the remaining taxa as ‗non-pelagic species‘. We compared the FAO reported catches with those reported at the national level to identify discrepancies between the two. We then identified the missing components (i.e., sectors, time periods, species, gears) not covered by the existing reported catch time series through literature searches and consultations with local experts. After a search for other available and reliable sources to supply the missing catch data, we developed data anchor points in time for missing data. Time series data were reconstructed using interpolations and extrapolations. Data used to form these anchor points range from fisher and human population data, tourists arrivals, to catch per fisher data.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  41  Human population  Population (x 104 )  Population (x 106 )  The marine environment of Mauritius and its outer islands has been exploited since first settlement, but this study is limited to the period of global FAO reporting, i.e., from 1950 onwards. Our reconstruction comprises the following components: a) ‗unreported‘ catches of the small-scale, near-shore fisheries carried out around the islands of Mauritius, Rodrigues and St. Brandon; b) ‗recreational‘ marine catches; c) estimates of the Mauritian banks fishery catches; d) illegal catches taken in the lagoon and non-lagoon waters of Mauritius and Rodrigues islands; and e) ‗discards‘ of the industrial tuna purse-seine fishery. For the purpose of the present study, and a) 1.4 according to the data that were 1.2 available to us, we distinguished between three different categories of 1.0 local fishers, namely the professional full-time fishers, the part-time 0.8 professional fishers and the part-time 0.6 subsistence fishers. Thus, we reconstructed the catches for each of 0.4 these categories, separating the 0.2 commercial and subsistence components for each. We 0.0 reconstructed separately the catches b) 4.01950 1960 1970 1980 1990 2000 for the different dependencies and fishing areas of the Mauritian state, 3.5 although FAO data do not distinguish 3.0 between them. 2.5 2.0  Tourists (x 105 )  1.5 Population statistics for Rodrigues and Mauritius Islands were extracted 1.0 from the Census Statistics Office 0.5 (CSO) website 0.0 (www.gov.mu/portal/site/cso) and c) 1960 1970 1980 1990 2000 reports. For years when data were 10 1950 unavailable, population numbers were derived by interpolating linearly 8 between adjacent figures. Thus, a complete time series of the human 6 population was derived from 1950 to 2008 (Figures 2a, b). The numbers of 4 tourist arrivals on Mauritius Island were extracted from several sources 2 (Paul, 1987; Gabbay, 1988; Anon., 2008; Anon., 2009b). Although 0 commercial flights to Mauritius began 1950 1960 1970 1980 1990 2000 in 1946, we assumed that tourist‘s Year arrivals were zero in 1950. Linear Figure 2. Human population data, 1950-2008 for: a) Mauritius interpolations were used to estimate Island; b) Rodrigues Island; and c) Tourist on Mauritius Island. tourist arrivals for intervening years when no data were available (Figure 2c). While the St. Brandon islands shelter no permanent human population, some 300 people live on the twin islands of Agalega (see below).  Mauritius Island small-scale fisheries Mauritius Island is almost totally surrounded by a fringing coral reef enclosing a lagoon of more than 300 km². Small-scale fishing takes place within the lagoon and non-lagoon areas beyond the reef on the  42  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  narrow shelf area around the island. Fishers use a wide spectrum of fishing gears which ranges from hand collecting to large nets, gill nets, canard nets, hooks and lines, basket traps and harpoons. To this must be added unreported catches taken by illegal fishing methods such as dynamite and pesticides. The main taxa caught include Serranidae (groupers), Siganidae (rabbitfish), Lethrinidae (emperors), Lutjanidae (snappers), Scaridae (parrotfish), Mullidae (goatfish), Mugilidae (mullets), Acanthuridae (surgeon fish), octopus and lobsters. With the use of hands, sticks and other simple gears, part-time fishers can take small fishes, molluscs, crustaceans and other edible marine species. Many use small cast and mosquito nets to catch large numbers of immature fishes (Paul, 1987). Professional full-time fishers Professional full-time commercial catches: On Mauritius Island, artisanal fishery landings have been monitored since 1946 (Ardill, 1979). However, the method employed as well as the managerial efficiency has fluctuated over the years, thus changing the reliability of the reported landings over time. Therefore, in order to provide a more accurate estimate of total artisanal catches, we retained the more reliable estimates, which we used as anchor points for our reconstruction of the artisanal fishers commercial catches. Linear interpolations were used between anchors points to derive a complete catch time series for the artisanal catches from 1950 to 2008. 1950-1958: From 1946 to 1958, officials have added to the ―controlled catch‖ an estimated amount of 560 t for the uncontrolled yield. This quantity included ―(1) an amount of fish artisanal fishermen retained for their own consumption, for their ‘curry’ or somewhere between 1 and 1.5 kg daily per fisher, (2) some 200 t thought to be landed at some 17 minor but uncontrolled landing stations and, (3) an estimated 50 t taken by sport fishers and countless amateurs‖ (Paul, 1987). Therefore, we used the official estimates of total catch presented by Paul (1987), from which we subtracted the official estimated amounts retained by the fishers for their own consumption (310 t) and the official estimated catch of the sport and countless amateur fishers (50 t). These will be considered separately (see below). 1959-1976: Two different samples of catch data were available to us. The first represented the artisanal catches estimated from 1960 to 1970 by multiplying the catches gathered by the Protection Service by the raising factor of 1.7. The second referred to a later method, in which the total catch was estimated from 1960 to 1977 by multiplying the controlled catch by 3.44, a correction factor arbitrarily chosen to bring total yield to ‗expected‘ levels (Ardill, 1979; Paul, 1987). Therefore, and given that the coverage of the landing stations became sporadic after 1967 (Ardill, 1979), we assumed that the earlier estimation method provided the more accurate and conservative estimates of catch from 1960 to 1966 and we retained those estimates as anchor points. For 1974, Moal (1975 in Paul, 1987) reported an estimate of 1,100 t, which was not consistent with our other data, so we assumed that this only represents a fraction of the total catch. 1977-2008: For this time period, we retained the artisanal catches estimated by the Fisheries Division as the more reliable estimates. Indeed, since 1977, a frame survey system for the collection of statistics on fish landings has been operational (Ardill, 1979), and the method was described in the literature (FAO, 1983; 1987; Samboo and Mauree, 1987; Anon., 2007b). Catch and effort data are collected on a monthly basis by a team of enumerators covering randomly selected landing stations divided into several strata, and raising factors are used to estimate catch and effort for each stratum separately. For 1982, Paul (1987) estimated the total catch of the professional artisanal fishers at 1,373 t. However, once the subsistence part of the catch was removed, the remaining amount, namely the commercial catch, was consistent with that reported by the Ministry for the same year. Therefore, landings collected by the Fisheries Division were taken as the best estimates for the artisanal commercial catches for this time period. To estimate the taxonomic composition of the professional fishers‘ commercial catches, we used data from Paul (1987) which provided the estimated changes in species composition of total controlled catch from 1957 to 1982 (Table 1).  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  43  Table 1. Estimated species composition of total professional full-time fishers‘ catches on Mauritius Island, based on estimated changes in species composition of total controlled catch (Paul 1987). Taxon Name  Common Name  Lethrinidae Lethrinidae Lethrinidae Siganidae Mullidae Mugilidae Serranidae Acanthuridae Octopodidae Lobsters Miscellaneous  Capitaine Dame Berri Battardet/Caya Cordonnier Rougets Mulets Vieilles Licornes Octopus Langoustes Others  a  1950 1957 3.9 5.4 5.3 17.4 5.2 4.5 4.9 8.6 13.2 1.8 29.8  1958 – 1964a 3.4 4.4 5.4 17.0 3.9 3.3 5.4 8.6 16.4 2.1 30.1  1965 1971 2.0 4.0 3.1 19 4.1 2.9 5.2 9.6 17.5 1.6 31.0  1972 1981 a 2.2 3.1 4.4 10.9 3.1 4.1 5.2 15.6 15.8 0.5 35.1  1982 2008 a 9.9 12.8 4.1 7.9 7.1 7.0 11.7 0.3 39.2  Note change in time period coverage.  The ‗others‘ category (Table 1) includes pelagic fishes. In the 1950s, the catch of clupeoids constituted an important seasonal activity of traditional fishers, but catch of those species followed a sharp decline while the catch of scombroid type fishes increased. Capture of sharks also occurred from the 1950s and 1970s, and 74 t of shark was caught in 1963 (Paul, 1987). Professional full-time subsistence catches: The custom is for professional fishers to keep a small amount of their daily catch for their own consumption (Gonzalez Manero, 1971; Paul, 1987). This amount has been estimated between 1946 and 1958 as part of the uncontrolled landings. However, we assumed that this amount would not have been included in the official reported catches since 1958. The amount retained by fishers is fairly constant, independent of the size or value of the catch (Gameiro, 2003). Therefore, for the 1959-2008 period, rather than using a percentage of the total commercial catch, we used the number of artisanal fishers, an effort of 176 fishing days per artisanal fishers, and the average amount of 1.25 kg retained per fisher per day as reported by Paul (1987). For 1974 and 1982, we used 1,000 and 1,500 professional artisanal fishers, respectively, as reported in Paul (1987), while the number of active professional fishers from 1999 to 2008 were provided by the CSO (Anon., 2009a). For the period 19501958, we rely on the description of the uncontrolled catch by Paul (1987), and we concluded that a total standard of 310 t was yearly retained by the artisanal fishers during this time period for subsistence purposes. We used linear interpolations between the figures reported or calculated as above to establish the subsistence catches of the artisanal fishers from 1950 to 2008. Professional part-time fishers In addition to the professional full-time fishers, numerous part-time fishers exploit the lagoon and nonlagoon environment. The common species in their catch are from the families Siganidae (rabbitfish), Serranidae (groupers), Lethrinidae (emperors) and Carangidae (jacks) (Samboo, 1987). A report by Roullot et al. (1988) mentioned that part-time fishers also visit Fish Attracting Devices (FADs), catching pelagic fish using trolling and hand lining. From 1977 to 2008, the official reported catches for the socalled ‗amateur fishery‘ consisted of the constant amount of 300 t. In a recent survey, the Ministry of Fisheries on Mauritius found that 23,400 persons were involved in ―recreational‖ fishing in the lagoon of Mauritius, from which about 1,000 were owners of a boat. Indications were that their catch could be more than the current estimate of 300 tonnes annually (Jehangeer, 2006). (Samboo and Mauree, 1987) mentioned that ―the quantities caught by the part-time fishers may exceed the catch by the commercial fishers‖. Paul (1987) and Moal (1975 in Paul, 1987) differentiated between two categories of part-time fishers. The first represents people who directly consume as well as sell a part of their catch; the second consists of local people who fish only for their own consumption. For clarity purposes, we will use the terms part-time professional fishers and part-time subsistence fishers for the first and second categories, respectively. Furthermore, in his estimate of catches for the part-time subsistence fishers in 1982, Paul  44  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  (1987) included catches from the fishing tourist population, but excluding the pelagic sport fishery catches. However, as we aimed at discriminating between the commercial, subsistence and recreational sectors, we rather included the fishing tourist catches in the recreational catches. Part-time professional fishers fish for their own consumption and sell the surplus of their catch. However, we assumed that their commercial catches are not included in the recorded commercial landings, for the following reasons. First, they are fishers who generally do not operate through a middleman 2 and many have developed their own sales outlets to commercialize the surplus of their catch (Paul, 1987). There are some indications that many hotels and restaurants which cater to the tourist trade are provisioned directly by contracted fishers, and that their catch escapes detection by the fisheries authorities and are unaccounted for in landings statistics (Paul, 1987; Sobhee, 2004). Moreover, many of them use illegal fishing gear. To respond to an increasing demand for seafood, reduced catches, and new regulations, many amateur fishers have currently resorted to illegal fishing methods, using fine meshed nets, illegal spearguns and landing of undersized fish (Paul, 1987; Hollup, 2000). Therefore, it was assumed that their commercial catches were not included in the reported artisanal catches. Table 2. Data sources and method used to estimate the part-time professional fishers population of Mauritius Island from 1950 to 2008. These data were converted to percentage of the total population which we used, together with assumptions and the total human population time series, to derive estimates of the part-time professional fishers population from 1950 to 2008. Year Number of part-time Human Ratio Source and method professional fishers population (%) 1950  1,566  506,663  0.31  25% decrease in ratio from 1950 to 1974  1974  2,000  857,063  0.23  Moal 1975 (in Paul, 1987)  1982  2,000  960,994  0.21  Paul (1987)  2008  2,562  1,230,995  0.21  1982 ratio maintained unaltered  Professional part-time fishers’ population: Paul (1987) and Moal (1975, in Paul, 1987) reported 2,000 part-time professional fishers in 1974 and 1982. We converted these figures to ratios of the total population for the corresponding years. The calculated ratios, together with the total island‘s population time series and assumptions as described below, were used to derive complete time series estimates of part-time professional fisher population from 1950 to 2008 (Table 2). We assumed that the proportion of the Mauritian population involved in such fishing activities decreased by 25% from 1950 to 1974. A greater proportion of the population likely relied on fishing for food or income purposes in the earlier time period. The progressive diversification of the economy away from sugar cane after independence in 1968 (Paul, 1987; Houbert, 2009), together with new and cheap supply of frozen fish from the offshore banks fishery in the late 1960s (Christy and Greboval, 1985; Ardill, 1986) would have contributed to reduce the need to fish for food or income complement. We also assumed that the same proportion of the population was involved in this activity from 1982 to 2008. This is a conservative estimate when we consider the ever-growing population and worsening poverty in the coastal regions (Sobhee, 2004). A high unemployment rate was also of concern in the 1980s as well as in the more recent period (Paul, 1987; 2010). Professional part-time fishers’ catch rate: According to Moal (1975 in Paul, 1987), 2,000 part-time fishers caught 700 t in 1974. This suggests a catch rate of 0.35 t∙fisher-1∙year-1 in 1974. Paul (1987) presented the change of the average catch rate for an artisanal fisher from 1948 to 1982. From the catches and number of artisanal fishers reported by the CSO (Anon., 2009b), we derived estimates of catch rates for the period 1999-2008. We assumed that the productivity of the part-time fishers would have changed similarly to that of the professional. Indeed, it is very likely that their productivity has declined similarly, as a consequence of increasing numbers of fishers exploiting the same areas. To the reported and calculated catch rates, we applied an exponential model. Using the growth constant of this model and the 2On  Mauritius, a large number of professional fishers are dependent on middlemen for equipment. Middlemen usually buy the entire catch, and finance the fishers who are then indebted due to these cash advances. Locally named banyan, middlemen are usually Muslim traders from the urban areas.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  45  catch rate of 0.35 t∙fisher-1∙year-1 for 1974, we derived estimates for part-time professional fisher‘s catch rates for the whole time period (Figure 3), declining from 0.56 tonnes∙fisher-1∙year-1 in 1950 to 0.18 tonnes∙fisher-1∙year-1 in 2008. Finally, using the fisher population and catch rates as estimated above, we derived the catches of the parttime professional fishers from 1950 to 2008.  Catch rate (t/fisher/year)  1.6 Professional part-time Professional full-time Model 1: y = 1.5e-0.0195x fishers’ commercial vs. 1.4 fishermen's productivity Model 2 : y = 0.6e-0.0195x subsistence catches: For (model 1) 1.2 1974, Moal (1975 in Paul, 1987) divided the part-time 1.0 professional fishers‘ total Part-time professional 0.8 catch into subsistence and fishermen's productivity 0.6 commercial components, (model 2) being 29 and 71 percent of (0.35;1974) 0.4 the total catch, respectively. 0.2 We assumed that the subsistence part of the catch 0.0 was higher in the earlier 1948 1961 1974 1987 2000 time period, thus reflecting Year the increasing development Figure 3. Data points and fitted models used to estimate part-time of the tourism industry professional fishers‘ productivity for Mauritius, 1950-2008. Data points are which likely provided new represented by open circles (for professional full-time fishers) or dark circles employment opportunities (for part-time professional fishers). Data sources: Paul (1987); Moal (1975 in as well as new sale outlets Paul 1987); Anon. (2009a). for the part-time fishers. Also, in the late 1960s, the development of the banks fishery contributed to provide a supply of local frozen fish at relatively low prices (Christy and Greboval, 1985; Ardill, 1986) which would have reduced the need to fish for subsistence by the low income population. Therefore, we inverted those percentages for 1950 so that the subsistence and commercial catches represented 71 and 29 percent of the total part-time professional fishers‘ catch in 1950, respectively. Linear interpolations were used to estimate the percentages for 19501974, while the 1974 percentages was carried unaltered to 2008.  Part-time subsistence (only) fishers Part-time subsistence fishers fish only for their own consumption. Most of them are coastal residents. This category of fishers likely comprises a substantial proportion of owners of the many houses along the island‘s shores (Paul, 1987). They usually fish close to shore with rod and reel or as a group on a boat. Part-time subsistence fishers’ population: In 1982, a sample survey of the entire coast indicated that about 10,000 people were fishing only for their own consumption (Paul, 1987). This amount was taken as a percentage of the total human population of Mauritius Island for the same year and came to about 1 percent. Moal (1975 in Paul, 1987) reported a larger estimate of 65,000 people or 7.6 percent of the island‘s population. However, Moal's estimate was considered unreliable. Indeed, to obtain this figure, Moal (1975 in Paul, 1987) simply added the coastal population based on the 1972 census and subtracted the estimated size of the vegetarian and artisanal fishing population. Therefore, we retained the one percent figure as the more conservative estimate. However, while we applied this percentage for 19822008, we assumed that it would have decreased from 1950 to 1982, similarly to that of the part time professional fishers, where the percentage of the number of part-time professional fishers to the total population declined from 0.31 to 0.21 percent from 1950 to 1982. This proportional decline of 67 percent was then applied to the reported 1 percent figure for 1982 back to 1950, so that the percentage of local people fishing only for their subsistence declined from an assumed 1.55 percent in 1950 to a reported 1 percent in 1982. Linear interpolations were used to estimate the percentages in the intervening years.  46  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Using the human population data for Mauritius Island, we derived the number of local people fishing only for their own consumption for each year from 1950 to 2008. Part-time subsistence fishers’ catch rates: For 1982, Paul (1987) estimated the catch rate of a local part time subsistence fisher at 10 kg∙fisher-1∙year-1. We maintained this catch rate unaltered from 1982 to 2008. However, we assumed that it would have been decreasing since 1950. From our precedent calculations, we derived that the part-time professional fishers‘ catch rate declined from 0.56 to 0.30 t per fisher per year from 1950 to 1982. This proportional declined of about 54 percent was then applied to the reported rate of 10 kg per fisher per year for 1982 back to 1950, so that the catch rate of the local people fishing for their subsistence declined from an assumed 18.7 kg∙fisher-1∙year-1 in 1950 to a reported rate in 1982 of 10 kg∙fisher-1∙year-1. Linear interpolations were used to estimate the catch rates in the intervening years. Using the fisher population as estimated above, these derived annual catch rates were expanded to determine the part time subsistence fishers‘ catches from 1950 to 2008. Other unreported catches Catches taken by illegal methods are common on Mauritius Island and are largely unreported (Ardill, 1979; Paul, 1987; Hollup, 2000; Sobhee, 2004). Ardill (1979) mentioned that sampled landings, aside from ignoring part-time fishers‘ catches, also excluded estimates of catch taken in illegal nets and by dynamite fishing, which are landed all along the coast. Other illegal and destructives fishing practices include fine meshed nets, illegal spearguns, night fishing, capture and landing of undersized fish, and pesticides. These practices involve both part- and full-time fishers, and have been rising due to the dwindling resources, reduced catches and new regulations (Hollup, 2000; Sobhee, 2004). In the absence of quantitative information specific to Mauritius Island, we relied on the knowledge acquired for Rodrigues (see below). We assumed that these illegal fishing methods evolved similarly on the two islands, and therefore used the same method as we did for Rodrigues to estimate this component of the unreported catch. Although underwater fishing was forbidden in 1982 in Mauritius, this practice is still used today (90 interventions occurs in 2006). Underwater fishers are also said to be responsible for the use of dynamite and anesthetics, and fishing at night (Paul, 1987). Estimated catches ranged from 150 t in 1974 (1975 in Paul, 1987) to 397 t in 1982 (Paul, 1987), but we assumed that such catches were included in our estimates of unreported catches.  Rodrigues Island small-scale fisheries Rodrigues Island (19°43‘S-63°25‘E, 110 km², 37,500 inhabitants) is located some 586 km northeast of Mauritius and it is a semi-autonomous island since 2001. Unemployment, poverty and illiteracy are high compared to the main island of Mauritius, and tourism is in its infancy. As a consequence, the lagoon fisheries are very important to Rodrigues, but they are currently severely overexploited (Blais et al., 2007). The island is surrounded by 90 km of fringing coral reef, enclosing a shallow lagoon of 240 km². Small-scale fishing takes place within and outside the lagoon, with the use of seine nets, hand-lines, basket traps and spears or harpoons for octopus and large fish. Professional fishers Professional commercial fishers: Total catches reported by the FAO on behalf of the Ministry do not distinguish between Mauritius and its outer dependencies. Nevertheless, comparison of the FAO data with other reports (Fisheries Division annual reports, 2003-2007; Harris, 1988) indicated that from 1977 to 2000, artisanal catches for Rodrigues have been estimated between 1,200 and 1,500 t annually. In Rodrigues, monitoring of artisanal catches started in 1994 (Reshad Jhangeer-Khan, Shoals Rodrigues, pers. comm.). Broad catch and effort data are collected at fish landing stations around the island on a regular basis by the Fisheries Research and Training Unit (FRTU), collated annually and transmitted to the Albion Fisheries Research Centre (Anon., 2007b)(E. Hardman, Shoals Rodrigues, pers. comm.). For the period 1994-2008, we obtained sampled catch data from the CSO annual digests of statistics on Rodrigues and from Dr Emily Hardman (for the octopus and lagoon fish catches reported by the CSO from 1994 to 2006). However, these reported catches refer to the registered fishers catches, and do not include unofficial catches landed by the amateurs and illegal catches (Anon., 2009c) (Dr Emily Hardman,  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  47  Shoals Rodrigues 2005-2008, pers. comm.; Reshad Jhangeer-Khan, Shoals Rodrigues, pers. comm.), and we retained those sampled catches as anchor points for the professional fishers‘ catches. For the earlier period, professional fishers‘ catches were compiled from various sources, using the more reliable estimates as anchor points. For 1955, 1962 and 1968, catches reported by Pearson (1988) were equivalent to the export data. Therefore, those amounts were only used as minimum. We interpolated linearly between our anchors points, in order to estimate the professional fishers‘ catches for the whole time period. Taxonomic breakdown: In order to determine the taxonomic composition of the artisanal commercial catches taken on Rodrigues Island, we used data from the Fisheries Research and Training Unit (FRTU) (unpublished data) and Central Statistics Office reports. The FRTU data provided the species composition of the seine net fishery for the years 1994 to 1999, and 2004 to 2006. We interpolated linearly between the 1999 and 2004 available figures to estimate the species composition of the seine net fishery for the period 2000-2003, while the 2006 figures were maintained unaltered until 2008. For the period 19942008, CSO data provided the total controlled catch on Rodrigues, discriminated by type of fishery, namely the octopus, lagoon and non-lagoon fisheries. This allowed us to calculate their respective contributions as percentages of the total controlled catch from 1994 to 2008. Similarly, for the period 1994-2006, FRTU data provided the seine net controlled catch, which we converted into percentage of the total controlled catch on Rodrigues. The calculated percentage for 2006 has been carried forward to 2008. Thus, knowing the respective contribution of the seine net, octopus, lagoon (other than seine net) and non-lagoon fisheries, we adjusted the FRTU data on the seine net species composition to estimate a breakdown of the total artisanal commercial catch on Rodrigues for the period 1994-2008 (Table 3). Furthermore, Sauer et al. (2011) indicate that the octopus fishery on Rodrigues is dominated by Octupus cyanea with the remaining catch being mainly O. vulgaris. The catches presented in Sauer et al. (2011) correspond to the CSO data for octopus. Professional fishers’ subsistence catches: Registered fishers will also keep some of the fish and octopus that they catch for their own consumption (Dr Emily Hardman, Shoals Rodrigues, pers. comm.). Therefore, these retained amounts were estimated and added to our total subsistence catches. We used the estimate of 1.25 kg of fish retained by each fisher per fishing day, which we multiplied with Paul‘s estimate of 176 fishing days for the professional fishers to obtain a catch rate of 0.22 t∙fisher-1∙year-1. This number of fishing days was consistent with these reported by (Pearson, 1988). Indeed, for 1987, he mentioned that the number of fishing days for the trap and handline fisheries rose to about 290 and 238 days, respectively, while that of the seine-net fishery fell from about 200 days in the earlier time period to 163 in 1987, due to regulations being implemented for this fishery. The professional fishers‘ population was reconstructed using various sources (Gonzalez Manero, 1971; Moal, 1971 in Paul 1987; Ardill, 1979; Paul, 1987; Pearson, 1988; Anon., 2010). Finally, the 0.22 t∙fisher-1∙year-1 catch rate estimate as calculated above, together with the artisanal fisher population data, were used to derive the subsistence catches amounts for the whole time period. Part-time fishers Fishing by part-time fishers, mainly for subsistence, on Rodrigues Island is unmonitored, although it is likely to be considerable, and it currently includes in and off-lagoon fin fishing using lines and traps, inlagoon octopus fishing and shell fishing (Reshad Jhangeer-Khan, Shoals Rodrigues, pers. comm.). In addition to the registered fishers, there are many part-time fishers (Gonzalez Manero, 1971; FAO, 1983; Paul, 1987; Anon., 2009c). Some of them are regular fishers who fish to feed their family or to sell, while others have a full-time job and fish for pleasure and their own consumption (Dr Emily Hardman, Shoals Rodrigues, pers. comm.). Therefore, their catches likely comprise commercial as well as subsistence components. However, catches of the part-time fishers were assumed to be mostly of subsistence purpose. In order to estimate these catches, we first assessed the annual number of part-time fishers, then we estimated the catch rates of those fishers, based on Pearson (1988).  Table 3. Estimated taxonomic composition of the total small-scale commercial catch for Rodrigues Island. Sources: Fisheries Research and Training Unit (FRTU) (unpublished data), Anon (2009a), Central Statistics Office of Mauritius. Portion of total catch (%) Fishery  2000  2001  2002  2003  2004  2007  2008  Rabbitfish  Siganus spp.  1.7  1.1  1.5  1.9  2.6  2.1  2.6  2.8  4.7  3.4  4.8  3.2  5.0  5.0  5.0  Spangled emperor  Lethrinus nebulosus  2.1  1.1  1.3  1.1  1.0  0.2  0.6  1.0  2.3  2.1  3.8  4.1  3.3  3.3  3.3  Unicornfish  Naso spp.  1.6  0.9  1.6  1.0  1.1  0.4  0.9  1.3  3.0  2.8  4.9  2.4  4.1  4.1  4.1  Mullet  Mugilidae  2.2  3.3  1.6  1.5  1.9  2.7  2.9  2.8  4.1  2.5  2.9  1.8  2.7  2.7  2.7  Trevally/jacks  Carangidae  1.8  2.2  1.0  1.6  1.0  1.0  1.1  1.1  1.6  1.0  1.3  1.1  1.3  1.3  1.3  Strongspine silverbiddy  Gerres longirostris  1.8  1.1  0.8  0.5  0.3  1.2  1.3  1.3  2.0  1.3  1.7  1.4  2.1  2.1  2.1  Goatfish  Mullidae  1.0  1.7  1.0  0.7  0.4  1.3  1.4  1.4  2.2  1.4  1.7  1.7  3.2  3.2  3.2  Parrotfish  Scaridae  0.9  0.2  0.5  0.3  0.2  0.1  0.2  0.2  0.4  0.3  0.5  0.6  0.9  0.9  0.9  Blackspot emperor  Lethrinus harak  0.8  0.4  0.3  0.0  0.1  0.0  0.0  0.0  0.0  0.0  0.1  0.0  0.0  0.0  0.0  Yellowfin bream  Rhabdosargus sarba  0.9  0.2  0.5  0.0  0.2  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  Rudderfish  Kyphosus spp.  0.8  0.5  0.6  0.4  0.1  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  Milkfish  Chanos chanos  0.8  0.7  0.4  0.3  0.2  0.1  0.1  0.1  0.1  0.1  0.1  0.0  0.0  0.0  0.0  Striped threadfin  Polydactylus plebeius  0.6  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  Others  Misc. marine fishes  0.6  0.6  0.9  0.4  0.2  0.0  0.1  0.3  0.8  0.8  1.4  1.7  2.2  2.2  2.2  Octopus  Octopus  Octupus spp.b  51.1  51.1  42.4  44.7  31.9  23.7  23.6  17.0  27.3  34.7  26.9  27.4  25.0  16.7  16.0  Other  Lagoon fishes  Lagoon fishes  31.0  34.0  43.0  45.0  48.0  53.0  50.0  46.0  39.0  41.0  46.0  36.0  35.0  33.0  36.0  Non-lagoon fishes  Pelagic and nonpelagic speciesa  1.0  1.0  2.0  -  11.0  15.0  15.0  25.0  13.0  9.0  4.0  18.0  15.0  25.0  23.0  Seine net  a  Common name  Taxon  1994  1995  1996  1997  1998  1999  .  2005  2006  including Serranidae, Lethrinidae, Lutjanidae, Acanthuridae, Labridae, sharks and pelagic species such as small tunas or barracudas (from Rodrigues’ Offshore Cooperative Society Landings, 1982-1983, in Paul, 1987); b Octopus spp. caught were O. cyanea (80%) and O. vulgaris (20%) based on Sauer et al.(2011).  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  49  Part-time fishers’ population: 1950-1979: We used the estimate of 7,000 part-time fishers given by (Ardill, 1979). This represented about 26% of the island population in 1979. We assumed that this percentage would also apply to the years before 1979, as most of Rodriguan people likely relied on subsistence fishing and agriculture at this time, and thus we maintained this percentage unaltered from 1950 to 1979. 1980-2008: For 1987, Pearson (1988) estimated that 10 percent of the population practiced occasional line-fishing (or about 4,000 Rodriguans). However, it is very likely that part-time fishers also used other gears such as traps and harpoons to catch fish and octopus and so that their population was underestimated by Pearson (1988). A more recent estimate consisted of 2,000 part-time fishers in 2008 (Blais et al., 2007) or about 5% of the island‘s population. However, some believe that more than twice the number of full-time professional fishers are engaged in part-time fishing activities (Anon., 2009c), which would give an estimate of more than 3,500 people. To remain conservative, we retained the estimate of 2,000 part-time fishers provided by Blais et al. (2007) and we interpolated linearly between the percentages of the total population in 1979 (26%) and 2008 (5%). Using the total island‘s population, we estimated the number of part-time fishers on Rodrigues Island from 1982 to 2008 (Table 4). Table 4. Data sources and method used to estimate the population of part-time fishers for Rodrigues Island from 1950 to 2008. Year Number of part-time Human Ratio (%) Sources and method fishers population 1950 1979 2008  3,353 7,000 2,000  12,971 27,081 37,570  25.8 25.8 5.3  1979 ratio maintained Moal unaltered (in Paul, 1987) Blais et al. (2007)  Part-time fisher’s catch rates: 1987-2008: We used the catch rate of 0.1 t∙fisher-1∙year-1 estimated by Pearson (1988). We maintained this catch rate unaltered for the period 1987-2008, assuming that the productivity of the part-time fishers would not have changed during the last two decades. This is likely to be a conservative estimate, as Pearson‘s catch rate estimate for 1987 referred to line-fishing, while subsistence fishers also use traps and fish octopus inside the lagoon (Reshad Jhangeer-Khan, Shoals Rodrigues, pers. comm.). 1950-1986: Over-fishing in Rodrigues may date from the 1800s (Bunce et al., 2008). The island has been affected by intensive land erosion and its heavy impact on the lagoonal fauna has been reported as early as 1962 (Baissac, 1962 in Paul, 1987). It has also been illustrated by a sharp downfall in exports to Mauritius since the 1970s (Paul, 1987; Pearson, 1988; Bunce et al., 2008). Therefore, we assumed that the average productivity of the part-time fishers declined from 1950 to 1986 by a half, so that in 1950 it was equal to 0.2 tonnes∙fisher-1∙year-1. Therefore, we applied a catch rate of 0.2 t∙fisher-1∙year-1 for 1950-1987 and 0.1 t∙fisher-1∙year-1 for 1986-2008. Other unreported catches Additional catches taken in reserves, during the closed seasons (implemented for the seine net and sea cucumber fisheries) and using illegal gears such as undersized seine net mesh sizes, spearguns to catch octopus were unreported (Anon., 2009c) (Dr Emily Hardman, Shoals Rodrigues, pers. comm.; Reshad Jhangeer Khan, Shoals Rodrigues, pers. comm.). These catches are either consumed by the fishers themselves or sold - on the beach or directly to the markets (Reshad Jhangeer Khan, Shoals Rodrigues, pers. comm.). To estimate these unreported catches we assumed that illegal gears were not used prior to the 1960s (Reshad Jhangeer Khan, Shoals Rodrigues, pers. comm.). For 1987, Pearson (1988) estimated that 150 illegal nets may have been operating within the lagoon. Pearson (1988) conservatively estimated unreported catches taken by illegal methods amounted to 10% of total legal landings in the lagoon. In contrast, these catches may be equal to or greater than legal landings (Reshad Jhangeer Khan, Shoals Rodrigues, pers. comm.). To remain conservative, and to take into account the fact that illegal fishing practices are used by both part-time and full-time fishers, we assumed that most of this catch would be  50  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  included in our total reconstructed catch from the small-scale fishery carried out around the islands. Therefore, for the 1987-2008 period, we estimated additional unreported catches in the coastal waters of Rodrigues island as 10% of our total reconstructed small-scale fisheries catches, including both professional and part-time fishers‘ catches. Finally, we interpolated linearly between zero catch in 1960 and our first calculated value for 1987.  Recreational fisheries Mauritius is well-known as a tourist destination. While it creates job opportunities for the local population, it also represents a potential threat to marine life (Paul, 1987; Sobhee, 2004). In Mauritius, the recreational fishery can be divided into two components, namely the extraction of marine resources for leisure purposes inside or close to the lagoon environment, and the pelagic sport fishery, which operates with boats in deeper waters. We estimated these catches separately. Pelagic sport fishery Mauritius is a popular destination for big game sport fishers. A number of leisure and sports fishers operate successfully around FADs (Roullot et al., 1988; Venkatasami and Sheik Mamode, 1995). For this activity, a number of lines are used with rods and outriggers baited for the large migratory carnivorous species. The catch comprises mainly blue marlin (Makaira mazara), black marlin (Makaira indica), Indo-Pacific sailfish (Istiophorus platypterus), striped marlin (Tetrapturus audax) and yellowfin tuna (Thunnus albacares), albacore (Thunnus alalunga) and bigeye tuna (Thunnus obesus). Other species like wahoo (Acanthocybium solandri), shortbill spearfish (Tetrapturus augustirostris), skipjack tuna (Katsuwonus pelamis) – used as bait for marlins, sharks (Sphyrna zygaena, Isurus oxyrinchus, Carcharinus albimarginatus, Carcharinus melanopterus) and dolphinfishes (Coryphaena hippurus) are also caught (Cayre and Stequert, 1988; Norungee et al., 2004; Jehangeer, 2006). Recently, a system of data collection has been set up at the Albion Fisheries Research Centre (Norungee et al., 2004). However, examination of the Ministry reports showed that the currently reported sport pelagic catches consist of the constant amounts of 400 t from 1977 to 1987 and 650 t from 1988 to 2008. Such amounts do not reflect the increasing trend of the tourist population that reaches the island each year. Thus, catches from the sport fishery were re-estimated. We first extracted catch estimates for the sport fishery from other reliable sources. For each reported value, we calculated a per tourist rate using the time series of number of tourists arrivals. Sport fishery likely already existed in 1950. According to Paul (1987), the sum of the countless amateurs and sport fishermen catches were estimated at 50 t by the officials from 1946 to 1958. However, to remain conservative (and in the absence of more detailed information), we assumed that the pelagic sport fishery catches were null in 1950. For the period 1950-1987, we interpolated linearly between the per tourist rates. Between 1974 and 1988, the calculated per tourist rates declines from 2.9 to 2 kg per tourist arrival. In order to reflect the decreasing catches of pelagic species since the 1990s, we carried this decreasing trend forward from the 1988 catch rate figure to derive the catch rates for the 1988-2008 period. Pelagic sport catches were finally deduced by multiplying the catch rates as estimated above by the number of tourist arrivals. Recreational catches in the lagoon In addition to the tourists involved in big game fishing, a substantial part of the tourist population is involved in recreational fishing in the lagoon of Mauritius. For 1982, Paul (1987) assumed that a conservative estimate of the total number of fishing tourists involved in the exploitation of the island‘s waters would be approximately 20,000 people, or about 17% of the tourist arrivals during the year. We assumed that the number of fishing tourists was proportional to the tourist arrivals, and we carried the 17 percent figures for the whole time period. Thus, using the number of tourist‘s arrivals time series, we established the number of fishing tourists from 1950 to 2008. For 1982, Paul (1987) estimated the catch rate of a fishing tourist at 5 kg∙tourist-1∙year-1. We assumed that this catch rate would not have changed from 1950 to 1982, but that the increasing degradation of marine resources of the island, together with an increasing number of people fishing in the lagoon area, would have caused the tourist‘s catch rates to decrease from 1982 to 2008. Therefore, we used the proportional decline of 60% between the part time professional fishers catch rates of 1982 and 2008, so that the tourist catch rates decreased from a reported  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  51  5 kg∙tourist-1∙year-1 in 1982 to an assumed 3 kg∙tourist-1∙year-1 in 2008. We then established the catches using the fishing tourist population time series.  Agalega Islands The twin islands of Table 5. Estimated taxonomic breakdown of the Mauritian banks handline fishery catches on the Nazareth, Saya de Malha, St Brandon and Chagos banks modified from Agalega are located Samboo 1989 (in Mees, 1996). In the Chagos, although it is avoided due to the potential some 982 km north of of ciguatera, the red snapper Lutjanus bohar can represent up to 50 percent of the catch. Mauritius between Taxon Nazareth & Saya St Brandon Chagos 10°28‘S and 56°40‘E. de Malha The islands are of 88.00 84.0 50 coralline origin and Lethrinus mahsena 2.00 2.0 2 cover a total land area other Lethrinidae 4.00 1.5 26 of 70 square Serranidae Carangidae 2.00 2.0 4 kilometers. 1.00 6 Colonization took place Aprion virescens (Lutjanidae) Siganidae 2.5 in the early part of the Scaridae 2.0 18th century and 0.5 fishing was the Mugilidae 0.5 mainstay of the Naso spp. (Acanthuridae) Lutjanus bohar (Lutjanidae) 0.75 (50) settlement‘s diet (Paul, Pristipomoides spp. (Lutjaidae) 0.75 10 1987). Since then, the Tuna 0.75 2 exploitation of the Others 0.75 5.0 coconut trees for copra production began, and it is still the main economic activity. Fish production being very limited due to the steep drop-off outside the reef, is only sufficient for consumption of the local workers. The current reported catches for Agalega consist of the constant amount of 30 t from 1977 to 2008. To arrive at these figures, it has been assumed that the per capita consumption of fish was 100 kg∙person-1∙year-1, giving an annual catch of 30 tonnes for a population of 300. The species composition of the catches is believed to be similar to that of Mauritius Island but this remains to be examined (FAO, 1983). As the population did not change much over the 1950-2008 period (Paul, 1987; Anon., 2006, Anon., 2010) and in the absence of contradictory information, we assumed that the current reported amount was accurate and we carried it back unaltered to 1950.  The Mauritian Banks fishery An important Mauritian fishery occurs on the shallow oceanic banks of the Mascarene Ridge. In addition to the banks located within Mauritius EEZ, the Mauritian fishing fleet fishes on the Saya de Malha bank (much of which is in international waters), and exercises traditional fishing rights within the EEZ of the British Indian Territory (BIOT, Chagos Archipelago). Since the 18th century, fishing on these grounds has been carried out by vessels engaged in inter-island trade, but it was not before the twentieth century that more systematic exploitation began when the Mauritius Fishing Development Company and its sister company the Raphael Fishing Company Limited gained control of the St. Brandon group. In the earlier years, the demersal stocks were exploited mainly for salting purposes, but since the Wheeler and Ommaney (1948-49) pioneer survey, these banks have gradually started to be the main suppliers of frozen fish to Mauritius (FAO, 1983; Ardill, 1986). Fishing on the banks is practiced using handlines from 7-8 m dories carried by refrigerated mother-vessels, 20-60m in length. The main targeted species is Dame Berri (Lethrinus mahsena) which contributes about 80% of the total catch, while the remainder of the catch is made up of serranids, lutjanids, siganids, and carangids. The catch is mostly gilled, gutted and frozen on board. In St. Brandon, however, temporary settlements of fishers continue to fish in the lagoon using seine nets and basket traps while the mother ships are away. Their catches are salted and dried while awaiting shipping to Mauritius (Moal, 1971 in Paul 1987; Samboo and Mauree, 1987). Also, 17 vessels are allowed to operate a semi-industrial chilled fishery on the Soudan, Albatross, St Brandon, Hawkins, Saya de Malha and Nazareth Banks. Their catch is either frozen or chilled at sea and comprises mainly emperors, snappers, groupers and tunas. Catch data for the banks fishery were obtained from the Ministry on behalf of the Albion Fisheries Research Centre (AFRC). However, such data only covered the  52  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  1977-2008 period. When compared to other sources, Table 6. Estimated taxonomic breakdown of the salted-dried fish catches taken in it also appeared that they the St Brandon lagoon (Anon., 1971; 2009a). represented landings in Family or group Taxon name Common name Catch (%) frozen weight, and for the Lethrinidae Lethrinus mahsena Dame Berri 75.0 greatest part of the time Lethrinidae Lethrinus nebulosus Capitaine Gueule Longue 10.0 period, catches of salted- Siganidae Siganus spp. Cordonnier 2.5 dried fish from St. Brandon Acanthuridae Naso spp. Licorne 2.5 were not included. Scaridae Callydon spp. Cateau 2.5 Therefore, in order to Serranidae Epinephelus spp. Vieilles/Babones 5.0 assemble the most accurate Octopus 2.0 estimates for the total banks Lobster 0.5 fishery catches, we used data of the Ministry converted to wet weight, complemented or adjusted with data extracted from other reliable sources to supply the missing years and areas (Gonzalez Manero, 1971; Moal, 1971 in Paul 1987; Ardill, 1979; Paul, 1987; Samboo and Mauree, 1987; Mees, 1996; Anon., 2009a). Conversion factors of 1.2 and 2 were used to convert the frozen weight and salted-dried weight respectively, to wet weight as reported in Paul (1987), Gonzalez Manero (1971) and in the Conversion Factors FAO software. Although the Chagos Archipelago is part of the British Indian Ocean Territory (BIOT), we included catches from the Chagos‘ EEZ by the Mauritian fleet in our reconstruction of the banks fishery catches. The estimated species composition of the Mauritian banks fishery catches was based on a report by Samboo (1989 in Mees, 1996) for the Nazareth, Saya de Malha, St Brandon, Albatross and Chagos banks (Table 5), while the taxonomic breakdown of the St Brandon salt and dried fish, octopus and lobsters catches was based on Gonzalez Manero (1971) and Anon. (2009a; Table 6).  Mauritian purse seine fishery Large-scale commercial purse seining was introduced in the Indian Ocean in 1979 as a Mauritian and Japanese joint venture, using a traditional Japanese technique of fishing schools of tuna associated with logs, developed in the Pacific Ocean. The reported landings from Mauritian purse seiners reached a peak of over 10,000 t in 1991. Since then, there has been a gradual decline of this fishery until 2001 when production dropped to zero, as all the seiners left the fishery (Jehangeer, 2006). Catch data for the Mauritian purse seiners were provided by the Indian Ocean Tuna Commission (IOTC). Amande et al. (2008) provided estimates of tuna discards and bycatch by large groups of species of the European purse Table 7. Data and methods used for estimating discards of the Mauritian tuna purse seine fishery from 1979 to 2000. To calculate the discard rate (ratio to tuna production) for the bycatch species, we converted the bycatch rate expressed in tuna production ratio into a percentage, which we multiplied by the reported percentage of discarded bycatch specific to each species group (except sharks). Source: Amande et al. (2008). Species or group Bycatch/tuna production Discarded bycatch Discard rate (t/1000 t) (%) (%) Tuna nei 26.5 100 2.65 MMFa 19.7 80 1.58 Sharks 6.0 0.00 Billfishes 0.7 65 0.05 Rays 0.2 100 0.02 a MMF = miscellaneous marine fishes  seiners operating in the Indian Ocean for the 2003-2007 period. Discards and bycatch were presented as tuna production ratios, for three different fishing methods. As the Mauritian purse seiners operated exclusively on artificial logs (Norungee and Lim Shung, 1995), we used the discards and bycatch ratios corresponding to the FAD-associated fishing mode. To estimate discard rates for each group of species, we multiplied the percentage of bycatch to tuna production by the percentage of discarded bycatch (Table 7). As no indications were provided regarding the amount of discarded bycatch for shark, and to remain conservative, we assumed that sharks species were accounted for in the landings and thus we did not  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Shark fishery Shark catches were presented by the FAO for the years 1977-2008. A review of the independent literature found no information regarding a shark fishery in Mauritius. These might have been incidental catches associated with the longline fishery, but which do not appear in the IOTC data. However, we accept the FAO landings data for the ‗sharks, skates and rays‘ category as it is presented in FAO FishStat and add it to the industrial component of our total reconstructed catch.  4.0  Illegal gears Professional part-time  3.0 2.5 2.0 1.5 1.0 0.5 0.0 1950  Professional full-time  1960  1970  1980  1990  2000  Year Figure 4. Reconstructed small-scale fisheries catches for Mauritius Island. Professional full- and part-time catches include both artisanal and subsistence components.  4.0  Sea cucumber fishery  Reconstructed  Catch (t x 103 )  3.5 3.0 2.5 2.0 1.5  Reported  1.0 0.5 0.0 1950  1960  1970  1980  1990  2000  Year Figure 5. Reconstructed small-scale fisheries catches for Mauritius Island compared to reported national landings data.  4.0  Illegal gears Full-time (subsistence)  3.5 Catch (t x 103 )  On Mauritius, the commercial exploitation of sea cucumbers started on a trial basis in late 2005 (Conand, 2008) and around 2006 on Rodrigues (Dr Emily Hardman, Shoals Rodrigues, pers. comm.). Species harvested are Actinopyga echinites (brownfish), A. mauritiana (surf redfish), Bohadschia marmorata (brown sandfish), Stichopus chloronatus (green fish), S. variegatus (curry fish), Holothuria scabra (sandfish), H. nobilis (black teatfish) and Holothuria spp. (Anon., 2009d). Although fishers regularly collect sea cucumbers in Mauritius Island mainly for domestic consumption (Laxminarayana, 2005), the harvesting is done by a limited section of the Mauritian population – mainly local Chinese people – and those catches were thus thought to be negligible (Chantal Conand, pers. comm.). Current management measures of the commercial sea cucumber fishery include collection restrictions, both spatial and temporal, and size limits (Anon., 2007a) (Dr Emily Hardman, Shoals Rodrigues, pers. comm.) although illegal fishing continues to occur in the closed season on Rodrigues (Dr Emily Hardman, Shoals Rodrigues, pers. comm.). We accepted the sea cucumber landings data presented by FAO on behalf of Mauritius, although it likely that this is an underestimate of actual catches.  Subsistence only  3.5 Catch (t x 103 )  include them in our discard estimates. Through this method, our total estimated discards for the Mauritian tuna purse seiner was consistent with the discard rate of 5% reported by Kelleher (2005).  53  3.0 2.5  Part-time  2.0 1.5 1.0  Full-time (commercial)  0.5 0.0 1950  1960  1970  1980 Year  1990  2000  Figure 6. Reconstructed small-scale fisheries catches for Rodrigues Island, with commercial and subsistence catches taken by full-time fishers, catches taken by part-time fishers whose catch is for subsistence purposes only and unreported catch from illegal gears.  54  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  RESULTS  Catch (t x 103 )  Total reported landings by the FAO from 1977 to 2008 followed the same pattern as those of the Ministry reports, implying a good data transfer mechanism between the Mauritius national level and FAO. The same conclusion was drawn from the comparison of the non-pelagic reported catches from both sources. The catch rate data combined with the 4.0 fisher population data yielded the Reconstructed 3.5 reconstructed small-scale catches for 3.0 1950-2008, which are presented here for Mauritius and Rodrigues Island. Also 2.5 represented are estimated catches for the 2.0 recreational sector including both pelagic 1.5 and lagoon fisheries, the Banks fishery, 1.0 and tuna catches and associated discards Reported from the purse seine fishery. Finally, we 0.5 compare our total reconstructed catch 0.0 estimate for the Republic of Mauritius for 1950 1960 1970 1980 1990 2000 1950-2008 to total marine landings Year presented by the FAO on behalf of Figure 7. Total reconstructed catch for the small-scale Mauritius.  Mauritius Island  fisheries of Rodrigues Island, 1950-2008 compared to the nationally reported landings, which start in 1977.  Catches (t)  Reconstructed small-scale fisheries catches for Mauritius Island, including catches by professional fulltime and part-time fishers, part-time subsistence (only) fishers and unreported catches taken by illegal gears and methods was estimated to be 170,825 tonnes over the 1950-2008 time period (Figure 4). Unreported catch taken by illegal gears and methods represented approximately 8,500 t over the period 1960-2008 (Figure 4). For the 1977-2008 a) period, our total reconstructed catches for 1400 the small-scale sector are about 1.6 times Total reconstructed greater than nationally reported landings 1200 (Figure 5). Catches by part-time fishers‘ 1000 (professional and subsistence only) made 800 up a substantial part of the reconstructed 600 catch representing, for the whole time period, 27% of our total reconstructed 400 Reported catches for the small-scale sector on 200 Mauritius Island. 0  1950  1960  1970  1980  1400  1990  2000  Tourist lagoon  1200  Catches (t)  The time series of catches taken by professional full-time fishers, the majority of which are artisanal, has two main periods. The first (1950-1977) represents a period where data were scarce, the second (1977-2008) corresponds to a period of improved sampling and collection of fisheries landings. Thus, the catch variability that appears in the latter time period may be more representative of a trend in catch levels. Catches taken by professional full-time fishers decreased substantially in 1979 from approximately 2,200 t to 1,500 t the following year (1980; Figure 4). The cyclone of 1980 may have played a role in this decrease e.g., through its impact on fishing effort. However, given  b)  1000 800 600  400  Pelagic sports fishery  200 0 1950  1960  1970  1980  1990  2000  Figure 8. a) Reported and total reconstructed recreational catches for Mauritius state 1950-2008; b) Components of total recreational catch.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  55  that the sampling methods for monitoring artisanal catches were implemented in 1977, these dropping catches could also be the result of adjustments to the method by the officials between 1977 and 1980.  Rodrigues Island Catches by small-scale fishers on Rodrigues Island were estimated over the 1950-2008 time period to be approximately 159,000 t (Figure 6). This estimate included commercial catches by professional full-time fishers which amounted to about 96,000 t, subsistence catches taken by full-time and part-time fishers amounting to almost 54,000 t and unreported catch from illegal gears estimated to be 8,600 t. Subsistence catches taken by part-time fishers (i.e., people who fish only for subsistence purposes) were over 2.5 times larger than subsistence catches taken by full-time fishers as the ‗take-home‘ portion (i.e., non-commercial) of their catch (Figure 6). However, this is mainly an artifact of the much larger number of part-time fishers than professional full-time fishers. The number of part-time fishers was up to 7 fold greater than the number of professional full-time fishers. Summed for 1977-2008, reconstructed small-scale fisheries data suggested a 1.9 fold difference between reconstructed estimates and the statistics reported by the Ministry (Figure 7). As for Mauritius Island, this discrepancy between the reported and our reconstructed catches is mainly due to the inclusion of our estimated catches by part-time fishers, which, for the whole time period, represented 25% of our total reconstructed catches for the small-scale sector.  Recreational fisheries  Catch (t x 103 )  Our total reconstructed recreational catch for Mauritius from 1950 to 2008 was estimated to be over 30,000 t, which is 1.7 times larger than the reported recreational catch (Figure 8a). Pelagic sports fishery catches accounted for approximately 21,800 t and tourist catches from the 8 Chagos lagoon fishery representing the remaining 7 8,700 t (Figure 8b). It is worth noting the Albatross 6 difference in the overall trend of the reported compared to the reconstructed 5 recreational catch, especially since 1990 4 (Figure 8a). Due to the method employed, 3 estimated catches of the fishing tourist Saya de Malha 2 population reflected the growing number St Brandon of tourists visiting the island each year, 1 Nazareth thus showing an increasing trend for the 0 whole time period (Figure 8b). Estimated 1950 1960 1970 1980 1990 2000 pelagic sports fishery catches increased Year constantly from 1950 to a peak of 825 t in Figure 9. Total catches estimated for the Banks fisheries of 2000, after which catches have been Mauritius, 1950-2008. These include fisheries on St. Brandon, decreasing. Chagos, Albatross, Saya de Malha and Nazareth banks.  Banks fishery Total estimated catches from the banks fisheries were estimated to be approximately 167,000 t over the study period. The St Brandon banks fishery operated from 1950-2008, the Nazareth bank fishery started in 1969 and the Albatross, Chagos and Saya de Malha banks fisheries started in the late 1970s. The Saya de Malha and Nazareth banks fisheries were the largest with catches of over 70, 000 t and 54,000 t, respectively over the study period (Figure 9).  Mauritian purse seine fishery Total catches of tuna taken by the Mauritian purse seine fishery were estimated for the period 1977-2008 to be approximately 127,000 t (Figure 10). The discarded bycatch associated with the tuna purse seine fishery were estimated to be 4,472 t (Figure 9).  56  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Shark and sea cucumber fisheries Total catches of shark and sea cucumber, taken directly from the FAO data were estimated to be 2,043 t and 1,055 t, respectively, over the 1950-2008 time period. Shark catches were considered part of the industrial sector while the sea cucumbers originate from the artisanal sector.  Total reconstructed catches  12  Catch (t x 103 )  Total reconstructed marine fisheries catches for Mauritius and its outer dependencies was estimated for the 1950-2008 time period to be 682,392 tonnes. This suggests our reconstructed estimate was 42% higher than the landing statistics reported by FAO on behalf of Mauritius. The discrepancy between our total estimated catches and those currently reported is most substantial for 1950-1970 (Figure 11). Summed for this period, our total reconstructed estimates represent about 2.4 times the reported catches. This implies that under-reporting of catches was particularly substantial for the earlier time period.  14  10  8 6 4 2 0 1950  Tuna 1960  1970  1980  1990  2000  Year Figure 10. Total catches of tuna and associated discards from the Mauritian purse seine fishery, 1950-2008.  30  Catch (t x 103 )  25  DISCUSSION  Discards  Total reconstructed catch  20  For a better understanding of the 15 fisheries impact on marine ecosystems, there is a great need for improved 10 reporting and verification of landings and catches. The present study as supplied to FAO 5 represents an alternative approach to estimate a more comprehensive total 0 catch for the island state of Mauritius, 1950 1960 1970 1980 1990 2000 including estimates of unreported Figure 11. Total reconstructed catch for Mauritius Island and its landings of the small-scale fisheries dependencies and catches presented by the FAO on behalf of the sector, recreational catches, and Republic of Mauritius, 1950-2008. discards. Summed for 1950-2008, marine fisheries catches for Mauritius and its dependencies as estimated in our reconstruction was 682,392 t, which is 42% larger than currently reported total catches of 478,305 t presented by FAO on behalf of Mauritius. This was largely due to the under-reporting of small-scale catches for Mauritius and Rodrigues islands. Estimated small-scale catches for Mauritius and Rodrigues islands represented respectively 25% and 23% of the total reconstructed catch, thus implying an important contribution of those fisheries to total national catches. Thus, our results confirmed that currently reported catches for the Republic of Mauritius are incomplete, and especially underrepresented catches from the small-scale fishing sectors. This situation of underreporting of catches for the small-scale sector is not specific to Mauritius and has been demonstrated for other countries (Zeller et al., 2006; Jacquet and Zeller, 2007). Indeed, although they are highly important in terms of income and food security, small-scale tropical fisheries are often marginalized world-wide (Pauly, 1997). Consequently, their contribution in terms of catches is also often substantially underreported.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  57  Nevertheless, it should be mentioned that the Mauritius government made efforts to decrease the pressure on lagoon fish stocks and to improve the fishers‘ livelihoods (such as the bad weather or closed net season allowances). Mauritius is also one of the few countries which accounts for recreational and ‗amateurs‘ fisheries catches in their total reported catches, even though they are represented by constant estimates over the years. However, as demonstrated in the present study, more efforts should be dedicated to monitoring recreational and small-scale fisheries. High levels of uncertainty are associated with reconstructions such as ours. However, as long as estimates for unaccounted catches are not substantially overestimated, the catch reconstruction will present a more accurate picture of total extractions in the marine environment compared to current practices of essentially allocating ‗zero catch‘ to IUU components (Illegal, Unregulated and Unreported catches) for which no hard time series data are available. The time series data presented do not include catches by foreign vessels which fish heavily – legally or illegally - the waters off Mauritius, which would likely add great amounts to the total extractions of marine resources in the Mauritius‘ EEZ. The Banks of the Mascarenes Ridge, for instance, have been, over the years, exploited by a number of foreign fishing units from diverse countries such as Korea, Japan, France (Reunion), Seychelles, Spain, Russia, Panama and Malaysia (Paul, 1987). Réunion based boats were fishing in the Soudan, Nazareth and Saya de Malha zone as early as 1962. Not all foreign vessels fishing on the banks use handlines but also longlines and trawls. The total catch of these foreign fishing vessels on the banks is not reported but estimated to be well in excess of 10,000 tonnes per year (Ardill, 1986; Paul, 1987). Discards of the deep-sea demersal trawlers are also considered to be high in many deep-sea fisheries (Kelleher, 2005) and therefore they would likely add important amounts to our estimated total catches. Kelleher (2005) reported a discarding rate for deep water trawls of 39.6%. However, this discarding rate was from fisheries operating in the Northeast Atlantic and Chile, and no similar estimate has been made for the Western Indian Ocean. Therefore, attempts have not been made to quantify trawl discards. In Mauritius as in Rodrigues Island, depletion of the marine lagoon resources is of concern and indications of growth overfishing were drawn early. One of the characteristics of growth overfishing is a decline in the mean size of a fish population, and this problem has been noted for Mauritius as early as the 1920s (Paul, 1987). Pearson (1988) also defined the lagoon of Rodrigues as clearly overfished, while (Bunce et al., 2008) mentioned that overfishing in Rodrigues could have occurred since the 1800s. In Mauritius, the reduced catches have to some extent been compensated for by rising prices and a high demand for fresh fish, and as a result there has been no substantial reduction of the fishing effort (Hollup, 2000). An ever growing population and increasing tourist arrivals is leading to a greater demand for seafood, and subsequent increased pressure on fish stocks. Fishers have also resorted to illegal fishing practices that are highly destructive for the marine environment (i.e., dynamite and fine mesh nets). Combined, population pressure and destructive fishing practices suggest that ‗Malthusian overfishing‘ (Pauly, 1997) is occurring on Mauritius Island. Although regulations have existed in the legislation since the colonial days, they only concern limitations on the use of specific gears, fish reserves and closed seasons for nets (Hollup, 2000). Pearson (1988) and Hollup (2000) both mentioned that regulations should include the limitation of access to the fish resources of the lagoon area. However, alternatives are needed for the numerous fishers who depend on these resources for their livelihoods. This study illustrates the need for better reporting of catches for all the different fisheries sectors of Mauritius state. Accounting for all fisheries components is fundamental to effectively managing the fisheries, which provide food and income for a large coastal population. We hope that the present study will encourage management agencies and policy makers to work in this direction.  ACKNOWLEDGMENTS We thank Dr Emily Hardman, Reshad Jhangeer-Khan from the Shoals Rodrigues, the Fisheries Research and Training Unit of Rodrigues and the Fisheries Division of Mauritius, who shared information and  58  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  knowledge for this study. We acknowledge Pierre Chavance, Miguel Herrera, and Vincent Defaux, for their help and contribution, as well as the support of the Sea Around Us Project, a scientific collaboration between the University of British Columbia and the Pew Environment Group.  REFERENCES Amande, J.M., Ariz J., Chassot, E., Chayance, P., Delago de Monlina, A., D., G., Murua, H., Pianet, R. and Ruiz, J. (2008) By-Catch and discards of the European purse seine tuna fishery in the Indian Ocean. Estimitation and Characteristics for the 2003-2007 period. Indian Ocean Tuna Commission, Bangkon, Thailand. Anon. (1987) Reports of the ad-hoc technical Commitee on fisheries research. Food and Agricultural Research Council (FARC), Maurice. Anon. (2006) Population and Vital Statistics Republic of Mauritius, Year 2005. Central Statistics Office, Republic of Maurititus, Port Louis. Anon. (2007a) Annual report 2007 Fisheries Division, Ministry of Agroculture Industry, and Food Security, 106 p. Anon. (2008) Digest of international travel 2007. Central Statistics office, Republic of Maritius, Port Louis. Anon. (2009a) Digest of Environmental statistics 2008. Central Statistics Office, Republic of Maritius, Port Louis. Anon. (2009b) Digest of International Travel. Central Statistics Office, Republic of Maritius, Port Louis. Anon. (2009c) Digest of statistics on Rodrigues. Central Statistics Office, Repbulic of Maritius, Port Louis. Anon. (2009d) Transboundary Diagnostic Analysis of Land-based Sources and Activities Affecting the western Indian Ocean Coastal and Marine Environment. 378. United Nations Environmental Programme, Nairobi Convention Secretariat, Nairobi. Anon. (2010) Population and Vital Statistics, Republic of Mauritius, Year 2009. Central Statistics Office, Republic of Maritius. Ardill, J.D. (1979) Country Statement on the marine fisheries in Mauritius. pp. 54-62. In: Organization, U.N.D.P.F.a.A. (ed.) Report of the FAO/ IOP workshop on the fishery resources of the western Indian Ocean south of the Equator. United Nations Development Programme, Mahe, Seychelles. Ardill, J.D. (1986) Current status of fisheries in Mauritius. Food and Agriculture Organization, United Nations, 31 p. Bunce, M., Rodwell, L.D., Gibb, R. and Mee, L. (2008) Shifting Baselines in Fishers' perceptions of island reef fishery degredation. Ocean and Coastal Management 51(4): 285-302. Christy, L. and Greboval, D. (1985) Fisheries Institutions in Mauritius. SWIOP Documents. Regional Office for Africa, Food and Agriculture Organization of the United Nations, Mahe. Conand, C. (2008) Population status, fisheries and trade of sea cucumbers in Africa and the Indian Ocean. pp. 143193. In: Toral-Granda, V., Lovatelli, A. and Vasconcellos, M. (eds.) Sea cucumbers: A global review of Fisheries and trade. FAO Fisheries and Aquaculture Technical papers. Food and Agrigulture Organization of the United Nations, Rome. Gabbay, R. (1988) Chapter 9: Tourism. p. 236. In: Appleyard, R.T. and Ghosh, R.N. (eds.) Indian Ocean Islands Development. 1. National Center for Development Studies, Australian National University, Canberra. Gonzalez Manero, C. (1971) Marketing of fishery products. Fishery Developmenment Project. United Nations Development Programme, Rome, 37 p. Hardman, E.R., Biais, F.E.I., Desire, M.S., Raffin, J.S.J., Perrine, S. and Gell, F.R. (2007) Marine Reserves for Sustainable Fisheries Managment in Rodrigues. Alternative Livelihood Options 1. Shoals Rodrigues & Wildlife and Conservation Division, Department of Agriculture, Fisheries and Forestry, Peel. Harris, A. (1988) Indian Ocean Islands Development. National Center for Development Studies, Australia National University, Canberra, 204 p. Hollup, O. (2000) Structrual and Sociocultural constraints for user-group participation in fisheries management in Mauritius. Marine Policy 24(5): 407-421. Houbert, J. (2009) Mauritius: a sea food hub? In: Rumley, D., Chaturveydi, S. and Vijay, S. (eds.) Fisheries Exploitation in the Indian Ocean. Threats and opportunities. ISEAS Institute of Southeast Asian Studies. Jacquet, J.L. and Zeller, D. (2007) National conflict and fisheries: reconstructing marine fisheries catches for Mozambique. Fisheries Center Research Reports. Fisheries Centre, University of British Columbia, Vancouver. Jehangeer, I. (2006) Review of the state of world marine capture fisheries management: Indian Ocean. Country Review: Mauritius. FAO Technical Paper. Food and Agricultural Organization United Nations, Rome, 393413 p. Kelleher, K. (2005) Discards in the World's marine fisheries. An update. Food and Agricultural Organization United Nations, Rome, 131 p. Laxminarayana, A. (2005) Induced spawning and larval rearing of the sea cucumbers, Bohadschia marmorata and Holothuria atra in Mauritius. Information Bulletin 22. Secretariat of the Pacific Community. Mees, C. (1996) The Mauritian Bank Fishery. A review and spatial analysis, Technical Report. Available at: www.dfid.gov.uk/r4d/pdf/outputs/R5484c.pdf [accessed August 2011], 62 p. Moal, R.A. (1971) Storage, processing, and distribution of fish. Indian Ocean Programme. Indian Ocean Fishery Commission, Rome.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  59  Norungee, D. and Lim Shung, C. (1995) Analysis of the purse seine fishery of Mauritius, 1990-1994, and comparison of catch rate and species composition of catches of Mauritian purse seiners to those of the French fleet (English). Expert Consultation on Indian Ocean Tunas, Colombo (Sri Lanka). Paul, E.C. (1987) Fisheries development and the food needs of Mauritius. A. A. Balkema, Rotterdam. 216 p. Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. and Torres, F. (1998) Fishing down marine foods webs. Science 279(5352): 860-863. Pearson, M.P. (1988) Rodrigues. Rapid Survey of the status of exploitation and environmental damage of the lagoon and coral reefs off Rodrigues. Report prepared for the project assistance to artisanal fisherment and development of outer-reer fishery. . Food and Agriculture Organization, United Nations, Rome. Roullot, J., Venkatasami, A. and Soondron, S. (1988) The first three years experience in the use of Fish Aggregating Devices in Mauritius. Food and Agriculture Organization, United Nations, 45 p. Sauer, W.H.H., Potts, W., Raberinary, D., Anderson, J. and Sylvio Perrine, M.J. (2011) Assesssment of the current data for the octopus resource in Rodrigues, Western Indian Ocean. African Journal of Marine Sciences 33(1): 181-187. Sobhee, K.S. (2004) Economic Development, Income Inequality and Environmental Degradation of Fisheries Resources in Mauritius. Environmental Managment 34: 105-157. Sobhee, K.S. (2006) Fisheries biodiversity conservation and sustainable tourism in Mauritius. Ocean & Coastal Management 34: 150-157. Turner, J. and Klaus, R. (2005) Coral reefs of the Mascarenes, Western Indian Ocean. Royal Society 363: 229-250. Van der Elst, R. (2005) Fish, fishers and fisheries of the Western Indian Ocean: their diversity and status. A preliminary assessment. Royal Society 363: 263-284. Venkatasami, A. and Sheik Mamode, A. (1995) Fish- aggregating devices (FADs) as a tool to enhance production of artisanal fishermen: problems and perspectives. 5 p. Zeller, D., Craig, P. and Pauly, D. (2006) Reconstruction of coral reef fisheries catches in America Samoa, 1950- 2002. Coral Reefs 25: 144-152. Zeller, D., Booth, S., Davis, G. and Pauly, D. (2007) Re-estimation of small-scale fishery catches for U.S. flagassociated island areas in the western Pacific: the last 50 years. Fisheries Bulletin, 105: 266-277.  60  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Appendix Table A1: FAO landings vs. total reconstructed catch (in tonnes) for Mauritius, 1950-2008. Years 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  FAO landings 2,000 2,000 2,000 2,000 2,000 2,000 2,200 2,200 2,200 2,500 2,501 2,501 2,801 2,801 2,801 3,000 3,001 3,000 3,300 4,001 5,400 5,200 6,600 6,400 7,679 7,038 6,660 7,667 7,108 6,525 6,348 7,132 9,780 9,434 10,346 12,175 12,848 17,279 17,116 16,896 14,098 18,576 18,861 20,576 18,145 16,395 11,870 14,025 12,093 12,205 9,615 10,986 10,706 10,968 9,971 9,855 8,681 8,087 6,152  Total reconstructed catch 5,786 5,982 5,913 6,057 5,958 5,995 6,013 5,920 6,121 6,300 6,486 6,757 6,263 6,783 6,238 6,274 6,343 6,258 6,421 7,328 7,586 8,337 9,347 8,663 9,536 9,248 8,649 10,932 10,697 9,362 9,109 10,032 12,638 12,222 13,613 15,248 15,817 20,656 20,604 20,530 17,603 22,298 22,687 24,595 21,911 19,620 15,104 17,707 15,068 16,264 13,684 14,748 14,414 14,836 13,126 12,995 13,653 12,627 11,430  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  61  Appendix Table A2: Total reconstructed catch (in tonnes) by major taxa for Mauritius, 1950-2008. Others grouping includes 38 taxa. Year  Lethrinus mahsena  1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  332 460 367 449 329 322 318 220 339 257 180 226 158 218 181 215 218 128 177 895 1,027 1,568 1,884 1,520 2,344 2,083 1,506 3,460 3,333 2,055 1,553 1,589 3,505 2,684 2,680 4,212 5,063 5,917 5,550 4,085 3,360 3,716 5,152 6,047 5,961 5,720 4,787 4,737 4,091 4,328 4,208 3,627 3,936 3,816 3,215 2,240 3,088 2,491 2,449  Octopodidae  Katsuwonus pelamis  1,544 1,563 1,582 1,602 1,621 1,640 1,649 1,656 1,762 1,813 1,864 1,907 1,840 1,925 1,844 1,883 1,896 1,909 1,922 1,937 1,969 1,986 2,229 2,079 2,090 2,088 2,077 2,072 2,043 2,042 1,926 1,923 1,806 1,842 1,823 1,798 1,773 1,785 1,747 1,749 1,758 1,744 1,766 1,736 1,742 1,614 1,504 1,503 1,209 1,037 1,042 791 964 1,221 853 792 752 622 643  0 15 41 1,004 1,746 2,430 1,421 2,537 2,080 1,899 4,397 5,049 5,614 4,195 6,735 6,126 7,074 5,209 3,936 1,589 3,150 1,692 2,481 425 140 140 140 140 133 133 133 133  Thunnus albacares  0 0 0 0 1 1 2 3 4 5 6 7 8 10 10 14 18 22 27 31 34 37 39 28 30 39 46 71 59 67 76 78 1,191 1,430 1,081 1,039 1,845 1,492 1,955 1,624 3,043 2,419 2,768 1,987 1,891 727 1,281 1,576 1,333 835 769 752 784 786 787 838 823 841  Siganidae  534 539 538 542 539 540 541 538 531 574 618 653 572 647 551 607 609 608 612 613 614 617 362 362 363 365 366 368 360 350 277 274 299 342 349 348 352 383 377 379 390 382 415 394 394 372 390 338 333 336 349 319 348 332 313 299 300 246 260  Thunnus alalunga  0 1 2 3 5 9 12 17 22 28 35 42 50 58 63 87 110 136 163 189 208 224 236 172 183 235 270 296 362 336 367 368 361 373 360 359 443 440 476 480 472 484 522 536 544 601 640 648 685 725 716 693 779 876 853 838 866 788  Serranidae  152 155 153 155 153 154 154 152 170 183 196 208 182 206 175 167 168 167 168 202 209 233 243 229 266 257 233 333 304 253 192 217 351 304 332 402 426 519 505 397 408 429 509 500 543 483 491 459 368 391 455 376 420 409 344 258 342 293 252  Acanthuridae  261 262 263 264 264 265 265 265 266 289 312 329 289 327 278 305 306 307 308 309 310 311 507 510 511 514 518 523 500 498 389 387 163 185 186 184 184 203 194 199 206 200 216 201 208 191 204 175 174 174 184 167 184 175 168 158 161 132 134  Mugilidae  Others  187 188 189 191 191 192 192 192 157 165 174 181 166 181 163 149 150 150 151 152 154 155 208 202 203 203 203 204 199 197 169 168 249 275 276 273 273 293 284 288 295 289 308 292 297 303 277 242 253 283 302 275 309 267 252 216 235 215 225  2,465 2,503 2,511 2,543 2,547 2,566 2,575 2,571 2,566 2,682 2,799 2,903 2,698 2,912 2,668 2,565 2,584 2,590 2,614 2,699 2,749 2,826 3,268 3,118 3,205 3,187 3,149 3,346 3,283 3,183 2,851 2,907 3,079 3,344 3,377 4,081 4,003 4,439 4,551 5,007 4,522 4,940 4,969 4,765 4,755 4,313 4,297 4,966 4,550 5,060 4,960 7,393 6,529 6,808 6,099 7,184 6,862 6,758 5,705  62  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  63  RECONSTRUCTION OF NAURU‘S FISHERIES CATCHES: 1950-20081 Pablo Trujillo, Sarah Harper and Dirk Zeller Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC, V6T 1Z4, Canada p.trujillo@fisheries.ubc.ca; s.harper@fisheries.ubc.ca; d.zeller@fisheries.ubc.ca  ABSTRACT Nauru is a small, single island country located in the western Pacific Ocean with a relatively large population. Historically, phosphate mining has destroyed much of the island‘s land surface. We reconstructed total fisheries catches for Nauru (1950-2008). The reconstructed catch consists of smallscale fisheries, including both commercial and subsistence components, as well as the offshore domestic catch. For 1950-2008, total reconstructed catches were 23,150 t, being 3.4 times higher than data supplied to FAO on behalf of Nauru. Of these catches, 9,000 t, 12,300 t and 1,850 t were small-scale commercial, small-scale subsistence and offshore catches, respectively.  INTRODUCTION The Republic of Nauru is a small, single raised limestone island located at 0˚ 32' S latitude and 166˚ 55' E longitude, with a land area of approximately 21 km2 (Figure 1) and an Exclusive Economic Zone (EEZ) of 308,000 km2 (www.seaaroundus.org). The island has an average height of 50 m above sea level, with an interior plateau that once held extensive deposits of phosphate bearing rock, resulting from the accumulation of seabird droppings over millenia. Phosphate mining was the island‘s largest source of revenue, but due to heavy mining the resource is now virtually depleted. This has left an estimated 80% of the land area uninhabitable, along with substantial Figure 1. Map of Nauru and its Exclusive Economic Zone (solid environmental degradation from silt and line). phosphate runoff, which is believed to have impacted large parts of the island‘s marine life (Jacob, 1998). With an already degraded terrestrial environment and an eroding coastline, the impact of global sea level rise for a small island such as Nauru may be devastating (Stephen, 2011) The original inhabitants of the island relied heavily on marine resources a source of animal protein. Nauru is surrounded by a coral belt that becomes exposed at low tide, ranging from 150 to 300 m in width, and the waters surrounding Nauru hold an abundance of both reef and pelagic fish species (Dalzell and Debao, 1994; Jacob, 1998). 1  Cite as: Trujillo, P., Harper, S. and Zeller, D. (2011) Reconstruction of Naurru‘s fisheries catches: 1950-2008. pp. 63-71. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727].  64  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  The colonial period In 1888, Germany annexed the island as part of the Marshall Islands Protectorate. At the turn of the century, a British company discovered phosphate, and mining for the deposits began in 1906. Immigration to the island, starting with the European influx during the colonial period (between 1889 and 1913), followed by migrant workers to labour in the phosphate mining operation, changed the island‘s traditional make-up. These immigration patterns were linked to the mining developments of the Nauru Phosphate Corporation (NPC), yet were also relevant to the island‘s fishing capacity and history. Initially, workers were brought to the island from other German-administered Micronesian islands, as well as New Guinea and China. Later, when Australia took control of the island in 1914, an increasing number of Chinese labourers were hired, and by 1939 their numbers (1,512) nearly equalled those of the native Nauruan population (1,733). This demographic make-up did not change until, during World War II, the Japanese occupiers of the island forcibly exiled Nauruans to Truk, where nearly 500 of them died (Underwood, 1989). At the start of the 1950s, virtually every able-bodied Nauruan adult male was gainfully employed both and once again began to receive their respective phosphate mining royalty payments. This enabled the inhabitants to purchase imported foods as well as improving their living conditions (Viviani, 1970). This purchasing power brought on numerous changes in the structure of the island population, differentiating it from other Pacific islands.  Independence At the time of its independence in 1966, Nauru had attained an economic status similar to that of rich oilproducing countries of the Middle East (Underwood, 1989; Vunisea et al., 2008). This in turn increased the immigration of labourers to the island to compensate for the growing number of retiring Nauruan nationals (Figure 2). The population is largely concentrated along a narrow coastal strip. Most of the nonnative Pacific islanders (largely from Kiribati and Tuvalu) that worked for the Nauru Phosphate Corporation (NPC) were also at least part-time fishers (Underwood, 1989; Dalzell and Debao, 1994). Nauru‘s economic prosperity translated into profound cultural changes. This was also reflected in the diet, not only in terms of dietary preferences, but also in the means by which dietary goods were acquired. Until the mid 1980s, the majority of fishing was done by the non-Nauruan Pacific islanders. Nonetheless, the economic decline that occurred in the early 1990s, following the downscaling of phosphate production, reduced the population‘s ability to purchase high-valued fish, and contributed to the subsequent emigration of non native fishers from the island. The development and management of the marine resources within the Republic of Nauru falls under the jurisdiction of the Nauru Fisheries and Marine Resources Authority (NFMRA). The NFMRA does not enforce the reporting of catches or issue any fishing quotas. Marine resources are open access, and records of catches are sparse (Dalzell and Debao, 1994). The NFMRA attempted to pursue industrial-scale fisheries by purchasing two purse-seine vessels. However, one ship was lost at sea, while the other was used sparsely due to inadequate fishing gear, and was subsequently sold. The aim of the present study was to gather available information on fisheries catches and fishing practices to reconstruct Nauru‘s total fisheries catches for the period 1950-2008. The catch reconstruction approach used here is based on the approach developed by Zeller et al. (2006; 2007).  METHODS At present, small scale marine fisheries in Nauru can be separated into two categories: 1) Subsistence fisheries, dominated by coastal reef fisheries, beach seining and reef gleaning (mainly by women [Chapman, 1987]), are known to be traditional fishing methods practiced throughout the  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  65  Pacific islands (Gillett, 2003). I-Kiribati, Tuvaluan and Nauruan fishers commonly operate outboard powered boats ranging from 3–7 m in size, using trolling and shallow-water bottom handlining as their main fishing methods. Some Nauruans carry out spear fishing equipped with scuba gear. This type of fishing may also be done at night, using battery powered flashlights (Chapman, 1998). These coastal fishing activities do overlap with commercial fisheries, if part of the catch is sold; and 2) Commercial fisheries, using fishing methods such as mid-water handlining and drop stone fishing, targeting offshore tunas and other pelagics (Gillett, 2003). The fishers involved in this type of fishery are generally Nauru Phosphate Corporation (NPC) workers from Tuvalu and Kiribati who fish whenoff from work. This fishing is often concentrated around the NPC‘s mooring buoys used for phosphate vessels that act as Fish Aggregation Devices (FADs) (Chapman, 1998).  Data sources 12  Human population numbers (1950-2008) were derived with assistance from G. Beccalossi; Programme Assistant at the Demography and Statistics division of the Secretariat of the Pacific Community (SPC), complemented with data from UN databases2 and work by Underwood (1989). Every decade had at least one demographic composition anchor point, with linear interpolations performed between anchor points (Figure 2).  10  Catch data  Population (103 )  Human population  Non Pacific islanders Pacific islanders  8 6 4  Nauruans  2 0 1950  1960  1970  1980 Year  1990  2000  Figure 2. Population composition of Nauru, 1 950-2008.  Data estimates of fisheries catches, and used here as anchor points, were obtained from Dalzell and Debao (1994), Gillett and Lightfoot (2002), Vunisea et al. (2008), as well as Bell et al. (2009), for the period between 1991 and 2008. Linear interpolations were used for time periods between anchor points. To reconstruct the 1950-1990 times series, we assumed that the per capita consumption rate for 1950 was twice the 1991 per capita catch rate of 45 kg∙person-1∙year-1 for Nauru, taken from Dalzell and Debao (1994). Hence, we assumed a consumption rate of 90 kg∙person-1∙year-1 for 1950. We also assumed negligible seafood imports for 1950. Nauruans, like Table 1. Fishing sector estimates for late 1990s, used as most Pacific islanders, have traditionally relied on anchor points, from Gillett and Lightfoot (2002). fish as a main protein source (Petit-Skinner, 1981; Fishing sector Catch (t) Sokimi and Chapman, 2001) and the assumed 1950 Coastal subsistence 110 rate is in line with other data for Micronesia (Bell et Coastal commercial 315 al., 2009). We linearly interpolated rates between Offshore locally based 50 1950 and 1991 and derive total catch estimates in Offshore foreign based 41,000 conjunction with population data (Figure 2). Total 41,475 For the 1990s, Gillett and Lightfoot (2002) estimated offshore and domestic commercial landings (accounting for approximately 77% of domestic supply) as well as subsistence catches (23%; Table 1). In addition, they documented offshore pelagic catches of around 41,000 t, taken by foreign vessels in the late 1990s. This breakdown by fishing sector provides a proxy for local per capita catch rates as well as domestically sourced consumption rates. Domestic fish landings were estimated based on coastal catch  2  United Nations Department of Economic and Social Affairs Population Division: http://www.un.org/esa/population/ [Accessed: February 2010]  66  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  estimates by Gillett and Lightfoot (2002) (425 t) summed with domestic offshore catch (50 t) for a total of 475 t (Table 1).  Catch (t)  The economic downturn of 1999 greatly diminished the population‘s purchasing power, decreasing their ability to pay for commercially sold fish. This is reflected in the increasing proportion of fish sourced via the subsistence sector which increased to 84% of the total domestic landings by 2004 (2008). In addition, a reduction in imported fish (22 t) was reported by Gillett and Lightfoot (2002) for 1999, which is considerably lower than the 55 t of imported canned and salted fish reported a few years earlier by Dalzell and Debao (1994). Nonetheless, a two month survey of the island‘s fishers and families in 2005 estimated that canned fish consumption had risen to an average of 16 kg∙person-1∙year-1 (Vunisea et al., 2008). Because Bell et al. (2009) report 600 that Nauruan fish consumption a) As submitted to FAO rates range between 55.8 and 62.3 500 kg∙person-1∙year-1 for the period between 2001 and 2006, we chose 400 the latter rate to account for the 300 canned fish consumption, along Subsistence with the 425 t of small scale catch 200 reported by Gillett and Lightfoot (2002), which translates into a 100 Commercial demand of approximately 600 t of fish per year for Nauru in more 0 recent years. 1950 1960 1970 1980 1990 2000  Catch composition  Year Figure 3a. Reconstructed catches breakdown for Nauru, compared with that reported to FAO from 1950 to 2008 .  Catch (t)  Dalzell and Debao (1994) list over 180 species of fish observed and 600 reported in the waters of Nauru. Other taxa They also report a catch breakdown b) 500 by gear type collected over an eight Lutjanidae month period between July 1992 400 and February 1993 where at the time, approximately 75% of the 300 Carangidae fishing was done by migrant workers and where approximately 200 70% of the total landings were Scombridae 100 commercially sold. Approximately 60% of the total landings were 0 taken by trolling gear, highlighting a 1950 1960 1970 1980 1990 2000 high prevalence of tunas (41% of total catch), mainly skipjack Year (Katsuwonus pelamis) and Figure 3b. Taxa layout for Nauru’s landings for the period between 1950-2008. yellowfin (Thunnus albacares). Scombridae species 56% of total catch by weight with substantial catches from Another 30% of the landings were the Carangidae (31%) and Lutjanidae (6%) families. Within the Scombridae caught using mid-water handlines, family 90% of the catch were comprised of Skipjack (katsuwonus pelamis) and almost exclusively catching Yellowfin tuna (Thunnus albacares) Carangidae such as rainbow runner (Elagatis bipinnulata) and using demersal handlines, that target various species such as squirrelfish (Holocentridae) and bluestripe snapper (Lutjanus kasmira). The Mid-water and demersal handline fisheries comprise 26% and 7% of the total landings, respectively. The remaining 7% corresponds to inshore reef fishing activities that are regarded as subsistence or semi-artisanal (Vunisea et al., 2008), where catches were composed predominantly of surgeonfish (Acanthuridae, 38.5%), squirrelfish (Holocentridae, 12.1%), groupers (Serranidae, 7.7%) and rainbow runners (Carangidae, 5%). More recently, lesser valued reef fish such as surgeonfish and triggerfish as well as many invertebrates found during beach and reef gleaning (octopus, turban shell and sea cucumbers etc.), account for an increasing  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  67  proportion of inshore subsistence catches (Vunisea et al., 2008). Those species were not necessarily favoured by the Nauruan locals a decade earlier (Dalzell and Debao, 1994; Gillett and Lightfoot, 2002). Presently, fishers indicated that most of their target species required further distances and hours at sea to catch, but regardless of where it came from, it was intended for their family consumption. Incidentally, all family members participated in one way or another in fishing activities (Vunisea et al., 2008; Gillett, 2009) and the increasingly important reef gleaning activity is generally undertaken mainly by women (Chapman, 1987). Overall, the more detailed breakdown by Dalzell and Debao (1994) was applied to the total catch throughout the 1950-2008 time period.  RESULTS Data supplied to FAO for Nauru would suggest incorrectly that essentially no fish were caught prior to the early 1960s (Figure 3a). Reported landings increased steadily from 100 t∙year-1 in 1963 to around 190 t∙year-1 in 1991, before increasing substantially to around 500 t∙year -1 in 1993 and 1994 (Figure 3a). Thereafter, data supplied to FAO suggested a rapid decline throughout the 1990s to around 40 t∙year -1 in the early 2000s. Reconstructed catch estimates suggested a distinctly different picture (Figure 3a). Overall, the 1950-2008 total catches were estimated at 23,150 t, being 3.4 times greater than the data supplied to FAO on behalf of Nauru. Note the slightly lower total estimates for 1993 and 1994 compared to data supplied to FAO. Reconstructed commercial catches increased from around 30 t∙year -1 in 1950 to around 360 t∙year-1 in 1999, before declining rapidly to just under 70 t∙year-1 by 2008 (Figure 3a). Conversely, reconstruction suggested a decline in subsistence catches from 290 t∙year-1 in 1950 to a low of around 110 t∙year-1 in 1991, before increasing rapidly to over 350 t∙year-1 by 2008 (Figure 3a). The taxonomic breakdown applied to Nauru‘s reconstructed fisheries catches (Figure 3b) is based on the reported catch composition given by Dalzell and Debao (1994) for the commercial catches, and on Vunisea et al. (2008) for the subsistence sector. Commercial catches were dominated by Scombridae (56%), composed of 80% skipjack (Katsuwonus pelamis) 10% yellowfin tuna (Thunnus albacares) and 10% other scombrids, followed by Carangidae (30%, mainly rainbow runner, Elagatis bipinnulata) as well as Lutjanidae (6%, mainly Lutjanus kasmiri), followed by coastal reef species such as surgeonfish (5%) and triggerfish (3%). Other taxa have been increasingly targeted more recently, due to ever more intensive fishing pressure. Overall, there is a predominance of pelagic species in local landings (Figure 3b). Nonetheless, far less of these landings are sold commercially; rather, they are intended for family consumption or sharing with other family members and neighbours. This shift to increasingly noncommercial basis was driven by the recent economical crisis and the weaker financial power of the Nauruan people (Vunisea et al., 2008; Gillett, 2009).  DISCUSSION Nauru‘s total domestic fisheries catches for the period 1950-2008 were estimated to be 23,150 t. This amount illustrates the historical importance fisheries have in meeting the island‘s dietary requirements, a fact which cannot be readily inferred from the data reported to FAO on behalf of Nauru. Small-scale fisheries are fundamental to many Pacific islands, nonetheless widespread lack of information on subsistence sector catches undervalues the social and economical importance of this sector (Zeller et al., 2006; Gillett, 2009) and may impact any successful form of ecosystem-based fisheries management (Pauly et al., 2002). The landings reported to the global community on behalf of Nauru substantially underestimate total catches as estimated here for all but two years during the early 1990s. We assume that this peak in reported landings coincides with the publication of the seminal work by Dalzell and Debao (1994) and Dalzell et al. (1996), who presented estimates of per capita catch rates. These studies were likely used to estimate Nauru‘s 1993 and 1994 fisheries landings that were reported to FAO. However, the human population counts for that decade were overestimated (Underwood, 1989), likely resulting in the higher reported landings estimates as presented by FAO compared to our reconstructed estimates. We identified  68  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  that for the early 1950s, population statistics were underestimated by as much as 60% (Underwood, 1989) and in the most recent decade, the population was overestimated by 20% to 40%, due to the rapidly increasing emigration of foreign NPC workers (Vunisea et al., 2008). Hence it is noteworthy to mention that the per capita fish consumption of 56 kg∙person-1∙year-1 reported by Bell et al. (2009), which we used for recent time periods, is likely underestimated, due to it being based on inflated population estimates. Despite the economic surge provided by the island‘s phosphate earnings during earlier decades, Nauru‘s fisheries did not develop in similar fashion, due in part to the absence of natural harbours to moor vessels. Three man-made channels and a small boat harbour have been excavated through the coral fringing reef, allowing only small outboard powered vessels to be launched. Industrial-scale fishing was attempted in the early 1980s involving two purse-seine vessels, but was not successful due to ineffective gear and through market loss which led to the sale of one vessel while the other was lost at sea (Chapman, 1998). Nauru has been a party to the US Multilateral Fisheries Treaty since it came into effect in June 1988. In July of 1994, Nauru entered into a bilateral fishing access agreement with Japan, granting four Japanese fishing vessels access to Nauruan waters and in June 1997, the first fishing access agreement was signed with the Philippines (Chapman, 1998). It is estimated that Nauru received about US$3.4 million in access fees in 1999 (Gillett and Lightfoot, 2002) and US$5.4 million in access fees in 2008 (Gillett, 2009), with 131 foreign fishing vessels (10 countries), licensed to fish in Nauru‘s EEZ, catching approximately 66,000 t of tuna. Attempts to farm milkfish (Chanos chanos) and tilapia (Oreochromis mossambicus) in the past had failed; these fish were introduced into the Baduan Lagoon. Milkfish, although part of the traditional Nauruan diet and culture, were used as bait for tuna fishing (Spennemann, 2002). Tilapia was introduced in the early 1960s but was not accepted as a staple food option and eventually infested all the milkfish ponds causing many farmers to abandon their traditional practice of raising milkfish (Gillett, 2009). Programs to eradicate the introduced fish from the island‘s lagoon have failed (Dalzell et al., 1996). In 2000, 10,000 milkfish fry from Kiribati were introduced into Buada Lagoon, reaping 5,000 adult fish (Gillett, 2009). At present, several milkfish grow-out ponds exist; these are backyard and mostly subsistence operations. No accurate production estimates exist (Gillett, 2009). Inshore fishing pressure appears to have increased dramatically since the late 1990s, with almost all households involved in fishing; women and children glean the beaches and reefs, collecting all invertebrate and finfish species they come across. Vunisea et al. (2008) surveyed invertebrate catches, estimating the total catch to be 23 t·year-1, dominated by genera such as Etisus, Octopus, Turbo, Thais, Tripneustes and Cardisoma and to a lesser extent Actinopyga, Panulirus, Grapsus, and Cypraea. All species and types are targeted and consumed, with the exception of lobsters, which are destined for sale (Vunisea et al., 2008). All sizes of fish are caught and consumed, however fishers have observed a decrease in size and volume of catches (Vunisea et al., 2008), suggesting that overfishing is occurring. There has been a steady increase in the intensity and frequency of fishing since the economic crisis in 1999, driven mainly by increasing subsistence efforts. Presently, pelagic fishing is dominated by canoes operated by Tuvaluans and I-Kiribati (Vunisea et al., 2008). Fish continue to form a large part of the Nauruan diet, increasingly so with the island‘s recent economic downturn. Food security may be jeopardized, leading to substantial dietary changes. Imported produce such as meat and poultry have been replaced with canned sardines and mackerel from abroad, seafood provides the main source of protein for more than 98% of Nauruan households (Vunisea et al., 2008). The Nauru Fisheries and Marine Resource Authority (NFMRA) have the responsibility of overseeing, managing and conserving the country‘s natural marine resources and environment. Yet, with fishing being the only major fallback option for the population, the task presents several challenges, especially when marine resources are vulnerable to overexploitation and the livelihoods of an entire nation are at risk.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  69  ACKNOWLEDGEMENTS We thank Ms. Gladys Beccalossi (SPC) for her assistance with population estimates and the Sea Around Us Project, a scientific collaboration between the University of British Columbia and the Pew Environment Group.  REFERENCES Bell, J., Kronen, M., Vunisea, A., Nash, W., Keeble, G., Demmke, A., Pontifex, S. and Andréfouët, S. (2009) Planning the use of fish for food security in the Pacific. Marine Policy 33: 64-76. Chapman, L. (1998) Feasibility study on infrastructure requirements and vessel parameters for tuna longlining in Nauru (17–22 November 1997). Noumea, New Caledonia 25 p. Chapman, M. (1987) Women's Fishing in Oceania. Human Ecology 15(3): 267-288.Dalzell, P., Adams, T. and Polunin, N. (1996) Coastal fisheries in the Pacific Islands. Oceanography and marine biology 34: 395-531. Dalzell, P. and Debao, A. (1994) Coastal fisheries production on Nauru. Noumea: Inshore Fisheries Research Project, Country Assignment Report, South Pacific Commission, Noumea, New Caledonia, 19 p. Gillett, R. (2003) Subregional review - Small island developing states of the Southwest Pacific. FAO, Rome, 121-139 p. Gillett, R. (2009) Fisheries in the economies of the Pacific isalnd countries and territories. Asian Development Bank, Mandaluyong City, Philippines, 484 p. Gillett, R. and Lightfoot, C. (2002) The contribution of fisheries to the economies of Pacific Island countries. Asian Development Bank, Manila, 242 p. Jacob, P. (1998) The status of marine resources and coral reefs of Nauru. Coral reefs in the Pacific: Status and monitoring, resources and management: 207-316. Pauly, D., Christensen, V., Guénette, S., Pitcher, T., Sumaila, U., Walters, C., Watson, R. and Zeller, D. (2002) Towards sustainability in world fisheries. Nature 418: 689-695. Petit-Skinner, S. (1981) The Nauruans. MacDuff Press, 321 p. Sokimi, W. and Chapman, L. (2001) Small-scale tuna longlining assistance and training for the Republic of Nauru. Fisheries Development Section. Fiend Report, Noumea, New Caledonia, 27 p. Spennemann, D. (2002) Traditional milkfish aquaculture in Nauru. Aquaculture International 10: 551-562. Stephen, M. (2011) On Nauru, a sinking feeling. New York Times Opinion Pages, July 18, 2011. Available at: www.nytimes.com/2011/07/19/opinion/19stephen.html?_r=1&ref=todayspaper [accessed July 21, 2011] Underwood, J. (1989) Population history of Nauru: A cautionary tale. Micronesica 22: 3-22. Viviani, N. (1970) Nauru: Phosphate and political progress. Australian National University Press, 215 p. Vunisea, A., Pinca, S., Friedman, K., Chapman, L., Magron, F., Sauni, S., Pakoa, K., Awira, R. and Lasi, F. (2008) Nauru country report: Profile and results from in-country survey work. SPC, Secretariat of the Pacific Community, Noumea, New Caledonia, 68 p. Zeller, D., Booth, S., Craig, P. and Pauly, D. (2006) Reconstruction of coral reef fisheries catches in American Samoa, 1950–2002. Coral Reefs 25: 144-152. Zeller, D., Booth, S., Davis, G. and Pauly, D. (2007) Re-estimation of small-scale fishery catches for US flag-associated island areas in the western Pacific: the last 50 years. US Fishery Bulletin 105: 266-277.  70  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Appendix Table A1: Total FAO landings vs. total reconstructed catch for Nauru, 1950-2008, in metric tonnes. Year FAO landings Total reconstructed 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  0 0 0 0 0 0 0 0 0 0 0 0 0 100 100 100 100 100 100 100 100 100 100 100 100 120 120 120 130 130 140 140 140 140 150 150 150 170 170 180 180 190 377 500 500 400 300 250 200 150 109 61 22 44 19 39 39 39 39  296 301 306 311 315 319 323 327 331 334 337 342 347 352 356 360 364 367 370 373 375 378 380 381 382 384 384 385 394 392 391 389 387 385 388 392 394 397 398 399 452 444 441 447 455 464 472 481 490 493 443 442 440 438 437 437 447 447 447  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  71  Appendix Table A2: Total reconstructed catch (t) by major taxa for Nauru, 1950-2008. Others category contains 7 taxonomic groups including miscellaneous marine fishes. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  Katsuwonus Elagatis bipinnulata pelamis 130 132 134 136 138 140 142 143 145 146 147 150 152 171 173 175 176 178 179 180 181 182 183 183 184 184 185 185 188 188 187 186 185 184 185 187 188 188 189 189 213 209 207 210 214 218 221 225 229 230 190 190 189 189 188 188 193 193 193  72 73 74 75 76 77 78 79 80 81 82 83 84 73 74 75 75 76 77 78 78 79 79 79 80 80 80 80 82 82 81 81 80 80 80 81 82 82 82 83 96 93 93 94 96 98 100 102 105 105 105 105 105 104 104 104 107 107 107  Thunnus albacares 16 17 17 17 17 17 18 18 18 18 18 19 19 21 22 22 22 22 22 23 23 23 23 23 23 23 23 23 24 23 23 23 23 23 23 23 23 24 24 24 27 26 26 26 27 27 28 28 29 29 24 24 24 24 24 24 24 24 24  Scombridae 16 17 17 17 17 17 18 18 18 18 18 19 19 21 22 22 22 22 22 23 23 23 23 23 23 23 23 23 24 23 23 23 23 23 23 23 23 24 24 24 27 26 26 26 27 27 28 28 29 29 24 24 21 12 21 21 21 21 21  Carangidae 18 18 19 19 19 19 20 20 20 20 20 21 21 18 18 19 19 19 19 19 20 20 20 20 20 20 20 20 21 20 20 20 20 20 20 20 20 21 21 21 24 23 23 24 24 25 25 26 26 26 26 26 26 26 26 26 27 27 27  Lutjanus kasmiri 14 14 14 15 15 15 15 15 16 16 16 16 16 14 14 14 15 15 15 15 15 15 15 15 15 15 15 15 16 16 16 16 16 15 16 16 16 16 16 16 18 18 18 18 19 19 19 20 20 20 20 20 20 20 20 20 21 21 21  Crustaceans nei 6 6 6 6 6 6 7 7 7 7 7 8 8 8 8 8 9 9 9 9 10 10 10 10 10 11 11 11 12 12 12 12 12 13 13 13 14 14 15 15 15 16 16 16 16 16 16 16 16 16 16 16 16 16 16 15 15 15 15  Others 24 25 25 25 26 26 26 27 27 27 28 28 28 25 25 25 26 26 26 26 27 27 27 27 27 27 27 27 28 28 28 28 27 27 28 28 28 28 28 28 33 32 32 32 33 34 34 35 36 36 37 37 39 48 38 39 39 39 39  72  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  73  MARINE FISHERIES OF PALAU, 1950-2008: TOTAL RECONSTRUCTED CATCH1 Stephanie Lingard, Sarah Harper, Yoshi Ota and Dirk Zeller  Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC, V6T 1Z4, Canada s.lingard@fisheries.ubc.ca; s.harper@fisheries.ubc.ca; d.zeller@fisheries.ubc.ca  ABSTRACT The small Pacific Island nation of Palau has a long history of human settlement. Palau maintained a predominantly traditional lifestyle until the post-war modernization after 1950s, with fishing being a preoccupation for the majority of its male population. This study estimated Palau‘s total marine fisheries catches for the 1950-2008 period to be just over 200,000 tonnes. This total was 43% higher than the official reported data as presented by the FAO on behalf of Palau. The discrepancy was mainly due to subsistence catches which were under-reported in the official statistics. The total coastal catches including subsistence and artisanal, were estimated to be 103,480 t 45% higher than the 46,615 t of coastal catches reported over the period. Our findings illustrate the importance of the subsistence sector, with catches representing 60% of coastal fisheries catches. Better monitoring or at least regular comprehensive estimation of the subsistence sector is key to properly account for the social and economic importance of fishing in Palauan society.  INTRODUCTION Palau, a small country in the Western Pacific, is comprised of 340 islands which lie between 131°-135°E and 2°8°N, 500 kilometers east of the Philippines (Figure 1). Palau is located within FAO statistical area 71, the Western Central Pacific, has a land mass of approximately 488 km2 and an exclusive economic zone (EEZ) of around 604,289 km2 (www.seaaroundus.org). The five main inhabited islands of Palau are Kayangel, Babeldoab, Korror, Peliliu, Angaur, and it has two de-populated outer islands, being Sonsorol and Hatohobei (Figure 1). The history of human settlement in Palau is long, with the earliest archaeological findings in Palau dating back over 2400 years (Clark, 2005). In recent human history, Palau has been successively colonized and under external stewardship by different states; Palau was ruled by Spain in the 1880s, Germany from 1899-1913 and Japan from 1914-1944. Following World War II, Palau was administered by the United 1  Figure 1. Map of Palau and its Exclusive Economic Zone (solid line).  Cite as: Lingard, S., Harper, S. Ota, Y. and Zeller, D. (2011) Marine Fisheries of Palau,1950-2008: Total reconstructed catch. pp.7384. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727].  74  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  States. In 1994, Palau was designated as an independent country under the Compact of Free Association (Anon., 2003a) (Ota, 2006). The Compact of Free Association entails the United States providing military defense to Palau, development aid in the form of annual grants, and scholarship programs for Palauan residence to attend pos- secondary education in the United States. Due to this external support gradual development of some local industries, such as the tourism sector, Palau has the highest per capita GDP among Micronesia, and higher than most small island states in other Pacific region2. Traditionally, Pacific islanders have relied on marine resources as their main source of food, and fishing skills and knowledge were recognized as the status symbol of both wisdom and masculinity among many Pacific cultures (Johannes, 1981; Colbert, 2000). Likewise, in Palau, fishers were highly esteemed members of the community (Ota, 2006). Both Johannes (1981) and Ota (2006) have described fishing to be central to the organization of Palauan communities, embodying the gender dichotomy and social stratification, which still supports the basis of the socio-cultural dynamics of the society. In Palauan tradition, for instance, women have traditionally engaged in reef gleaning activities (Chapman, 1987), and farming of taro, while men capture fin fish which make up the majority of Palauan diet (Johannes, 1981; Mathews and Oiterong, 1991). Sharks and larger pelagic species are usually only caught for special occasions (Johannes, 1981). Prior to colonization, Palauan society was organized into separate communities, each with its own respective chief. Between communities, land and sea barriers were continuous, and each community had proprietorship of sections of the coast line and reef. From ownership of and responsibility for a defined geographical area came a natural conservation ethic (Johannes, 1981). With the succession of colonizing countries, this system was increasingly disturbed by external political control and eventually replaced by modern democracy, which maintains the traditional chieftainship merely as a façade (Ota, 2006). More recently, the increase in the importance of the cash economy has led rapid urbanization of the country as people started seeking work in the capital, Koror (Johannes, 1981; Ota, 2006). In 1990, an estimated 70% of Palauans were living in Korror (Nichols, 1991). Even in the midst of this modernization, fishing has remained both economically and culturally important and is practiced regularly for subsistence purposes, though not commercially, by many Palauans both in their urban and rural settings. However, the centralization of marine resource management to the contemporary governing body has had negative effects on Palauan fishing culture and fish populations from the early days of the Palauan modernization as it slowly replaced the power of traditional community based management (TCBM) system which was based on the indigenous chieftainship (Johannes, 1981). For instance, Johannes (1981) reports that species of the Serranidae (e.g., groupers) were quickly overexploited due to fishers targeting spawning aggregations, which were previously controlled through the traditional community-based management (TCBM) system. Dynamite, poison, and other unsustainable fishing practices have also been employed since the breakdown of the TCBM, resulting in negative impacts on the reef fisheries of Palau (Johannes, 1981). Since the Japanese occupation of Palau in 1914, tuna have been exploited in Palauan waters, and in recent decades, this has provided an important source of income for Palau, as fishing access fees for foreign vessels were introduced. Gillett(2009) estimated that in 2007 approximately 1.2 million USD were paid to Palau for access to fish in their waters. However, during WWII, all off-shore tuna fishing was halted until 1964 when an American company, Van Camp Seafoods, opened a processing facility in Koror (Lawson, 1991). Van Camp Seafoods operated until 1982. Since then, there has only been one locally based off-shore pole-and-line vessel operating under the Palauan Flag (Nichols, 1991). The subsistence fishery has been largely unaccounted for in the FAO data for many island countries, despite its significant contribution to food security and local economies (Zeller et al., 2006). It has been noted that Palau‘s inshore coral reef fishery continues to provide the main source of protein, and financial income for the majority of Palauan people (Johannes, 1981; Ota, 2006). In the recent period it has been estimated that 87% of Palau‘s population is engaged in coastal artisanal and subsistence fishing activities (Palau International Coral Reef Center, unpublished data in Golbuu et al, 2005). The artisanal and subsistence reef fisheries are carried out with a variety of gears. Fish pots, drop lines, trolling, hand 2  http://www.adb.org/Documents/Fact_Sheets/PAL.pdf [accessed June, 2011]  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  75  spears, spear guns, gill nets, set nets (kesokes), and cast nets are the major gear types employed. The major species fished for in the small-scale fishery are snappers (Lutjanidae), emperors (Lethrinidae), groupers (Serranidae), parrot fishes (Scaridae), wrasses (Labridae), rabbitfishes (Siganidae), surgeon fishes (Acanthuridae), trevallies (Carangidae) and herring (Clupeidae) (Nichols, 1991). Invertebrate fisheries are important components of the local diet and economy, but are poorly represented in official fisheries statistics. In earlier time periods most of the invertebrate fisheries were reported as being handled by women and children, predominantly for subsistence use (Johannes, 1981; Mathews and Oiterong, 1991), but they are now caught by commercial fishers and were largely sold at the local market mainly to the tourism sector (Anon., 2003b; Pakoa et al., 2009). Bêche-de-mer is an important part of the Palauan diet, and is often collected by women during reef gleaning activities (Johannes, 1981; Mathews and Oiterong, 1991; Pakoa et al., 2009). Bêche-de-mer is considered a boom and bust fishery, as the catches can fluctuate substantially year-to-year due to its ease of harvest and the open access nature of this fishery (Dalzell et al., 1996). Other important invertebrate resources in Palau are the mud crab (Scylla serrata), land crab (Cardiosoma hirtipes and Cardisoma carnifex), and coconut crab (Birgus latro) (Johannes, 1981; Nichols, 1991; Dalzell et al., 1996). Since the 1990s, there have been some improvements to the documentation of the offshore and artisanal sectors of Palauan fisheries also reflected in better taxonomic breakdown in FAO data; however, to date no studies exist that include all fisheries components in a single estimate with a complete estimated time series (although Gillet [2009] does provide a very comprehensive estimate for the recent time period). The aim of this study is to make a comprehensive estimate of total marine fisheries catches for Palau that includes the invertebrate, subsistence, artisanal, offshore, and baitfish fisheries sectors over the 19502008 time period.  METHODS For this report, officially reported landings were acquired from FAO (FishStat), and data were obtained from government reports, and independent reports published by the Secretariat of the Pacific Community (SPC) and Asian Development Bank (ADB). For the time period 1990-2007, several estimates of subsistence, artisanal and offshore commercial fisheries catches were available for some years (Nichols, 1991; Adams and Dalzell, 1994; Kitalong and Dalzell, 1994; Dalzell et al., 1996; 2001; Gillett and Lightfoot, 2002; Gillett, 2009). Prior to 1990, few studies on commercial reef and off-shore fisheries were available (Johannes, 1981; Perron et al., 1983; 1984). Data for certain years and for certain sectors were taken from Johannes (2009), Kitalong and Dalzell (1994), Dalzell et al.(1996), Gillett and Lightfoot (2002), and Gillett (2009). To derive a complete time series of data (1950-2008), we interpolated linearly between years of known data. Human population data and per capita fish consumption rates were used to calculate total seafood demand and secondarily derive subsistence sector catches for the 1950-2008 time period. 25 Visitors  Human population Population (x 103)  Human population census data were obtained from the Palau Office of Planning and Statistics (http://www.palaugov.net/stats) and used in combination with per capita consumption rates to calculate Palau‘s seafood demand. Linear interpolations were used to create a continuous time series of human population data from 1950 to 2008, as data were not available for all years (Figure 2). To account for the temporary increase in population due to tourists, we converted number of tourists  (full-time equivalents)  20 15  10  Resident population 5  0 1950  1960  1970  1980  1990  2000  Figure 2. Population of Palau, 1950-2008 with the resident population and the number of visitors converted to full-time equivalents.  76  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  into full-time equivalents (Figure 2). To do this, we multiplied the number of visitors by the average number of nights stayed. The length of stay was derived from the estimated number of full-time equivalents presented in Gillette (2009) for the 1990s and the average number of annual visitors for the 1990s taken from the Palau Visitors Authority, published in Yamashita (2000). We estimated the average length of stay to be 3.6 days∙visitor-1∙year-1 and applied this to the time series of visitors, which was for the period 1980-2008. The number of full-time equivalents was added to the resident human population and the total population was then used in conjunction with per capita seafood consumption rates to calculate the total demand of fishery products over the study period.  Commercial Fisheries Inshore fisheries Several independent studies on commercial inshore fisheries estimate annual catches (Table 1). The PCS (2000) estimate of 865 t has been the most widely agreed upon estimate for coastal fish production in Palau (Gillett, 2009). This estimate was compared to total catches presented by FAO for all species except large pelagics (tuna and billfishes). In 2007, FAO (non-pelagic) landings were found to be 11% higher than the PCS estimate of 865 t∙year-1 of coastal commercial catch. We assumed that this additional catch (11%) presented in the FAO data was an estimate of nonTable 1. Comparison of commercial inshore fisheries commercial (subsistence) catches. Therefore, we catch estimates for the recent period (1990s & 2000s) assumed that 89% of FAO reported landings from various independent sources. (excluding large pelagics) represent the coastal Catch (t/year) Source commercial (artisanal) catch, while the remaining 300-400 Shimada (1987 in Nichols, 1991) 11% were considered the reported or estimated 250 Kitalong and Dalzell (1994) component of subsistence catches. 865 PCS (2000 in Gillett and Lightfoot, 2002) 736 Anon. (1993 in Dalzell et al., 1996) Offshore fisheries Palau‘s offshore fishery began in the 1920s when Japan occupied the islands. After WWII, tuna fisheries in Palau ceased and it was not until the early 1960s, after a fisheries development program was launched to jump start Palauan offshore fisheries, that tuna fishing resumed. In 1964, a joint-venture company (Van Camp Seafood of the United States) began operating a pole-and-line tuna fishery in Palau, which lasted until 1982 (Anon., 1984). These catches were considered domestic as they were caught and landed by Palau, even though the financing for this fishery was from the United States. Aside from Van Camp Seafood, offshore tuna fisheries in Palau have been mainly foreign fleets, which land their catches outside Palau (Anon., 1944; Gillett and Lightfoot, 2002). After the closing of Van Camp Seafood, only a single domestic pole-and-line vessel remained in operation. Based on our comparison of FAO tuna and billfish landings with other reports (e.g., Gillett, 2010), we concluded that the FAO data were the best available representation of domestic, large pelagic fishery catches. Due to the minimal bycatch associated with pole-and-line fisheries (Bailey and Williams, 1996; Kelleher, 2005) we did not estimate bycatch, either landed or discarded. However, pole-and-line fisheries do require considerable amounts of live bait which are often caught in reef areas adjacent to the tuna fishing grounds. These catches are rarely accounted for in catch statistics, and we assumed that these were not estimated or included in the FAO data. Therefore, we estimated the amount of fish that was likely caught in order to provide bait for the domestic tuna fishery. Gillette(2011) gives an average tuna-to-baitfish ratio of 26:1 for Palau‘s pole-and-line fishery between 1964 and 1972. We assumed a similar ratio for the entire 1964-1982 period, when Van Camp Seafood was in operation. This ratio was applied to the total tuna landings from 1964-1982. In 1982, the bait fishery in Palau ceased operations, likely as a result of Van Camp Seafood discontinuing operations in Palau (Anon., 1984). Although one domestic tuna vessel continued to fish after 1982, we did not make any further estimates of baitfish catches. Catch of baitfish was likely dominated by short head anchovy (Encrasicholina heteroloba), delicate round herring (Spratelloides delicatulus), and Samoan silverside (Hypoatherina temmincki). From the estimated catch by species given in the SPC (1984) report on the bait fishery, we derived a species breakdown of 56% Encrasicholina heteroloba, 21% Spratelloides delicatulus and 8% Hypoatherina temminckim, with the remaining 15% being Clupeiformes.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  77  Subsistence fishery To independently estimate the subsistence catch in Palau, we used per capita fish consumption estimates for three separate years, and interpolated linearly between these to derive a complete time series of per capita fish consumption rates. The per capita consumption rates used were 122 kg∙person-1∙year-1 for 1974 (Johannes, 1981), 135 kg∙person-1∙year-1 for 1999 (Gillett and Lightfoot, 2002), and 115 kg∙person-1∙year-1 for 2007 (Gillett, 2009). Johannes‘ (1981) estimate of per capita fish consumption was calculated using the weight of protein consumed daily, and the landed weight of fish (with weight of bones and scales removed). The estimate from Gillett and Lightfoot (2002) was calculated using the PCS (2000) in-shore catch estimate of 2,115 t adjusted for imports, exports and full time visitor equivalents. The consumption rate calculated from Gillett (2009), which used the same methodology as the Gillett and Lightfoot(2002) and population and visitation data from 2007, was adjusted for imported and exported fishery products and excluded pelagic species which do not form a significant portion of the Palauan diet. Other estimates of per capita fish consumption were available (Perron et al., 1983; Preston, 1990) but were general estimates for the South Pacific region or were partially derived using FAO or commercial landings data. We disregarded these estimates based on information regarding the Palauan diet which suggested that although imported food, pigs, and bats contribute in small parts to the Palauan diet, fish still remains the main source of protein (Johannes, 1981; Anon., 2003b; Ota, 2006). Our earliest estimate of per capita fish consumption (Johannes, 1981) was carried back fixed from 1974 to 1950 and linear interpolations were used between 1974 and the two later estimates to derive a complete time series of per capita fish consumption rates. These rates were then multiplied by the annual human population to derive the total domestic seafood demand. This was then used to Table 2. Taxonomic composition of the non-pelagic catches (as proportion of total catch) by family or grouping based on Kitalong and Dalzell (1994) for 1950determine whether the 1990 and Friedman et al. (2007) for 2007-2008. A linear interpolation was used demand was met through the to derive a complete time series between 1990 and 2007. supply of reported landings. Family 1950-1983 1984-1990 1991-2006 2007-2008 The discrepancy found Acanthuridae 0.141 0.129 linear interpolation 0.104 between these two numbers Carangidae 0.026 0.040 linear interpolation 0.022 was then considered Gerridae 0.007 0.002 linear interpolation 0.011 unreported subsistence catch. Haemulidae 0.002 0.001 linear interpolation 0.015 To this we added the 11% of Holocentridae 0.004 0.001 linear interpolation 0.011 FAO non-pelagic catches that Labridae 0.010 0.008 linear interpolation 0.008 were considered reported Lethrinidae 0.119 0.139 linear interpolation 0.268 subsistence catches. The total Lutjanidae 0.085 0.135 linear interpolation 0.140 subsistence catch (reported + Mugilidae 0.019 0.005 linear interpolation 0.024 unreported) was then Mullidae 0.010 0.012 linear interpolation 0.023 compared to our estimate of Scaridae 0.163 0.187 linear interpolation 0.154 reconstructed coastal Serranidae 0.091 0.093 linear interpolation 0.105 0.116 0.105 linear interpolation 0.105 commercial catches. On Siganidae 0.185 0.123 linear interpolation 0.005 average, subsistence catches Others 0.020 0.020 linear interpolation 0.005 represented approximately Crustaceans 60% of the total coastal catches (subsistence and artisanal combined). For the South Pacific in general, the ratio of subsistence to total catch can be as high as 80% (Adams and Dalzell, 1994) with the average being around 70%(Gillett, 2009). We thus considered our approach to be conservative.  Invertebrate fisheries Estimates of invertebrate fisheries catches were not readily available; however, sea cucumber (bêche-demer) has been a component of the Palauan diet for centuries (Pakoa et al., 2009). Sea cucumber fisheries often fluctuate considerably from year-to-year (Dalzell et al., 1996); however, catches reported by FAO for this fishery are minimal (<0.5 t), despite it being an important component of the subsistence fishery. Sea cucumber and invertebrates are collected in reef gleaning activities by women (Mathews and Oiterong, 1991). Sea cucumber landings were estimated to be 11.3 t∙year -1 between 1989 and 1998 (Pakoa et al., 2009). Over 50% of this was for subsistence consumption, 48% was sold in local markets and less than 1%  78  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  was exported. Pakoa et al. (2009) estimate that in 2007, sea cucumber catches were approximately 65.5 t∙year-1. Given the substantial fluctuations characteristic of sea cucumber fisheries and the limited catch data, we used the average of Pakoa et al.‘s (2009) estimate for the 1990s (11.3 t) and for 2007 (65.5 t). The average of these two was approximately 38.4 t (19.2 t each from artisanal and subsistence sectors), which we applied as the catch for the year 2000. This was converted to a per capita rate, and using the population time series, sea cucumber catches were estimated for the entire study period. Sea cucumber catch estimates were then subtracted from the ‗other species‘ category of the artisanal and subsistence catch. (Anon., 2003a) Crustacean fisheries are also an important part of subsistence and, to a certain extent, commercial fisheries in the South Pacific (Dalzell et al., 1996). Palau‘s catches of crustacean in the early to mid-1990s were estimated by Dalzell et al. (1996) to be 14.4 t∙year-1, which represents approximately 2% of artisanal catches. Prior to 1990, FAO catches for Palau present all catches as either miscellaneous marine fishes or pelagic fish species. As of 1990, the FAO provides greater taxonomic detail, which includes several invertebrate categories. Between 1990 and 2008, FAO crustacean categories represent on average 0.8% of the total non-pelagic landings. Crustacean catches were subtracted from the ‗other species‘ category of the artisanal and subsistence catches.  Taxonomic composition  Table 3. Species composition of the families, which dominate the coastal fisheries of Palau based on Friedman et al. (2007). Family Taxonomy Proportion of catch by family Acanthuridae Acanthurus xanthopterus 0.22 Naso lituratus 0.34 Naso unicornis 0.21 Acanthuridae 0.23 Carangidae Caranx ignobilis 0.24 Caranx melampygus 0.43 Carangidae 0.33 Gerreidae Gerres macrosoma 0.97 Gerreidae 0.03 Haemulidae Plectorhinchus albovittatus 0.51 Plectorhinchus spp. 0.49 Holocentridae Holocentridae 1.00 Labridae Cheilinus undulatus 0.30 Choerodon anchorago 0.70 Lethrinidae Lethrinus harak 0.09 Lethrinus olivaceus 0.19 Lethrinus rubrioperculatus 0.06 Lethrinus xanthochilus 0.15 Lethrinidae 0.51 Lutjanidae Aprion virescens 0.09 Lutjanus bohar 0.15 Lutjanus gibbus 0.63 Lutjanidae 0.13 Mugilidae Mugilidae 1.00 Mullidae Mullidae 1.00 Scaridae Cetoscarus bicolor 0.17 Hipposcarus longiceps 0.45 Scarus ghobban 0.15 Scarus oviceps 0.11 Scaridae 0.12 Serranidae Plectropomus areolatus 0.22 Plectropomus leopardus 0.25 Epinephelus spp 0.44 Serranidae 0.09 Siganidae Siganus lineatus 0.57 Siganus spp. 0.43  FAO data for Palau have poor taxonomic resolution until the 1990s. For the period 1950-1989, landings are mainly reported as the aggregated grouping, ‗miscellaneous marine fishes (MMF)‘. Perron (1983) presents a breakdown of the artisanal fishery by family, but these estimates were highly variable. Dalzell et al. (1996) has a partial breakdown of the 1992 artisanal fishery including an estimate for the crustacean fishery. Kitalong and Dalzell (1994) have a taxonomic breakdown by family for the years 1976-1990, with two separate breakdowns for the periods 1976-1983 and 1984-1990. Friedman et al. (2007) have the most detailed breakdown of the artisanal fishery including species-level composition for catches in four major fishing districts in Palau. Here, we used the breakdown in Kitalong and Dalzell (1994) for the period 1950-1990 and Friedman et al. (2007) for 2007 and 2008 (Table 2).  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  For each family, a species-level breakdown was derived using Friedman‘s (2007) estimates (Table 3). This species breakdown was applied to each family group throughout the 19502008 time period for the artisanal and subsistence fishery catches.  Catch (t x 103 )  12.0  Bait fish  Tuna  10.0 Reported subsistence  8.0 6.0  Unreported subsistence  FAO  4.0 2.0 0.0 1950  1960  1970  1980  1990  2000  Figure 3. Palau total reconstructed catch including subsistence, artisanal, locally based tuna, baitfish, 1509-2008.  12 10 Catch (t x 103 )  Using Dalzell et al.’s (1996) estimate of crustacean catch for the early 1990s, we assumed 2% of artisanal catches were crustaceans for the period 1950-1990. This was incorporated into our artisanal sector taxonomic breakdown by subtracting 2% from the ‗others‘ category in the original breakdown. In 2007, we assumed a lower proportion of crustaceans (0.5%) in the artisanal breakdown as after 1990 crustacean categories begin to appear in the FAO data. The taxonomic breakdown with crustaceans incorporated was then applied to the unspecified portion of the artisanal catch (miscellaneous marine fish category in the FAO data). We assumed the same species breakdown for the subsistence catch. Sea cucumber catch which was estimated using a per capita rate was then subtracted from the ‗others‘ category for the artisanal and subsistence catch in all years.  14.0  8 6  Baitfish  4  Large pelagic fish  2 0 1950  1960  1970  1980  1990  2000  Figure 4. Total catch of large pelagic species and the baitfish associated with the main gear used in this fishery: pole-and-line.  3.0 2.5 Catch (t x 103 )  Using Friedman et al. (2007) we calculated the taxonomic composition using the weighted average of each fishing district as a proportion of the total catch in the study, and each area (lagoon, sheltered reef, outer reef) as a proportion of the total in each district. The proportion of each species in each region was then weighted according to the fishing area and district. The proportion each species represented in the total catch was taken as the sum of the proportion of the catch by weighted area. To compare this breakdown with that presented in Kitalong and Dalzell (1994), we grouped the species by family (see Table 3). In order to derive a complete time series between 1990 and 2007, we interpolated linearly between the Kitalong and Dalzell (1994) and Friedman (2007) estimates for each taxonomic family. The 1976 estimate was carried back, unaltered to 1950 and the 2007 estimate was carried forward unaltered to 2008.  79  reconstructed  2.0 FAO  1.5  1.0 0.5 0.0 1950  1960  1970  1980  1990  2000  Figure 5. Palau reconstructed and FAO coastal catches 1950-2008.  80  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  RESULTS Our estimate of the total reconstructed catch for Palau, which includes subsistence, artisanal, locallybased tuna fisheries and baitfish totalled 200,817 t for the 1950-2008 time period. This estimate was 43 % higher than the total landings presented by FAO on behalf of Palau for 1950-2008, which was estimated to be 140,483 t (Figure 3). The FAO reports 93,868 t of pelagic species, which were considered to be from the domestic offshore fisheries (Figure 4).  Subsistence portion of total coastal catch  The remaining 49,609 t 0.8 reported by the FAO were considered to be the coastal 0.7 fisheries catches. Of the coastal catches presented by the FAO, 0.6 89% were considered to be from 0.5 the commercial (reported artisanal) sector, while the 0.4 remaining 11% were considered to be non-commercial (reported 0.3 subsistence) catches (Figure 3). 0.2 Reported artisanal catches totalled approximately 41,488 t 0.1 and reported subsistence catches were just over 5,000 t 0.0 (Figure 5). In addition to the 1950 1960 1970 1980 1990 2000 reported subsistence, we Figure 6. Subsistence portion of the total coastal catch, which includes estimated another 56,800 t of artisanal and subsistence catches, 1950-2008. unreported subsistence (Figure 3). Subsistence catches were estimated to be almost 62,000 t for the 1950-2008 time period. Total coastal catches were estimated to be 103,480 t, 45% higher than the total reported coastal catches (Figure 5.) Thus, total subsistence catches accounted for 60% of total coastal catches. Subsistence catches in 1950 represented 70% of the total coastal catch (total artisanal + total subsistence), decreasing to approximately 50% during the 1970s and 1980s and increasing again to approximately 65% in the recent period (1990s and 2000s; Figure 6). The amount of baitfish caught for use in the pole-and-line fishery was estimated to be almost 3,500 t over the period 1964-1982 (Figure 4).  DISCUSSION The total reconstructed catch for Palau from 1950-2008 was estimated to be 200,817 tonnes. This estimate is 43% higher than the landings reported by FAO on behalf of Palau for this same period (140,483 tonnes). The difference between these two estimates is mainly attributed to the addition of subsistence catches, which were predominantly unaccounted for in the reported data. Our work showed that subsistence catches represented an average of 60% of the total domestic catch. For the Pacific Islands region, in general, Adams and Dalzell (1994) estimate that the subsistence catch can represent as much as 80% of the total catch. Currently, an estimated 87% of people are involved in the fishing industry in Palau (Palau International Coral Reef Center, unpublished data in Golbuu et al., 2005). In recent years there has been an increase in the amount of imported food in the Palauan diet (Johannes, 1981; Ota, 2006), but the results of this study show reliance on local reef fish remains high in Palau. The subsistence catch has increased, since 1950, mostly attributed to increase in population size. The subsistence catch will likely continue rising due to a growing population (Anon., 2003a). The majority of the domestic supply of fish in the 1950s was from the subsistence sector (70%) and although the portion of the total domestic catch supplied by subsistence fisheries declined during the 1970s and  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  81  1980s,—likely due to an increase in commercial exploitation of fisheries resources—the subsistence sector continues to dominate domestic fisheries catches. The under-reporting of subsistence catches may adversely affect food security in Palau given the importance of this sector to the Palauan culture and diet. Nichols (1991) notes that one species in particular, the humphead wrasse (Cheilinus undulatus), is rarely sold due to a strong preference for this fish in the local diet. It is caught in unknown amounts and usually kept for home consumption. A lack of monitoring and reporting of species important to the subsistence fishery, such as C. undulatus, could lead to the silent extirpation of species from Palau‘s reefs. The commercial offshore catches in Palau were well documented from 1937-1944 by the Japanese Office of the Chief of Naval Operations. The average catch during this time was approximately 3000 tonnes (Anon., 1944). During WWII, all commercial offshore fishing ceased. Catches prior to WWII, while being quite substantial, are not included as they fall outside the time period considered for this study. Recording of pelagic species, mostly skipjack, began in 1964. From 1964-1982 Van Camp Seafoods operated in Palau, and the catches are well documented in Lawson (1991). Our estimate of total catches is likely still an under-representation of the actual annual catch due to the conservative approach employed in our estimations and the omission of the recreational fishery sector. The South Pacific Commission (1999) estimates roughly 8.3 t of billfish are taken annually in game fish tournaments in Palau. The major species taken in this fishery are blue marlin (Makaira nigricans), black marlin (Makaira indica), striped marlin (Tetrapturus audax), and sailfish (Istiophorus platypterus). Recreational fishing is on the rise in Palau as tourism increases (Nichols, 1991). Better documentation of this fishery is recommended for the conservation of billfish in Palau‘s EEZ. Multiple studies have estimated Palau‘s offshore, artisanal, subsistence and invertebrate fisheries separately; however, this study is the first to combine all these sectors into one comprehensive estimate with a continuous time series from 1950-present. This study revealed some major deficiencies in the availability of fisheries data covering all sectors, with a particular scarcity of data on the subsistence sector, which could be argued is the sector most important to day to day existence of the Palauan people. Our comprehensive estimate of total marine fisheries catches for Palau will hopefully serve as a more realistic baseline from which fisheries management options and tradeoffs can be assessed.  ACKNOWLEDGMENTS We thank the Sea Around Us Project, a scientific collaboration between the University of British Columbia and the Pew Environment Group. We would also like to thank Dr. Takashi Mita of the Global Collaboration Centre in Osaka University for graciously providing input for this study.  REFERENCES Anon. (1944) The Fishing Indistry of the Japanese mandated islands. United States Foreign Economic Administration Supply and Resources services, Office of the Economic Programs, Foreign Economic Administration for Occupied areas section, Cheif of Naval Operations, Washington, D.C. Anon. (1984) An Assessment of the skipjack and baitfish resources of Northern Marianna Islands, Guam, Palau, Federated States of Micronesia, and Marshall Islands. Skipjack Survey and Assessment Programme. South Pacfic Commission, Noumea, New Caledonia, xi+ 111 p. Anon. (2001) Profiles of Palau's Inshore Fisheries, 1989-1998. Palau Conservation Society, Korror, Palau. Anon. (2003a) 2000 Census of Population and Housing of the Republic of Palau. Office of Planning and Statistics, Government of Palau, Korror, Palau. Anon. (2003b) Community Consultations on Marine and Terrestrial Resources Uses. Palau Conservation Society, Korror, Palau. Bailey, K. and Williams, P.I. (1996) By-Catch and Discards in the Western Pacific Tuna Fisheries: A review of SPC Data holdings and literature. Oceanic Fisheries Programme. South Pacific Commission, Noumea, New Caledonia, vii+ 153 p. Chapman, M. (1987) Women's Fishing in Oceania. Human Ecology 15(3): 267-288.Dalzell, P., Adams, T. and Polunin, N. (1996) Coastal fisheries in the Pacific Islands. Oceanography and marine biology 34: 395-531.  82  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Clark, G.R. (2005) A 3000- year culture sequence from Palau, Western Micronesia. Asian Perspectives 44(2): 349380. Colbert, D. (2000) City, Seas and Storms: Managing Change in Pacific Island Economies. Adapting to Climate Change 4. World Bank, Washington, D.C, ix+ 119 p. Dalzell, P., T.J., A. and Polunin, N.V.C. (1996) Coastal Fisheries in the Pacific Islands. Oceanography and Marine Biology 34: 395-531. Friedmand, K., Kronen, M., Pinca, S., Lasi, F., Pakoa, K., Awira, R., Bonblin, P., Tardy, E., Chapman, L. and Margon, F. (2007) Palau Country Report: Profiles and Results from Surfey work at Ngarchelong, Ngatpang, Airai and Korror. . Pacific Regional Oceanic and Coastal Fisheries Development Programme. Secretariat of the Pacific Community, Nomeau, New Caledonia, xxxiv+ 411 p. Gillett, R. (2009) Fisheries in the economies of the Pacific Islands Countries and Territories. Asian Development Bank, Mandaluyong City, Philippines, 254 p. Gillett, R. (2010). Marine fishery resources of the Pacific Islands. FAO Fisheries and Aquaculture technical paper 537, Rome, 71 p. Gillett, R. (2011) Replacing Purse Seining with Pole-and-Line fishing in the central and Western Pacific: Some aspects of the Baitfish requirements. . Marine Policy 35(2): 148-154. Gillett, R. and Lightfoot, C. (2002) The Contribution of Fisheries to the economies of the Pacific Island Countries. Asian Development Bank, Manilla, Philippines. Johannes, R.E. (1981) Words of the Lagoon: Fishing and Marine Lore in the Palau District of Micronesia. University of California Press, Berkeley, California, 245 p. Kelleher, K. (2005) Discards in the World's Marine Fisheries: An Update. FAO Fisheries Technical Paper. Food and Agriculture Organization of the United Nations, Rome, 131 p. Kitalong, A. and Dalzell, P. (1994) A Preliminary Assessment of the Status of Inshore Coral Reef Fish Stocks in Palau. South Pacific Commission, Noumea, New Caledonia. Lawson, T. (1991) Status of Tuna Fisheries in the SPC Area During 1990, with annual catches since 1952. South Pacific Commission, Noumea, New Caledonia, 40 p. Mathews, E. and Oiterong, E. (1991) The Role of Women in the Fisheries of Palau. Master of Science, University of Oregon, Corvallis, Oregon. Nichols, P. (1991) Republic of Palau Marine Resource Profiles. Fisheries Development Section. Forum Fisheries Agency, Noumea, New Caledonia, iv+ 112 p. Ota, Y. (2006) Custom and Fishing: Cultural meaning and Social Relations of Pacific Fishing, Republic of Palau, Micronesia. PhD, University College of London, England, London. Pakoa, K., Lasi, F., Tardy, E. and Friedmand, K. (2009) The status of Sea Cucumbers exploited by Palau's Subsistence Fishery. Secretariat of the Pacific Community, Noumea, New Caledonia, vi+23 p. Perron, F., Nauro, A. and Patris, S. (1983) The Palau reef fish production study: A baseline study of the commercial reef fishing industry in Palau and a blueprint for the development of permanent fisheries managment system. . Division of Marine Resources, Korror, Palau. Preston, G. (1990) Inshore fishery resource management in Palau. South Pacific Commission, Noumea, New Caledonia. Yamashita, S. (2000) The Japanese Encounter with the South: Japanese Tourists in Palau. . The Contemporary Pacific 12(2): 26. Zeller, D., Booth, S. and Pauly, D. (2006) Fisheries Contributions to the Gross Domestic Product: Understanding small-scale fisheries in the pacific. Marine Resource Economics 21(4): 39.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  83  Appendix Table A1: FAO landings vs. total reconstructed catch for Palau, 1950-2008, in metric tonnes. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  FAO landings 300 300 300 300 400 400 400 400 500 500 500 500 600 600 1,866 3,370 3,386 4,106 5,756 5,462 8,882 2,943 2,319 3,150 7,608 7,069 6,136 4,883 10,602 6,699 7,516 10,276 5,053 1,041 1,037 1,100 1,100 1,100 1,100 1,090 1,076 1,093 1,271 1,211 1,086 1,027 1,000 913 952 962 1,097 1,086 1,030 1,050 1,081 935 969 986 1,008  Total reconstructed catch 863 903 942 982 1,021 1,061 1,100 1,140 1,167 1,195 1,222 1,250 1,277 1,304 2,542 4,130 4,174 4,915 6,621 6,309 9,811 3,703 3,118 3,971 8,588 8,026 7,051 5,705 11,647 7,507 8,423 11,396 5,874 1,796 1,808 1,926 2,002 2,083 2,132 2,065 2,201 2,170 2,447 2,388 2,436 2,421 2,512 2,585 2,640 2,672 2,807 2,727 2,714 2,740 2,685 2,517 2,473 2,437 2,492  AppendixTableA2: Total reconstructed catch (t) for Palau by major taxa.Others category represents 58 taxonomic groups including miscellaneous marine fishes. Year  1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  Katsuwonus pelamis  1,025 2,497 2,615 3,354 5,039 4,629 8,081 2,133 1,463 2,309 6,647 5,971 4,911 3,592 9,391 5,687 5,580 6,931 3,438 75 32 82 112 139 119 72 80 61 -  Lethrinidae  53 55 58 60 62 65 67 70 71 73 75 76 78 80 81 83 85 84 84 84 87 90 94 94 93 93 93 93 92 92 96 99 102 105 126 130 133 136 140 141 161 168 177 189 206 220 234 253 278 283 290 296 305 317 313 317 321 325 336  Hipposcarus longiceps  63 66 68 71 74 77 80 83 85 87 89 91 93 95 97 99 101 100 100 99 103 108 112 111 111 111 110 110 110 110 114 117 121 125 148 152 156 160 165 166 159 163 167 171 178 178 184 195 195 196 193 187 187 186 177 172 167 163 166  Lutjanus gibbus  46 48 51 53 55 57 59 61 63 64 66 67 69 70 72 73 74 74 74 73 76 79 83 82 82 82 82 81 81 81 84 87 90 92 151 156 159 164 168 169 162 169 175 182 192 193 203 217 221 225 225 220 223 225 217 214 211 209 212  Thunnus albacares  141 173 71 52 17 133 1 10 56 41 161 298 412 420 303 1 996 2,480 615 15 19 22 38 5 8 62 39 31 3 2 1 1 1 63 41 3 19 28 0 -  Siganus lineatus  57 60 62 65 67 70 73 75 77 79 81 83 84 86 88 90 92 91 91 90 94 98 102 101 101 101 100 100 100 100 104 107 110 114 106 110 112 115 118 119 114 119 123 127 134 135 142 151 154 156 156 152 154 155 149 147 144 143 145  Siganus spp.  43 45 47 49 50 52 54 56 58 59 60 62 63 64 66 67 69 68 68 68 70 73 76 76 75 75 75 75 75 75 77 80 82 85 80 82 84 86 89 89 86 89 92 95 100 101 106 113 115 117 116 114 115 116 112 110 108 107 109  Others  601 629 657 684 712 739 767 794 814 833 852 871 890 909 972 1,048 1,068 1,090 1,148 1,133 1,297 1,112 1,132 1,157 1,319 1,297 1,268 1,234 1,495 1,361 1,372 1,495 1,316 1,199 1,164 1,199 1,227 1,260 1,295 1,303 1,428 1,460 1,588 1,583 1,592 1,594 1,640 1,654 1,677 1,694 1,762 1,721 1,728 1,733 1,689 1,557 1,521 1,491 1,525  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  85  RECONSTRUCTION OF SRI LANKA‘S FISHERIES CATCHES: 1950-20081 Devon O‘Mearaa, Sarah Harpera, Nishan Pererab and Dirk Zellera Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC, V6T 1Z4, Canada bLinnaeus University, SE 39182 Kalmar, Sweden d.omeara@fisheries.ubc.ca; s.harper@fisheries.ubc.ca; nishan.perera@lnu.se; d.zeller@fisheries.ubc.ca aSea  ABSTRACT Sri Lanka has a long history of reliance on the sea for the nutritional and economic well-being of its people. Fishing has long been an important industry and, while detailed fishing records exist dating back to the early 1900s, they are incomplete. In this study, we estimated total marine fisheries catches for the 1950-2008 time period by accounting for all fisheries sub-sectors and components and compared this to the reported landings as provided to FAO. Our total reconstructed catch which included commercial and subsistence catches, and discarded bycatch was estimated at almost 18 million tonnes over the 1950-2008 time period. This estimate was over 2 times larger than the total landings reported by Sri Lanka to the FAO. The majority of this discrepancy was due to catches from the subsistence sector and discarded bycatch associated with shrimp trawl fisheries. Improved monitoring of, and record-keeping for, these fisheries components is crucial to the longterm management of Sri Lanka‘s fisheries and to maintaining livelihoods and food security of the Sri Lankan people.  INTRODUCTION The Democratic Socialist Republic of Sri Lanka is an island country southeast of India within the Bay of Bengal (Figure 1). The climate is tropical with seasonal monsoon and cyclones, but no upwelling. In 2009, the population was 20 million (Anon, 2009) with 32 percent living in coastal areas (UNEP, 2001). The Sri Lankan Exclusivity Economic Zone (EEZ) lies within FAO statistical area 57 (FAO, 2011). The island was colonized by the Portuguese and the Dutch, but most influentially by the British. Sri Lanka, or ―Ceylon‖ as it was known prior to 1972, was a strategic military and trade link between West Asia and Southeast Asia. It acquired independence from the British Empire as the Dominion of Ceylon in 1948, just after World War II. In 1972, Ceylon became a republic and the name was changed back to the pre-colonial name: Sri Lanka (De Silva, 1981). Attempts to record fisheries data in Sri Lanka may have begun during British rule; however, a rigorous island-wide attempt to estimate total landings did not start until after independence. Since 1910, general fisheries information was recorded by the resident marine biologist as part of an annual fisheries administration report. These reports included descriptions of traditional fisheries, destructive practices, fisheries regulations, results of test fisheries, policy changes, and financial record keeping; yet, information regarding landings on the island was incomplete (Pearson, 1911; 1922). By the 1930s, the importance of quantifying total landings was recognized, and by the 1940s, efforts to quantify landings were well underway with the appointment of 12 fisheries inspectors (FIs) within 20 fisheries districts. In the early 1950s, the number of FIs was increased to 24. The first comprehensive annual report of total landings was published in 1952 by the Department of Fisheries (DOF); the reports were, from then on, published annually (reviewed in Sivasubramaniam, 1997). Records of landings in the 1950s focused mainly on the traditional practice of beach seining as it accounted for approximately 40% of total landings (Canagaratnam and Medcof, 1956). The use of the large beach seine, madella, began in the mid to late 1800s and continued to be the most commonly used traditional fishing techniques throughout the twentieth century (Alexander, 1977). Gillnetting began in the 195os, and eventually took over as the most widespread fishing method for small-scale fishers. Incidents of 1  Cite as: O‘Meara, D., Harper, S., Perera, N. and Zeller, D. (2011) Reconstruction of Sri Lanka‘s fisheries catches: 1950-2008. pp. 8596. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727].  illegal dynamite fishing and fish poisoning were also reported. The DOF showed great interest at this time in test fisheries, with special attention to experimental dredging for pearl and windowpane oysters, as well as trawler surveys (Sivalingam, 1961). Artisanal and traditional fisheries in the 1950s could not meet the island‘s domestic demand for marine fish. Thus, markets were supplemented with cheap imports of predominantly dried fish products from Pakistan, Japan, and India. Sri Lanka was not a large exporter of marine fish with the exception of a small market in Thailand and Singapore for shark fins, sea cucumbers and ornamental shells called ‗chanks‘ (Turbinella pyrum). Domestic marine fish production and export capacity were limited by poor infrastructure, most importantly the lack of ice and salt at landing sites, and inefficiencies attributed to the traditional nature of the fishery. In an attempt to improve upon traditional methods, the DOF imported nylon nets and implemented the development of a craft motorization program (Canagaratnam and Medcof, 1956). Subsidies for 11,000 outboard motors and Figure 1. Map of Sri Lanka and ts Exclusive Economic Zone. the introduction of 17-23 foot fibrereinforced plastic (FRP) boats were credited for the subsequent high annual growth rate of Sri Lanka‘s fisheries that lasted until the beginning of civil war in 1983 (RAPA, 1989). In the last few decades, there has been an effort to augment pelagic fisheries though government assistance to increase the number of multiday vessels capable of fishing offshore and in international waters. With the aid of the FAO, statistical methods again improved in the 1970s with the removal of the position of statistical officer and the appointment of an additional 143 FIs, while a new sampling system was also adopted that utilized landing centers as primary sampling units, and boats as secondary sampling units. In 1981, the National Aquatic Resources Research and Development Agency (NARA) was established with the mandate to improve research and development, with an emphasis to better understand tuna biology and catch statistics by way of a collaborative effort with the Indo-Pacific Tuna Programme (IPTP), the Bay of Bengal Programme (BOBP), the Food and Agriculture Organization (FAO) and the Asian Development Bank (ADB) (Dayaratne and Maldeniya, 1996). Gillnetting, a practice that had become popular in the 1960s, continued as a favorite of Sri Lankan fishers and by the 1970s, was accountable for 60% of reported fisheries catches. Shortly after the establishment of NARA, civil war broke out between the Liberation Tigers of Tamil Eelam (LITE) and the Government of Sri Lanka (GoSL). The effect of the war on fisheries was considerable, especially in the north where restrictions (e.g., a ban on outboard motors greater than 40 hp, Maldeniya, 1997b) on fishers were put in place to prevent fuel and weapons from being illegally brought from India by the LITE. Additionally, the conflict led to the destruction of boats, gear, and infrastructure which included ice making facilities and highways important for fish transport to distant markets (Silucaithsam and Stokke, 2006). The northern fishing grounds, once responsible for producing over 40% of the country‘s reported landings, were the most productive and accessible fishing grounds in Sri Lanka due to the presence of a large continental shelf and a trawlable bottom (Engvall et al., 1977). The 1990s saw an increase in reported landings due to improvements in the security situation in some areas of the north and the expansion of the fishing fleet offshore and internationally. By the 1990s, government officials recognized coastal resources were fully exploited, and efforts were shifted to expanding the potential of deep sea fisheries by providing boat and equipment subsidies (Mallikage,  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  87  2001). For billfish, this was attributed to improvements in gear and the expansion of fisheries into offshore and deep sea areas (Maldeniya et al., 1996). Methods for improvement of catch statistics have been made in the 2000s, but overall, they remained the same since the changes made in 1981. The demand for marine fish has remained high, with a catch that was insufficient to meet demand. Despite the increase in multiday fishing vessels and other larger craft a large component of the marine fishing fleet continues to consist of small FRP boats with outboard motors as well as non-motorized traditional craft (FAO, 2006). The tsunami in December 2004 seriously affected 90% of the fishing community through losses of boats, fishing nets, housing, and lives. Eighty percent of fishing villages were completely destroyed, along with 12-14 fishing harbors (ITDG, 2005). Post-tsunami efforts to rebuild fisheries have resulted in an overabundance of fishing boats in some areas raising concerns for overfishing (Jayasuriya et al., 2005). With the end of the civil war in 2010, efforts to increase fisheries production in the north were a high priority for the DOF. Growing domestic demand for seafood and the potential for substantial earnings from seafood exports appear to be the driving force behind current fisheries policy, with plans to double marine fisheries production in the future. Apart from increasing landings, offshore fisheries have been identified as a more viable source of high value export oriented species such as tuna. The lack of adequate offshore fishing capacity has been seen as a major obstacle to fisheries expansion, and there have been initiatives to allow commercial fishing by foreign vessels in exchange for access fees and prescribed landings in order to increase domestic fish supply (Anon., 2010). Small-scale subsistence fisheries are often not considered when collecting fisheries statistics; however, they can constitute a large portion of actual catches (Zeller et al., 2007). The goal of this study was to more accurately quantify total marine fisheries catches, by taking into account all fisheries sub-sectors and components, including subsistence catch and discarded bycatch. The importance of fisheries to the livelihoods of Sri Lankan‘s, particularly coastal dwellers, requires a more comprehensive estimate and accounting of the true magnitude of fisheries extractions.  METHODS 25 Population (x106 )  Total marine fisheries catches were estimated using information obtained from national reports, independent studies, local experts and grey literature. Landings data presented by the FAO on behalf of Sri Lanka were compared to national landings data, and household surveys were used to estimate total demand for domestic seafood as compared to local supply. We also estimated discarded bycatch for the shrimp trawl and tuna longline fisheries. In this report we refer to ‘landings‘ as the amount of fish caught, brought to shore and recorded, while ‘catch‘ refers to the total amount of fish caught, and includes Illegal, Unreported and Unregulated (IUU) catches and discarded bycatch.  20 15  10 5 0 1950  1960  1970  1980  1990  2000  Year Figure 2: Human population trend for Sri Lanka. Data source: www.populstat.info and World Bank (Anon, 2009).  Population Human population data were obtained for the 1950-1959 period from Populstat (www.populstat.info) and for the 1960-2008 time period from the World Bank (Anon, 2009). Population estimates were used to derive per capita marine supply and subsistence catch rates. The population of Sri Lanka has increased steadily from 7 million in 1950 to over 20 million in 2008 (Figure 2).  Commercial fisheries  Catch (t x 104 )  Total commercial landings for Sri FAO 14 Lanka were available in nationally 12 published reports as well as by the National FAO; however, the national data 10 contained a statistical error causing 8 landings to be high for years prior to 1970 (Pathirana, 1972); landings 6 reported to FAO and obtained from 4 FAO FishStat were lower than 2 nationally reported landings prior to 1970 (Figure 3). Therefore, it was 0 assumed that the statistical error in 1950 1960 1970 the national data was accounted for Year and corrected in landings presented in FAO‘s FishStat. Landings Figure 3: Comparison of landings data as presented by FAO and the national data source, indicating the statistical error in presented by the FAO were also more the national data, and its correction in data presented by FAO complete from 1980-1990, where on behalf of Sri Lanka. national landings data were sparse. FAO data for crustaceans were compared to prawn and lobster landings presented by NARA. For the 1994-2002 time period, prawn and lobster landings were used in place of the FAO‘s ‗miscellaneous marine crustaceans‘ grouping, as they were deemed to be a better representation of total crustacean catches (Figure 5). Marine crab fisheries, although known to occur in Sri Lanka, were assumed to be contained within a new, but smaller miscellaneous crustaceans category as no data was available to determine catch. With the exception of the amendment to crustaceans landings, the remainder of the FAO data was considered a good representation of commercial fisheries landings, both for the artisanal and industrial sectors. These landings were used as a baseline, to which we added components not accounted for in the officially reported data. Noteworthy are two nonfishery related events which are correlated with a noticeable decrease in landings over the time period considered; the beginning of the civil war in 1983 and the tsunami which occurred on December 26th, 2004. Discards Shrimp trawl fisheries are typically associated with considerable bycatch, which can either be landed or discarded at sea. A study in the late 1970s estimated bycatch associated with the shrimp fishery in two of Sri Lanka‘s main shrimp trawling grounds, Jaffna and Mannar (Subasinghe, 1981). The study provided estimates for both the landed and discarded components of the bycatch. Subasinghe (1981) presents discard rates for both areas, which gave an average rate of 10.2 kg of discards per kg of shrimp landed for 1979 (Discard rates for Mannar and Jaffina were 8.92 and 11.48 kg discarded per kg landed, respectively). These two regions were responsible for 60% of the commercial production of shrimp that year (Subasinghe, 1981; Saila, 1983). Therefore, we assumed that this discard rate was representative of Sri Lanka‘s shrimp trawl fisheries and applied the rate of 10.2 kg discards per Table 1. Estimated seafood kg of shrimp landed across the entire time period. Discards may have consumption rates derived from been even higher in earlier time periods due to greater benthic biomass the Department of Census and Statistics 2007 Household and/or less storage capacity on vessels for non-target species; however, to Income and Expenditure remain conservative we held the discard rate constant back in time to Survey. 1950. For the recent time period, we carried the 1979 discard rate Year Per capita demand forward, unaltered, to 2008. This same study reported that over 80% of -1 (kg∙person ∙year-1) the discarded catch was silverbellies (Leiognathidae); we considered the 1981 19.39 remainder to be miscellaneous small pelagic fishes and miscellaneous 1986 18.24 sharks. 1991 2002 2005 2007  14.64 19.86 24.12 24.12  Depending on the type of gear used, bycatch is also of concern for tuna fisheries. The majority of tuna catches in Sri Lanka are skipjack tuna (Katsuwonus pelamis), representing roughly 60% of tuna catches and yellowfin tuna (Thunnus albacares), representing approximately 20% of the tuna catches. Tuna are predominantly caught using gillnets, although, longlines are becoming increasingly popular for catches  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  89  aimed at the export market (Maldeniya, 1997b). Kelleher (2005) estimates discards by tuna longline in Sri Lankan waters to be 0.05%. Given that this was a very low discard rate, and given that we were unable to determine the portion of the tuna catch taken by longline, we did not estimate this component of the bycatch. As for bycatch associated with the tuna gillnet fisheries, information was also quite sparse. Due to the size of the nets used, incidental catch in the tuna gillnet fishery is mainly seerfish, billfish and shark. Given that these are marketable species, we assumed that the majority of the non-targeted catch for the tuna gillnet fishery was retained and that this portion of the catch was accounted for in the landings data.  Subsistence fisheries 35 30  Catch (t x 104 )  We assumed that the subsistence component of small-scale fisheries was unaccounted for in the reported data. To estimate this component of the total catch, we calculated the island-wide marine seafood demand using per capita consumption data from the 2007 Department of Census and Statistics Household Income and Expenditure Survey (Anon, 2007); and compared this to the reported (commercial) landings presented by the FAO. We considered the difference between the supply of marine products for human consumption and the demand for seafood to be the subsistence catch.  25  20 15  Artisanal  10 5 0 1950  Industrial 1960  1970  1980  1990  2000  Year Figure 4: Total commercial fisheries catches for Sri Lanka,  separated by industrial and artisanal fisheries, 1950-2008. The supply of marine products available for consumption by the local population was estimated as the commercial landings (FAO data) adjusted for imports and exports (W. Swartz, unpublished data, UBC Fisheries Centre). These adjusted landings were then converted to per capita supply rates using human population data.  Catch (t x 10)3  To estimate marine demand, the per capita marine fish consumption was obtained from the 2007 Sri Lanka Department of Census and Statistics Household Income and Expenditure Survey (HIES). A detailed breakdown of per capita consumption of marine products was available for 2007 only. The per capita consumption of fish, which included 18 aquaculture and freshwater products, was 16 summarized in the 2007 survey for the 14 years: 1981, 1986, 1991, 2002, and 2005. In Add-on 12 order to remove aquaculture and Marine Crustaceans 10 freshwater consumption and calculate 8 marine consumption, we assumed that the 6 ratio of freshwater and aquaculture 4 consumption to marine fish consumption 2 remained the same over the entire survey 0 period. This assumption resulted in a conservative estimate of per capita marine 1950 1960 1970 1980 1990 2000 consumption as aquaculture and Year freshwater fish consumption have likely Figure 5: Reported landings of marine crustaceans (light increased since the 1980s. However, in colour) and the additional estimated catches during the order to remain conservative, the amounts 1990s and early 2000s. removed were assumed to be proportional to those in 2007. Conversion factors provided by the FAO for Indonesia (FAO, 2000) were used to convert product weight from the 2007 HIES into live weight. The resulting per capita seafood consumption rates for 1981, 1986 and 2007 were used as anchor points to derive a complete time series of consumption rates for the 1950-2008 study period (Table 1). We did not use the 1991 and 2002 estimates of per capita consumption since these points exactly matched FAO reported landings when they were multiplied by the human population. These points were likely estimates of per caput consumption (reported landings divided by the population) and hence left out  Catch (t x 103 )  of the analysis. We assumed that the consumption rate in 1950 was the same as that in the 1980s, and therefore carried the 1981 rate of 19.39 kg∙person-1∙year-1 back, unaltered to 1950. The 2007 estimate was carried forward to 2008. Years between anchor points were interpolated linearly. Finally, we subtracted the per capita marine supply (FAO landings adjusted for imports and exports) from the total per capita seafood demand to determine the per capita subsistence catch rate. Human population data were then used to convert per capita subsistence catch rates into total subsistence catch amounts. This calculation was not done for 2005 since, although the 2005 consumption estimate was thought to be reasonable, the reported landings were low due to the tsunami, which was likely the result of both fewer catches and poor reporting. The subsistence catch rate for the year following the tsunami (2005) was estimated by linear interpolation between the 2004 and 2006 160 subsistence catch rates and then was reduced by the same percent decline in catch (42%) as 140 reported by the FAO for landings between the 120 years 2004 and 2005. It is possible that 100 subsistence was underestimated for anchor 80 points following the beginning of civil conflict in 1983 as it is unlikely surveys included 60 regions at war. The 2007 HIES states that 40 Trincomalee and the Northern Province, 20 known for high marine productivity and possibly higher per capita consumption, were 0 not sampled in 2007 due to active conflict in 1950 1960 1970 1980 1990 2000 these areas; consequently, it is likely the per Year capita consumption and hence the Figure 6: Reconstructed total subsistence catches for subsistence catch estimates are conservative. Sri Lanka, 1950-2008.  Subsistence catches were assumed to be composed of small pelagic species (50%), demersal species (40%) and invertebrates such as crabs and cephalopods (10%). The small pelagic species caught were mainly clupeids and scombrids, with the most common species being Sardinella gibbosa, S. albella, Amblygaster sirm, A. clupeoides, Rastrelliger kanagurta, and Auxis thazard. Demersal species catches were mainly represented by Lethrinidae, Carangidae, Myliobatidae, Sciaenidae, Haemulidae, Leognathidae, and Acanthuridae (Canagaratnam and Medcof, 1956; Maldeniya, 1997a; MFAR, 2008). Industrial and artisanal catches were also improved for FAO ―crustaceans nei‖ utilizing assumptions based on Jayawardane et al. (2003). The species breakdowns for lobster and sea cucumbers were also improved based on local expert opinion (N. Perera, pers. obs., Linnaeus University)  Other IUU components While catches of sea cucumbers and sharks are reported in the official landings data, they are likely underestimates. Unreported catches of sea cucumbers and sharks are common in Sri Lankan waters; however, data on these were not readily available. Although we were unable to account for this unreported component as part of the reconstructed catch, it should be noted that IUU fishing is known to occur in Sri Lanka and should be further investigated (P. Ganapathiraju, pers. comm., UBC Fisheries Centre).  RESULTS Commercial fisheries Total marine fisheries catches by the commercial sector (artisanal and industrial) were estimated to be 8.4 million tonnes over the 1950-2007 time period (Figure 4). Catches in 1950 were approximately 20,000 t∙year-1 and increased steadily to over 300,000 t∙year-1 in 2004. This was followed by a substantial decrease in catches to around 15,000 tonnes in 2005, the year after the tsunami devastated Sri Lanka. Total commercial catches were composed of small- (artisanal) and large-scale (industrial) sectors, which represented 55% and 44%, respectively of the total commercial catch. The total commercial catch included over 50,000 tonnes of additional crustaceans, which were not represented in the reported landings as presented by FAO (Figure 5). Catches of marine crustaceans were estimated to be 320,000 tonnes for the period 1950-2008. These were mainly shrimp (75%) and lobster (9%), with the remainder being  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Subsistence Fisheries Total catches by the subsistence sector were estimated to be over 7 million tonnes from 1950-2008 (Figure 6). Subsistence catches remained relatively stable over the entire study period with an average annual catch of around 120,000 t·year-1 (Figure 8). A decrease in subsistence catches was observed for the late 1970s and early 1980s, but they increased again after that.  600  Supplied to FAO 500  Catch (t x 103 )  miscellaneous marine crustaceans. Discards associated with the shrimp trawl fishery were estimated over the study period to be approximately 2.4 million tonnes (Figure 8).  Total reconstructed catch  400 300 200  100 0 1950  1960  1970  1980  2000  60 Subsistence  Catch (t x 104 )  50  Discards  40 Artisanal  30 20  10  Industrial  0 1950  1960  1970  1980  1990  2000  Year Figure 8: Total reconstructed catches for Sri Lanka by component or fisheries sector, 1950-2008.  Scomber spp. Carangidae  60  Thunnus albacares Catch (t x 104 )  50 40  30  Katsuwonus pelamis Clupeoids Leiognathidae Clupeidae  20 10  DISCUSSION  1990  Year Figure 7: Total reconstructed catches compared to the data submitted by Sri Lanka to FAO, 1950-2008.  Total reconstructed catch The total reconstructed catch of marine fisheries in Sri Lanka was estimated to be almost 18 million tonnes over the 19502008 time period (Figure 7). This estimate of total catches was 2.13 times larger than the landings officially reported by Sri Lanka to the FAO. Reported landings, as presented by the FAO on behalf of Sri Lanka were 8.4 million tonnes. The subsistence catch represented 40% and discards represented 13% of the total estimated catch (Figure 8). The remainder of the total catch was from the artisanal (26%) and industrial (21%) sub-sectors of commercial fisheries. The estimate for commercial catch was almost entirely based on reported landings, while the subsistence and discards were entirely unreported components. Major contributing taxa in the reconstructed catch included silverbellies (Leiognathidae), skipjack tuna (Katsuwonus pelamis), herrings, sardines, and anchovies (Clupeoids), jacks (Carangidae), and yellowfin tuna (Thunnus albacares; Figure 9).  91  Others  0  1950  1960  1970  1980  1990  2000  Total marine fisheries catches for Sri Year Lanka were estimated to be approximately Figure 9: Total reconstructed catches with main taxa 18 million tonnes over the 1950-2008 caught. All other taxa (88 total) were grouped into ‗Others‘ time period. This estimate was over 2 category. times larger than the landings reported by Sri Lanka to the FAO, which was approximately 8.4 million tonnes. This large discrepancy indicates a clear need for improvements in the collection and reporting of fisheries statistics in Sri Lanka. Our investigation into the fisheries of Sri Lanka revealed that information on subsistence fisheries, discarded bycatch and other IUU components was quite limited, even though these fisheries components contributed substantially to overall marine fisheries  catches. Subsistence fisheries catches were the largest unreported component of the catch, and represented 40% of the total reconstructed catch. Discards from the shrimp trawl fishery were also a substantial contributor (13%) to the total catch, and was unaccounted for in the official data. The high rate of discarding in Sri Lankan shrimp trawl fisheries has been attributed to the limited cold storage facilities on multi-day boats. Economically important species are often stored while other less valuable species are discarded. However, the majority of the bycatch consists of low-valued species of Leiognathidae, which are typically discarded (Subasinghe, 1981). Tuna longline fisheries, on the other hand, have a much lower discard rate (0.05%) according to Kelleher (2005). This low rate of discarding is thought to be due to fishers targeting and landing multiple species of high economic importance, thus reducing the amount of discarded fish (Kelleher, 2005). Beach seining in the early period (1950s) was reported to have few discards, with the exception of jellyfish which were known to seasonally clog nets (Canagaratnam and Medcof, 1956). The year following the tsunami, reported landings were significantly lower, even though seafood consumption remained constant. Although DOF offices were badly damaged by the tsunami and efforts were directed into emergency measures rather than accounting for landings, it is likely catches also decreased, especially in small-scale and subsistence fisheries as they sustained a large amount of damage. An assessment of the impacts of the tsunami on coastal fishers suggested that fishing pressure may have initially decreased in 2005, but then increased to pre-tsunami levels caused by excessive replacement gear and vessels donated to local fishers as part of the relief effort (De Silva and Yamao, 2007). Illegal, unreported and unregulated fisheries are known to occur in Sri Lankan waters. For example, the transshipments of shark fins caught in Sri Lanka‘s EEZ occur regularly. Most Sri Lankan vessels lack adequate refrigeration capacity and will therefore trade sacks of shark fins to foreign vessels at sea, which allows them to empty their hold and continue fishing while at sea. Such catches are not included in any reporting mechanism. Sri Lankan vessels also participate in the poaching of sharks and sea cucumbers, which are caught illegally outside of Sri Lanka‘s EEZ in the poorly regulated waters of Somalia, Madagascar, and the Seychelles, and are then landed in Sri Lanka. These catches are reported as domestic landings, but this is not necessarily the case. A lack of enforcement and proper reporting has allowed these IUU fisheries to continue and possibly expand over time (P. Ganapathiraju, pers. comm., UBC Fisheries Centre). Additionally, foreign vessels, in particular Indian vessels, engage in illegal fishing within Sri Lanka‘s EEZ and these catches are not reported for Sri Lanka. Previous attempts to estimate the potential sustainable yield in Sri Lankan waters suggested harvest rates of 250,000 t∙year-1, with around 80,000 t allocated to demersal species catches and 170,000 t for pelagic species (RAPA, 1989). Our reconstructed catches indicate that this level was likely surpassed as far back as 1974. In this study we highlighted the lack of proper accounting for total fisheries catches, which in the case of the subsistence sector accounted for almost half of the domestic marine food supply. Without a realistic estimate of what is being extracted, fisheries are likely to be mismanaged and possibly overexploited. Although human and financial resources may not be available to establish and maintain in depth monitoring programs, regular surveys conducted every few years have been found to be very effective in estimating subsistence and small-scale catch in other developing countries (Brouwer et al., 1997; Zeller et al., 2006).  ACKNOWLEDGMENTS We thank the Sea Around Us Project, a collaboration between the University of British Columbia and the Pew Environment Group. The present work was funded by FAO through the Bay of Bengal Large Marine Ecosystem Project (www.boblme.org/). We thank Rudolf Hermes (FAO-BOBLME) for critical comments.  REFERENCES Alexander, P. (1977) Sea Tenure in Southern Sri Lanka. Ethnology 16 (3): 231-251. Anon (2007) Household income and expenditure survey. Sri Lanka Ministry of Finance and Planning- Department of Census and Statistics, Colombo, 117 p. Anon (2009) World Bank: World Databank. http://data.worldbank.org/country/sri-lanka Accessed: Nov 4, 2010.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  93  Anon. (2010) Saudi Arabia to fish in Sri Lankan waters. ColomboPage. http://www.colombopage.com/archive_10C/Nov28_1290968294KA.php. Brouwer, S.L., Mann, B.Q., Lamberth, S.J., Sauer, W.H.H. and Erasmus, C. (1997) A survey of the South African shore- angling fishery. South African Journal of Marine Science: 165-175. Canagaratnam, P. and Medcof, J.C. (1956) Bulletin No. 4: Ceylon's Beach Seine Fishery The Fisheries Research Station, Ceylon. Dayaratne, P. and Maldeniya, R. (1996) Status report on the development of tuna research and data collection National Aquatic Resources Agency. Colombo, Sri Lanka. ftp://ftp.fao.org/fi/CDrom/IOTC_Proceedings(19992002)/files/proceedings/miscellaneous/ec/1996/EC604-01.pdf. Accessed: Jan 10, 2010. 4 p. De Silva, D. and Yamao, M. (2007) Effects of the tsunami on fisheries and coastal livelihood: a case study of tsunamiravaged southern Sri Lanka. Disasters 31 (4): 386-404. De Silva, K.M. (1981) The V.O.C in Sri Lanka: The Last Phase 1767-1796. Pp 609. In A History of Sri Lanka. C. Hurst & Company Ltd. Engvall, L.O., Ratcliffe, C. and Scheepens, T.J. (1977) Assessment of problems and needs in marine small-scale fisheries. Ministry of Fisheries. Development of small-scale fisheries in Southwest Asia. ftp://ftp.fao.org/docrep/fao/006/ad739e/ad739e08.pdf [Accessed: August 24, 2010]. Colombo, p1-9 p. FAO (2000) Conversion factors - landed weight to live weight. FAO Fisheries Circular No.847, Revision no. 1, Rome, 176 p. FAO (2006) Fishery Country Profile- Sri Lanka. fao.org/fi/oldsite/FCP/en/LKA/profile.htm. Accessed: June 24, 2010. FAO (2011) FAO major fishing areas. FAO Fisheries and Aquaculture Department. http://www.fao.org/fishery/area/search/en Accessed: Jan 24, 2011. ITDG (2005) Rebuilding fisheries livelihoods in Sri Lanka post-tsunami. Intermediate Technology Development Group (ITDG). http://practicalaction.org/docs/region_south_asia/rebuilding-fisheries-livelihoods(draft).pdf. Accessed: Feb 14, 2011. Jayasuriya, S., Steele, P., Weerakoon, D., Knight-John, M. and Arunatilake, N. (2005) Post-tsunami recovery: issues and challenges in Sri Lanka. The Institute of Policy Studies (IPS) of Sri Lanka and the Asian Development Bank Institute (ADBI) and the Asian Economics Centre, University of Melbourne. http://www.ips.lk/news/newsarchive/2005/01122005_p_tsun/tsunami_recovery.pdf. Accessed: Feb 14, 2011. Jayawardane, P.A.A.T., Mclusky, D.S. and Tytler, P. (2003) Population bynamics of Metapenaeus dobsoni from the western coastal waters of Sri Lanka. Fisheries Management and Ecology 10: 179-189. Kelleher, K. (2005) Discards in the world's marine fisheries. FAO Fisheries Technical Paper 31-32, 111 p. Maldeniya, R. (1997a) The coastal fisheries of Sri Lanka: resources, exploitation, and managment. In Silvestre, G. and Pauly, D., (eds.), Status and Management of Tropical Coastal Fisheries in Asia. Asian Development Bank, Makati CIty. Maldeniya, R. (1997b) Small boat tuna longline fishery north-west coast of Sri Lanka. National Aquatic Resources Agency (NARA), Colombo. ftp://ftp.fao.org/fi/CDrom/IOTC_Proceedings(19992002)/files/proceedings/miscellaneous/ec/1996/EC602-04.pdf. Accessed: Jan 27, 2011. Maldeniya, R., Dayaratne, P. and Amarasooriya, P.D.K.D. (1996) An analysis of billfish landings in the pelagic fisheries in Sri Lanka. National Aquatic Resources Agency, Colombo. ftp://ftp.fao.org/fi/CDrom/IOTC_Proceedings(19992002)/files/proceedings/miscellaneous/ec/1996/EC602-31.pdf. Accessed: Jan 10, 2010. Mallikage, M. (2001) The effect of different cooling system on quality of pelagic species. Department of Fisheries and Aquatic Resources, Colombo, 34 p. MFAR (2008) Major marine fish types by commercial group. Ministry of Fisheries and Aquatic Resources. http://www.fisheries.gov.lk/Data/Fish%20Types.pdf. Accessed: January 27, 2011. Pathirana, W. (1972) Administration report of the acting director of fisheries for 1969-1970. Ceylon Government Press, Colombo. 146 p. Pearson, J. (1911) Report of the marine biologist for 1910-11. Ceylon administration reports: part IV- education, science and art., Colombo. Pearson, J. (1922) Report of the government marine biologist for 1922. Ceylon administration reports, Colombo. RAPA (1989) Marine fishery production in the Asia-Pacific Region- Sri Lanka. Regional office for Asia and the Pacific (RAPA) Food and Agricultural Organization (FAO) of the United Nations, Bangkok. Saila, S.B. (1983) Importance and assessment of discards in commercial fisheries. FAO Fisheries Circular 765: 62. Silucaithsam, A.S. and Stokke, K. (2006) Fisheries under fire: Impacts of war and challenges of reconstruction and development in Jaffina fisheries, Sri Lanka. Norsk Geografisk Tidsskrift- Norwegian Journal fo Geography 60: 240-248.  Sivalingam, S. (1961) The 1958 Pearl Oyster Fishery, Gulf of Mannar. Ceylon Fisheries Research Bulletin No 11. 28 p. Sivasubramaniam, K. (1997) One hundred years of fisheries management in Sri Lanka: lessons for the future. The department of Fisheries and Aquatic Resources, Colombo. 156p. Subasinghe, S. (1981) Fish By-catch...Bonus From the Sea- Sri Lanka. 141-142. UNEP (2001) Sri Lanka state of the environment report: coastal resources. United Nations Environment Programme. http://www.rrcap.unep.org/pub/soe/srilankasoe.cfm. Accessed: Nov 3rd, 2010. Zeller, D., Booth, S., Craig, P. and Pauly, D. (2006) Reconstruction of coral reef fisheries catches in American Samoa, 1950-2002. Coral Reefs 25: 144-152. Zeller, D., Booth, S., Gerald, D. and Pauly, D. (2007) Re-estimation of small-scale fishery catches for U.S. flagassociated island areas in the western Pacific: the last 50 years. Fishery Bulletin 105: 266-267.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  95  Appendix Table 1. FAO reported landings vs. total reconstructed catch for Sri Lanka, 1950-2008, in metric tonnes. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  FAO landings (t) 20,622 24,103 24,709 25,016 26,433 27,265 32,702 29,638 35,737 40,434 50,775 59,717 71,137 71,256 87,796 68,836 72,083 78,225 93,080 91,936 83,855 69,074 85,438 93,972 96,608 99,110 122,870 126,000 142,768 150,934 167,594 179,398 184,664 183,005 137,909 158,065 159,437 164,998 175,347 183,773 159,173 193,989 189,939 219,447 240,307 222,170 242,031 248,790 259,746 288,301 284,314 265,749 271,927 302,082 303,168 160,142 239,292 270,176 285,028  Total reconstructed catch (t) 151,813 156,210 158,421 160,496 166,998 166,716 176,616 172,670 185,640 189,190 198,438 206,182 227,544 222,926 259,676 222,888 230,049 238,398 245,640 255,295 242,466 224,345 243,631 247,627 264,940 273,005 302,695 295,302 319,436 310,821 320,543 348,908 374,713 352,116 302,847 298,884 298,798 307,197 320,621 331,133 305,788 349,006 360,916 390,223 381,993 399,668 411,686 419,199 443,709 449,153 483,307 428,117 432,235 524,880 533,482 251,821 432,512 468,803 503,501  Appendix Table 2. Total reconstructed catch (t) by major taxa for Sri Lanka, 1950-2008. Clupeoids include herrings, sardines, and anchovies. Others includes 86 taxa Year  1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008  Leiognathidae  10,457 10,511 11,125 11,748 14,638 13,537 16,390 15,303 19,852 18,749 19,200 18,094 24,895 21,509 35,647 23,203 26,023 27,706 22,596 30,482 27,612 25,861 28,621 26,281 37,502 42,447 47,751 41,644 48,182 37,957 34,769 47,292 63,048 49,864 48,481 33,183 33,794 34,848 36,061 36,590 35,154 40,141 50,504 49,542 26,277 48,498 42,941 42,681 50,824 35,840 59,392 35,461 33,768 79,929 83,900 21,620 55,219 59,779 73,544  Clupeoids  8,000 8,000 9,000 7,100 9,500 11,600 18,300 14,100 14,900 17,000 20,400 28,400 31,500 22,400 36,300 24,200 24,300 20,000 27,500 21,700 22,300 18,400 20,100 20,600 24,900 32,530 38,541 46,278 54,412 59,276 69,061 64,479 66,764 70,971 52,153 27,682 28,471 29,460 30,608 31,064 27,958 33,426 35,097 37,379 38,870 49,785 48,221 47,200 50,800 51,370 53,250 49,270 52,310 56,390 54,410 24,870 56,230 63,520 66,890  Katsuwonus pelamis  771 890 807 723 720 717 981 1,245 1,410 1,576 2,063 2,551 3,960 5,369 5,227 5,084 5,830 6,576 7,448 8,322 6,554 4,785 8,250 9,919 8,792 6,937 12,392 11,583 12,933 9,692 14,117 15,196 14,172 14,649 12,348 13,699 13,697 14,442 15,004 16,500 19,495 21,990 25,786 29,692 35,755 33,915 41,000 50,012 50,124 64,316 70,957 66,692 64,425 75,146 75,795 44,938 54,341 73,240 78,860  Clupeidae  12,715 12,807 12,887 12,983 13,087 13,137 13,180 13,253 13,294 13,341 13,151 13,192 13,217 13,228 13,230 13,224 13,212 13,190 13,156 13,105 13,034 12,942 12,830 12,700 12,552 12,387 12,207 12,010 11,795 11,557 11,296 11,151 10,935 10,705 10,465 10,217 9,962 10,108 10,255 10,400 10,543 10,683 10,820 10,951 11,075 11,190 11,296 11,393 11,481 11,562 11,636 11,698 11,786 11,954 12,146 6,489 12,433 12,524 12,618  Thunnus albacares  774 1,150 903 655 606 557 720 883 970 1,055 1,347 1,639 2,493 3,348 3,222 3,096 3,515 3,931 4,416 4,901 3,841 2,783 4,266 5,244 4,610 3,771 6,908 5,806 5,915 6,555 7,304 8,068 8,682 9,264 6,694 7,160 7,416 7,785 8,089 8,727 9,929 11,934 14,185 16,478 21,045 16,499 21,308 27,094 26,122 32,767 29,512 26,522 28,085 34,425 35,512 24,887 35,842 32,998 33,027  Carangidae  Scomber  Others  spp.  2,543 2,561 2,577 3,697 4,717 4,027 3,436 3,351 4,859 5,968 7,030 6,838 8,443 9,846 10,146 6,545 6,342 8,638 8,631 8,621 9,007 7,588 7,566 8,440 11,610 8,637 10,076 11,192 12,717 12,440 12,307 12,796 10,579 10,726 12,594 10,139 10,319 10,638 11,003 11,165 9,831 11,112 11,112 13,068 10,215 9,148 8,347 9,179 10,796 10,992 12,777 12,290 13,117 17,331 16,009 7,248 12,057 13,885 13,684  0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3,000 3,000 3,000 3,000 4,000 3,600 5,100 6,200 4,900 4,300 7,994 11,018 9,179 7,747 13,388 13,888 12,906 11,302 15,518 12,773 13,000 13,000 13,000 13,000 13,000 10,500 12,000 13,557 10,854 16,450 17,642 17,700 20,000 20,900 21,350 22,180 16,760 17,250 17,760 18,440 9,680 15,570 16,290 18,260  116,553 120,290 121,122 123,591 123,730 123,140 123,609 124,536 130,356 131,502 135,247 135,467 143,036 147,226 155,905 147,536 150,828 158,357 161,893 168,164 160,118 151,986 161,997 164,443 164,974 166,294 174,819 166,788 173,482 173,344 171,689 189,926 200,533 185,937 160,112 196,803 195,139 199,916 209,601 216,687 192,878 219,721 213,413 233,113 238,756 230,633 238,573 231,641 243,562 242,307 245,783 226,183 228,743 249,705 255,711 121,769 206,391 212,857 224,878  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  97  FROM LOCAL TO GLOBAL: A CATCH RECONSTRUCTION OF TAIWAN‘S FISHERIES FROM 1950-20071 Daniel Kuo and Shawn Booth  Sea Around Us Project, Fisheries Centre University of British Columbia, 2202 Main Mall, Vancouver, BC, V6T 1Z4, Canada danieltykuo@gmail.com; s.booth@fisheries.ubc.ca  ABSTRACT Taiwan‘s coastal fisheries were considered overfished in the 1950s, and, as a result, Taiwan expanded its fisheries into offshore areas and to distant-waters beginning in 1959. Taiwan‘s fisheries catches within its own EEZ increased until 1980, and total catches were estimated to grow from approximately 95,000 tonnes in 1950 to 575,000 tonnes in 1980. However, since 1980, total catches taken by Taiwan within its own EEZ have been in decline and were estimated to be only 290,000 tonnes in 2007. Although the catches taken within the EEZ have been in decline, the distant-water fisheries (DWF) have expanded into all major oceans. By the 2000s, they were 3 to 5 times larger than those from its own EEZ. Total catches of the distant-water fleet were estimated to increase from 110,000 tonnes in 1959 to 1.5 million tonnes in 2007. However, the continued expansion of the DWF is not likely sustainable due to the increased pressure on fisheries stocks globally, and because of the large amount of fuel subsidies the DWF receives.  INTRODUCTION Taiwan (―Republic of China‖) is an island country located off the southeast coast of the People‘s Republic of China in the South China Sea (Figure 1). The west coast of Taiwan is separated from China by the relatively shallow Taiwan Strait and its east coast lies along the much deeper Philippines Sea. Taiwan has always relied on the sea for resources, and its fisheries are well developed and provide an essential source of food and income for its people. The population of Taiwan grew from approximately 7.5 million people in 1950 to approximately 23 million people in 2007 (Anon., 2009). The increasing population, and its seafood demand, the decline of near-shore resources, and the high level of pollution of inshore areas, have resulted in Taiwan‘s fisheries expanding into offshore waters and eventually to distant waters. Taiwan‘s diplomatic isolation has often caused problems during negotiations with neighbouring countries resulting in territorial disputes. Consequently, Figure 1. Map of Taiwan and the exclusive economic zone (EEZ) it Taiwan‘s exclusive economic zone (EEZ) is claims. highly disputed and frequently challenged (Figure 1; Chen, 2007). 1  Cite as: Kuo, D. and Booth, S. (2011) From local to global: a catch reconstruction of Taiwan‘s fisheries from 1950-2007. pp. 97-106. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727].  It is widely suspected that Taiwan, like a number of other countries, tolerates extensive Illegal, Unreported and Unregulated (IUU) fisheries, and thus the annual catches reported to the Food and Agriculture Organization (FAO) are minimum estimates that may omit significant catches. This report aims to provide a more accurate estimate of Taiwan‘s annual fisheries catches from 1950–2007 by including estimates of unreported catches and discards. Some catches, which contribute to the local or global supply of seafood, are often unreported for tax avoidance purposes, and are usually not included in assessments or quota considerations, which undermine fisheries management policies. Discards increase economic benefits for fishers by increasing profits, and are a form of fishing mortality that may not be accounted for, and therefore may lead to detrimental effects on exploited fish populations.  EEZ fisheries Taiwan‘s fisheries have traditionally been considered as two distinct fisheries—the coastal and the ―distant-water fisheries‖. The coastal fisheries operate up to the 12 nm territorial sea boundary, whereas the distant-water fisheries operated past the 12 nm demarcation of the coastal fisheries. However, Taiwan‘s definition of ‗distant-water‘ changed with the expansion of the fisheries into all of the major oceans beginning in 1959, using gears such as trawl, longline, and purse seine. Thus, the distant-water fisheries were re-categorized as those operating in the far seas, and the fisheries around the waters of Taiwan were re-defined as ‗coastal‘ and ‗offshore‘. Throughout this report we refer to the coastal and offshore fisheries as those operating in what is today considered to be within the Taiwanese exclusive economic zone (EEZ), and reserve the term distant-water fleet (DWF) for those fisheries operating outside the EEZ. As an island country, Taiwan has extensively developed its fisheries for food security, but it has also used its fisheries to gain political influences. After World War II, Taiwan was faced with a growing population, food shortages, and a heavily damaged fishing industry (Huang and Chuang, 2010). During that time, Taiwan‘s inshore fishing grounds were already considered to be overfished, and in order to meet domestic demand, Taiwan began to expand its fisheries into distant waters because of the declining fisheries resources within its local waters (Chen, 2007).  Distant water fisheries The growth of the post-war Taiwanese distant-water fishery after WWII was dictated by urgent economical and political needs. During the post-war period, there was a push towards an ―Every Fisher Has His Boat‖ policy. As part of this policy, the government offered subsidies to encourage the purchase of fishing boats and to stimulate the launching of more boats (Chang et al., 2010). The government also attracted investors for the DWF with the support of the United States Agency for International Development (USAID) and the World Bank during 1953 – 1973 (Chang et al., 2010). Beginning in 1961, investments were directed towards Taiwan‘s DWF to support the production of large-scale longliners and trawlers (Chang et al., 2010). At the same time, the construction of small trawlers was suspended in 1967 because of the drastic increase of fishing boats in the coastal region and depletion of coastal fisheries resources. During the mid-1970s to early 1980s, the DWF was affected by two international oil crises, which increased the costs of operation and resulted in decreased landings in some years (Chang et al., 2010). Also, many coastal nations declared EEZs starting in the late 1970s or early 1980s, and Taiwan‘s offshore fleets were prevented access to former areas of operation (Chen, 2007). Consequently, Taiwan began to focus on further expansions of the DWF resulting in an increase in landings from this fleet, while at the same time there was a decrease in landings from the offshore area of the Taiwanese EEZ. From the mid-1980s to 1990s, the Taiwanese government continued to push for the growth of the DWF (Chang et al., 2010). In 1983, a policy was implemented to suspend the building of all small-scale fishing vessels (< 1000 GRT) in order to develop the large-scale distant-water vessels (Chang et al., 2010), i.e., high seas capable vessels. Beginning in the 1980s, the Taiwanese DWF also explored new gear types such as the squid drift net and tuna purse seine. The main fishery gear types consisted of tuna long line, purse seine, squid jig, saury stick held dip net, trawl, and large-scale tuna drift net fisheries (Chang et al., 2010). The main species targeted during this time were yellowfin tuna (Thunnus albacares), bigeye tuna (Thunnus obesus), albacore (Thunnus alalunga), skipjack (Katsuwonus pelamis), and argentine squid  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  99  (Illex argentinus) (Chang et al., 2010). The massive deployment of drift nets, however, was short-lived as drift net fishing was banned by the UN in 1992 (Chen, 2007). From the mid-2000s to the present, due to increasing problems of IUU fishing and vessels operating under flags of convenience, the Taiwanese longline fishery was sanctioned by the International Commission for the Conservation of Atlantic Tunas (ICCAT) in 2005. This was devastating to the Taiwanese distant-water fisheries, since longline vessels targeting bigeye tuna had their quota substantially reduced and Taiwan also lost its Cooperating Status in ICCAT. The sanctions imposed by ICCAT can, however, also be seen as an opportunity for Taiwan to reform its fisheries policies (Chang et al., 2010).  Discards Discarding is a common practice within the fishing industry, and unwanted fish that have been caught are simply thrown back to sea. Discards can include damaged fish, non-target species, spoiled fish, and undamaged target species that are discarded to make room for more valuable ones (i.e., ‗high grading‘). Much of the global discards currently are influenced by economic incentives, especially in DWFs. Depending on the markets of the fishing country, catches may be fully retained or largely discarded. For Taiwan, only a rough estimate of discards is possible, as neither the FAO nor the Taiwanese Fisheries Agency (FA) documents discards.  MATERIALS AND METHODS FAO reports, on behalf of Taiwan, annual landings statistics through their fisheries capture database (Fishstat) since 1950. Internally, Taiwan has two sets of data relating to fisheries. National landing statistics by area and gear are available from 1959 to 2007 through Taiwan‘s Fisheries Agency (FA; www.fa.gov.tw) and a separate set of statistics are available from the Taiwanese Council of Agriculture, Executive Yuan that describe the food balance for the country from 1984 to 2007 (www.ttdais.gov.tw/view.php?catid=20242). The food balance statistics include the human population, domestic production for fish and seafood, imports, exports, and domestic supply, and are expressed in live weight equivalents. The sub-categories under the domestic production for fish and seafood include ‗fish‘, ‗shrimps and crabs‘, cephalopods, ‗shell fish‘, ‗others‘, and ‗dried (salted) fish‘. A comparison between the national landings and FAO landings was conducted to determine the quality of data transfer from the Fisheries Agency (FA) to FAO. Freshwater fishes, corals, marine mammals, and seaweeds were removed from FAO landing statistics for the purpose of this comparison. Similarly, we excluded the freshwater fisheries and all aquaculture statistics from the FA national landings. The data from FA were also compared to the food balance data for the period 1984 – 2007 when published statistics are available. The domestic production from the food balance data was converted into estimates of annual national catches by the following formula: Catch = Domestic Production + Export - Import - Marine Culture - Freshwater Culture - Inland Catch  ...1)  The marine culture, freshwater culture, and inland catch statistics were taken from the FA‘s national landing statistics. The annual catch determined from equation (1) was compared to the annual FAO and FA total landings to check for inter-agency discrepancies.  Unreported catches We assume that discrepancies found between the food balance data and that supplied to the FAO via the FA represent unreported catches. For the years 1954–1983, when there are no data available from the FA, we use the average rate of unreported catches (~33%) reported from 1951 to 1953 (Sung, 1972) to estimate unreported catches. We further assume that the unreported catches are proportional to the spatially reported landings. Unreported catch rates were applied to the reported landings from the coastal, offshore and distant water fisheries. The unreported catches for each FAO area are distributed in proportion to the reported taxa, except in FAO area 61 and 71. For FAO 61, the ―sharks, rays, skate, etc, nei‖ category were  distributed in proportion to the species reported in Lawson (1997). The unreported catch rates for FAO area 71 were applied in proportion to the species reported for the Taiwanese distant water longline catch (Lawson, 1997).  Discards The discard portion of the Taiwanese catch reconstruction is based on estimates, since no official records of Taiwanese discards exist. Discard rates were applied to the sum of the reported and unreported landings. Longline discard rates were based on data presented in Kelleher (2005; Table 1). The average rates of longline discards calculated from this publication were applied according to their reported areas and year. The discard rate of purse seine, trawl, squid jig, and nets are rates provided for other countries operating in the same area (Kelleher, 2005). We used reported discards by gear and target species as a proportion of retained catches by gear and target species to derive discard rates. We applied these derived discard rates to the sum of the reported and unreported landings to estimate discard amounts. Table 1. Summary table of discard rates used for the various oceans and gear types of the Taiwan fisheries fleets. Year Ocean Gear Discard rate Source 1950-1997 Pacific longline 0.113 Mejuto et al. (1997) 1998-2007 Pacific longline 0.124 Mejuto et al. (1997) and Kelleher (2005) 1950-1994 Atlantic longline 0.160 Cramer 1998 and Mejuto et al. (1997) 1995-2004 Atlantic longline 0.400 Kelleher (2005) 2005-2007 Atlantic longline 0.130 Mejuto et al. (1997) 1950-1989 Indian longline 0.340 ½ the 1990-1999 rate 1990-1999 Indian longline 0.680 Mejuto et al. (1997) 2000-2004 Indian longline 0.460 Mejuto et al. (1997) 2005-2007 Indian longline 0.390 Mejuto et al. (1997) 1950-2007 all purse seine 0.040 Amande et al. (2008) and Anon. (2003) 1950-2007 all shrimp trawl 0.970 Clucas (1997) 1950-2007 all demersal trawl 0.123 Kelleher (2005) 1950-2007 all squid jig 0.500 Kelleher (2005) 1950-2007 all other nets 0.014 Kelleher (2005)  An additional 279 t∙year-1 of whale shark catches are applied from 1990 to 2000, and 126 t∙year-1 applied from 2001 to 2007 to the offshore catches due to the unregulated nature of the whale shark fisheries (Chen et al., 1997; Chen and Phipps, 2002). Whale sharks are usually sold outside of the market system to avoid fees, and only two markets have recorded landings statistics for whale sharks (Chen et al., 1997).  RESULTS The total reconstructed catches amount to 54.7 million tonnes from 1950-2007, as  Catch (x 103 t)  The discard rates thus derived were applied spatially to each major fishery based on gear type. The longline and purse seine discards were assigned to both the EEZ and distant water fisheries. Discards associated with squid jigging in the Pacific and net discards (excluding purse seines), which are almost exclusively used in coastal and offshore areas, were assigned to EEZ fisheries only. The longline and purse seine fisheries within the EEZ had discard rates applied in proportion to the reported landings by the two gear types using the FA‘s national statistics. For the DWF, the longline and purse seine discards were also assigned proportionally, based on reported landings by the two gear types. Discards from the trawl fishery and the squid fisheries in oceans other than the Pacific were assigned to the DWF. 2000 1800 1600 1400 1200 1000 800 600 400 200 0 1950  Discards Unreported  Reported  1960  1970  1980  1990  2000  Year Figure 2. Total reconstructed catches including reported landings, unreported catches and discards.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  The FAO landings statistics and the FA national landings statistics are largely indistinguishable (Figure 3), indicating that the data transfer between FA and the FAO was efficient. However, discrepancies in landings were found between 1950 and 1953, and from 1984 onwards using alternative sources. Catches for Taiwan from 1950 – 1953 were, on average, about 28% higher than the national FA statistics (Sung, 1972) and the discrepancies increased from approximately 11% in 1950 to 38% in 1953. Discrepancies between the food balance data and the FA data for the later time period (1984-2007) averaged 37%, and also showed an increasing trend through time (Figure 4). Over the entire period (1950-2007), unreported catches estimated from the discrepancies between the food balance data and the FA amount to 13.6 million tonnes. Although the yearly discrepancies were relatively small during the first 4 years (1950 -1953 average: ~25,000 t∙year-1), after 1983, these differences increased over time and range from 225,000 t∙year-1 to 567,000 t∙year-1.  1200  Landings (x 103 t)  Total unreported catches  1400  800  FAO  600  Taiwan Fisheries Agency 400  0 1950  1960  1970  1980  1990  2000  Year Figure 3. National statistics as reported by FAO and by the Fisheries Agency of Taiwan. 70  60 50 40 30 20  10 0 1950  1960  1970  1980  1990  2000  Year  Figure 4. Discrepancies (%) between the food balance data and reported landings. Dots are reported values and the solid line indicates an estimated value (mean of 1951-1953). 600  Total discards  Discards  Catch (x 103 t)  Our estimated total discard amounts were fairly constant from 1950 to 1959 and averaged 12,000 t∙year-1. However, with the expansion of the DWFs, discards increased and peaked at 80,000 tonnes in 1988, representing 5% of the total reconstructed catch. Total discards declined to approximately 47,000 tonnes in 2007 and represent 3% of the total reconstructed catches. From 1959 onwards, the average discard rate applied to reported landings and unreported catches was 5.3%. We estimate that between 1959 and 2007, the highest rate of 8% occurred in 1972, but has since dropped to 3% in 2007.  1000  200  Discrepancy (%)  compared to the landings reported to FAO of approximately 38.2 million tonnes. The difference of 16.5 million tonnes is estimates of IUU in the form of unreported catches and discards. Unreported catches are the dominant form of IUU accounting for 84% of the IUU catches assessed here (Figure 2).  101  500  Offshore unreported  400  Coastal unreported  300  Offshore  200 100 0 1950  Coastal 1960  1970  1980  1990  2000  Year Figure 5. Catches taken within the EEZ waters of Taiwan, including coastal (with catches considered here to be taken by the small-scale sector) and offshore fisheries, unreported catches and discards.  EEZ fisheries Fisheries in the EEZ of Taiwan, which consist of the coastal and offshore sectors, increased from the 1950s until the 1970s by approximately 100,000 t∙decade -1(Figure 5). Reported landings increased from 75,000 t∙year-1 in 1950 to 401,000 t∙year-1 in 1979. Annual catches within the EEZ peaked in 1980 at 407,000 t∙year-1, but have steadily declined and were only 190,000 t∙year -1 in 2007. The declines in landings from the EEZ waters have occurred in the offshore waters. Coastal fisheries, relying mostly on fixed nets, have been subject to overfishing since the 1950s (Chen, 2007) and apparently may also have been affected by the expansion of aquaculture. Coastal fisheries, roughly corresponding to small-scale fisheries, averaged reported landings of approximately 31,000 t∙year-1 between 1950 and 1979, and increased to average 47,000 t∙year-1 between 1980 and 2007 (Figure 5). Offshore reported landings increased from 55,000 tonnes in 1950, peaked in 1980 at about 370,000 tonnes, but have since declined to 135,000 tonnes in 2007 (Figure 5).  Catch (x 103 t)  From 1950 to 2007, unreported catches 1600 within the EEZ were estimated to be over 5 1400 million tonnes (Figure 5). Unreported catches increased from 8,400 tonnes in 1200 1950, peaked at about 136,000 tonnes in 1000 Discards 1980, and declined to 92,000 tonnes in 2007. Discards within the EEZ were 800 estimated to be highest in the first decade Unreported 600 due to the amount of trawling. Trawl Reported landings declined from 89% to 33% of 400 reported landings between 1950 and 1959, 200 but due to the increased landings trawl discards were estimated to average 12,000 0 t∙year-1 between 1950 and 1958. With the 1950 1960 1970 1980 1990 2000 expansion of fleets into distant waters Year beginning in 1959, we estimate that the Figure 6. Total reconstructed catches for the DWF, including discards within the EEZ waters declined, but reported landings, unreported catches and discards. those from DWF began to increase. Discards were estimated to average approximately 21,000 t∙year-1 from 1959-2007, but increased from 7,600 tonnes in 1959, peaked during the 1980s at 33,000 t∙year-1 before declining to 8,000 t∙year-1 in 2007 (Figure 5).  Distant water fisheries Ine more recent times, distant-water fisheries represent the most significant component of Taiwanese fisheries. The DWF accounts for approximately 62% of the total reported Taiwanese landings from 1950 to 2007, and mainly targets tunas and squids using longline, purse seine and jig. Since 1985, the DWF have consistently been the largest, contributor to Taiwan‘s reported total landings. With decreased access to other neighbouring countries waters starting in the late-1970s due to the increasing declaration of EEZs, the DWF was rapidly developed in the 1980s, and reported landings from these fisheries during the 2000s are now 3 to 5 times larger than those from the EEZ in terms of annual reported landings (Figure 6). Unreported catches from the DWF were estimated to contribute over 8.7 million tonnes to the total catches from 1959-2007 (Figure 6). Unreported catches increased from 26,000 tonnes in 1959 to 97,000 tonnes in 1971. Between 1972 and 1987, unreported catches were between 100,000 and 200,000 tonnes, and averaged 120,000 t∙year-1. Unreported catches have increased from 253,000 tonnes in 1988 to 477,000 tonnes in 2007. Discards from the DWF were estimated to have increased from 5,000 tonnes in 1959, peaked at 53,000 tonnes in 2002, but have since declined to approximately 39,000 tonnes in 2007.  DISCUSSION From 1950-2007, the total reported landings were over 38 million tonnes. Reported landings increased from about 75,000 tonnes in 1950 to over 900,000 tonnes in 1987. Since 1988, annual reported landings have been approximately 1 million t∙year-1. Unreported catches from 1950-2007 were 13.7 million tonnes,  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  103  and increased from about 8,000 tonnes in 1950 to 568,000 tonnes in 2007. Discards rose from 10,000 tonnes in 1950 to 80,000 tonnes in 1988 and then began to decline, amounting to approximately 47,000 tonnes in 2007. As estimated here, our total reconstructed catches including reported landings, unreported catches and discards amount to 54.7 million tonnes and are 43% larger than reported landings. This estimation is probably conservative. While the unregulated portion of IUU catches in Taiwan may be small, the unreported illegal portion may be large, leaving out a potentially significant portion of IUU catches. The unreported catches that we estimated, based on the food balance data, may represent unreported bycatch from the EEZ fisheries and DWF. With the exception of FAO fisheries statistical area 61, the only species reported from other fisheries statistical areas were tunas, marlins, squids, unidentified sharks, and unidentified marine fishes (―marine fishes nei‖). The ―marine fishes nei‖ category accounts for a small proportion of the overall catch statistics in FAO, and thus would be an underestimate because it doesn‘t account for all bycatch. It was not until 1991 that the government enforced an observer and vessel monitoring program to track fisheries catches (Chang et al., 2010). However, the fact that most of the reported DWF landings are high-value target species (e.g., tunas) makes it possible that the estimates of unreported catches reported here are mainly comprised of bycatch species. The expansion of Taiwan‘s fishing fleet into offshore areas of the EEZ beginning in 1959 increased annual landings until many of Taiwan‘s neighbours declared EEZs, increasingly limiting the fishing areas accessible by Taiwan. The loss in catches from traditional fishing grounds caused a fairly rapid expansion of the DWF in the early years, but has since slowed. As estimated here, the total reconstructed DWF catches doubled from 113,000 tonnes to 251,000 tonnes in 8 years (1959-1966), doubled again from 251,000 tonnes to 500,000 tonnes in 7 years (1966-1972), but it took 17 years for DWF catches to double from 500,000 to 1 million tonnes (1972-1988). Total reconstructed DWF catches were estimated to be 1.5 million tonnes in 2007. Taiwan‘s fisheries contribute to national food supply, and presently the DWF catches compensate for the declining catches from the Taiwanese EEZ. However, the DWFs are highly vulnerable to fluctuations in the energy sector in the form of rising fuel costs. Decreased catches were the result of the oil crisis in the 1970s and again in the 1980s. Fuel used by the DWF is subsidized by the Taiwanese government, and in the year 2000 the $US 0.09∙litre-1 subsidy amounted to $US 120 million (Sumaila et al., 2008). With increased pressure on fisheries stocks worldwide, this strategy is not likely to be sustainable, and the Taiwan DWF may thus experience difficulties expanding, or even maintaining their present catch volumes.  ACKNOWLEDGEMENTS This is a contribution of the Sea Around Us Project, a scientific collaboration between the University of British Columbia and the Pew Environment Group.  REFERENCES Amande, J.M., Ariz, J., Chassot, E., Chavance, P., Delgado, M.A. de, Gaertner, D., Murua, H., Pianet, R., and Ruiz, J. (2008). By-catch an d discards of the European purse seine tuna fishery in the Indian Ocean. Estimation and characteristics for the 2003-2007 period. IOTC-2008-WPEB-12. Ecosystem and by-catch wroking group. Bangkok, Thailand. 20-22 October 2008. 26, p. Anonymous (2003). Inter-American Tropical Tuna Commission Fishery Status Report No. 1. Tunas and billfishes in the eastern Pacific Ocean in 2002. 96, p. Anonymous (2009) Total Population, annual increase, and numbers and rates of natural increase, births and deaths. Department of Household Registration, M.O.I. Available at: www.ris.gov.tw/web_eng/eng_sta_hs.html [Accessed: August 2010]. Chang, S.-K., Liu, K.-Y. and Song, Y.-H. (2010) Distant water fisheries development and vessel monitoring system implementation in Taiwan--history and driving forces. Marine Policy 34: 541-548. Chen, C., Liu, K. and Joung, S. (1997) Preliminary Report on Taiwan's Whale Shark Fishery. TRAFFIC Bulletin 17: 5357. Chen, T. (2007) Taiwanese Offshore (Distant Water) Fisheries in Southeast Asia, 1936 - 1977. Doctor of Philosophy Dissertation, Murdoch University, Perth, Australia 368 p.  Chen, V. and Phipps, M. (2002) Management and Trade of Whale Sharks in Taiwan, TRAFFIC East Asia Report. Taipei, 1-25 p. Clucas, I. (1997) A study of the options for utilization of bycatch and discards from marine capture fisehries. FAO Fisheries Circular. No. 928. FAO, Rome, 59, p. Cramer, J. (1998) Pelagic longline billfish by-catch. Col. Vol. Sci. Pap. ICCAT, 47: 143-153. Huang, H.-W. and Chuang, C.-T. (2010) Fishing capacity management in Taiwan: experiences and prospects. Marine Policy 34: 70-76. Kelleher, K. (2005) Discards in the world's marine fisheries. An update. FAO Fisheries Technical Paper. No. 470. FAO, Rome, 131 p. Lawson, T. (1997) Estimation of Bycatch and Discards in Central and Western Pacific tuna Fisheries: Preliminary Results. South Pacific Commission, Nouméa, 12-30 p. Mejuto, García-Cortés, B., and Ramos-Cartelle, A.M. (2007) Preliminary approach to evaluate the importance of discards and other uses of billfish in the Spanish surface longline fishery carried out in differeent oceans betweeen 1993-2005. Col. Vol. Sci. Pap. ICCAT, 60(5): 1547-1554. Sumaila, U.R., Teh, L., Watson, R., and Pauly, D. (2008) Fuel price increase, subsidies, overcapacity, and resource sustainability. ICES Journal of Marine Science 65: 832-840. Sung, T.A. (1972) Taiwan Distant Waters Fishery [Partial photocopy of an internal report provided in confidence; in Chinese].  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  105  APPENDIX Appendix Table A1. Total Taiwanese catches (t) including coastal, offshore and DWF catches, and estimates of unreported catches and discards. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  Coastal  Coastal unreported  20,885 25,291 29,696 30,250 30,804 31,358 31,911 32,465 33,019 33,573 31,761 34,525 35,214 40,489 35,597 33,688 28,615 29,158 27,985 29,254 30,338 30,515 27,383 26,130 25,565 29,540 32,732 33,079 32,024 36,380 35,645 37,615 39,445 45,257 49,650 52,817 55,087 52,255 47,439 47,594 46,162 39,031 43,201 40,540 43,496 41,033 40,576 43,609 39,911 39,911 44,016 51,869 49,669 63,739 56,290 52,956 54,381 54,280  2,313 7,293 9,991 11,372 10,276 10,460 10,645 10,830 11,015 11,199 10,595 11,517 11,747 13,506 11,875 11,238 9,545 9,727 9,335 9,759 10,120 10,179 9,134 8,716 8,528 9,854 10,919 11,035 10,683 12,136 11,891 12,548 13,158 15,097 14,840 17,175 20,169 16,425 17,205 13,784 14,669 14,170 10,892 10,573 14,822 15,238 13,202 12,306 14,689 11,527 14,701 21,669 23,090 26,983 26,822 29,166 32,809 26,206  Offshore 54,297 58,146 61,365 65,194 69,023 72,852 76,681 80,509 84,338 88,167 91,170 112,927 128,176 139,081 156,419 156,566 167,718 182,628 204,064 218,211 232,541 247,429 238,833 260,298 238,499 293,259 314,452 324,379 355,094 364,708 370,906 346,203 345,471 320,495 334,131 316,417 306,179 300,649 308,114 333,799 292,670 267,224 280,792 258,880 242,553 256,560 256,933 247,854 209,721 205,645 169,520 159,989 185,939 193,482 197,722 201,700 154,873 135,440  Offshore unreported 6,082 16,768 20,645 24,509 23,025 24,302 25,579 26,856 28,134 29,411 30,413 37,670 42,757 46,395 52,178 52,228 55,948 60,921 68,072 72,791 77,571 82,538 79,670 86,831 79,559 97,826 104,895 108,207 118,453 121,660 123,727 115,487 115,243 106,911 99,868 102,890 112,101 94,500 111,743 96,676 93,005 97,011 70,797 67,520 82,655 95,165 83,599 69,944 77,188 59,394 56,620 66,839 86,441 81,908 94,213 111,087 93,436 65,389  DWF n/a n/a n/a n/a 11,672 31,319 42,582 50,364 65,177 79,232 88,763 109,044 120,505 123,024 129,641 141,784 170,469 190,468 242,674 256,394 274,857 291,983 344,309 360,695 314,568 323,170 322,634 343,927 329,542 340,642 350,744 321,833 316,514 320,621 369,453 413,608 463,361 574,942 697,246 736,582 768,945 715,513 738,388 833,041 680,082 712,034 669,567 746,183 841,794 853,599 879,804 792,751 806,549 875,200 725,849 762,390 758,058 984,431  DWF unreported n/a n/a n/a n/a 3,894 10,447 14,205 16,801 21,742 26,430 29,610 36,375 40,198 41,039 43,246 47,297 56,865 63,537 80,952 85,528 91,687 97,400 114,855 120,321 104,934 107,803 107,625 114,728 109,929 113,632 117,002 107,358 105,583 106,953 110,425 134,494 169,650 180,716 252,868 213,331 244,357 259,755 186,173 217,269 231,753 264,423 217,860 210,573 309,825 246,535 293,857 331,188 374,956 370,506 345,863 419,989 457,343 476,622  Discards 10,091 11,743 12,185 12,547 12,085 12,164 12,256 12,277 12,578 12,594 14,811 17,981 19,403 20,981 22,057 23,757 28,230 30,319 39,559 41,646 45,363 49,168 62,451 62,519 51,703 55,819 58,122 60,116 58,025 60,494 59,177 56,717 57,166 59,447 61,066 62,358 69,506 73,149 80,051 65,891 62,047 56,332 50,352 59,005 53,388 65,588 64,938 59,152 67,427 57,921 64,585 65,149 67,028 61,342 61,952 52,969 50,032 46,619  Appendix Table A2: Total reconstructed catch (t) for Taiwan within its EEZ, 1950-2007. Others category includes 92 taxa plus miscellaneous marine fishes grouping. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  Katsuwonus pelamis  Elasmobranchii  0 0 0 0 0 0 0 0 596 481 384 1,143 2,140 1,496 1,267 839 659 453 526 536 292 577 536 1,497 1,048 1,575 1,382 1,962 3,514 2,554 2,546 1,872 2,850 5,605 5,818 12,241 12,658 13,909 16,988 26,106 38,661 32,881 39,795 49,823 65,505 81,935 91,178 58,985 76,961 66,962 68,207 70,155 86,254 62,978 84,513 69,930 72,955 60,188  7,793 10,416 12,071 13,144 11,700 11,447 12,696 12,680 12,524 10,908 10,740 11,026 12,449 11,081 11,697 12,782 14,354 16,930 21,978 22,525 23,229 23,676 21,116 20,561 24,982 38,465 37,171 35,664 32,624 29,720 36,983 30,727 34,832 30,954 31,454 39,056 32,378 30,018 22,493 26,216 32,716 32,123 30,159 23,912 18,864 21,428 19,680 18,503 15,523 19,803 17,884 14,463 16,422 23,402 21,604 21,173 21,676 16,789  Shrimps and prawns 2,105 2,436 3,180 3,216 2,178 3,095 3,088 2,765 2,685 3,167 2,960 3,315 4,111 5,996 6,665 9,164 9,982 12,624 13,748 19,928 17,167 18,323 19,034 22,678 23,286 35,359 34,538 35,432 44,258 47,963 45,550 39,428 41,449 42,896 47,243 52,032 54,105 44,525 32,489 29,816 17,966 14,713 13,361 8,394 9,647 14,481 14,788 13,199 8,075 7,909 5,373 6,334 6,128 611 7,308 12,847 6,576 4,655  Scomber japonicus 3,340 5,208 6,706 6,918 5,611 7,563 9,164 6,983 6,645 7,058 6,450 5,868 5,502 8,582 12,752 12,125 9,883 6,498 14,266 17,002 21,940 16,118 11,629 5,064 2,052 2,735 3,214 6,061 4,086 4,489 5,084 15,529 13,284 15,501 19,633 23,308 25,953 20,472 19,220 18,246 17,916 15,392 23,137 21,226 32,767 32,530 33,567 28,238 18,270 23,998 13,573 12,839 19,855 28,636 38,264 43,292 30,974 21,673  Thunnus albacares 5,567 6,510 6,706 6,918 6,327 6,132 5,250 4,870 4,856 4,652 4,454 4,649 5,273 4,803 5,069 5,490 5,888 6,575 8,737 10,178 7,123 5,607 3,266 14,061 5,581 10,106 10,357 14,470 18,880 21,237 18,631 15,626 14,742 13,784 13,244 13,454 12,573 16,236 13,743 10,007 13,211 13,192 35,331 36,165 28,406 27,565 22,632 33,512 34,680 27,530 25,092 27,285 24,292 21,259 14,237 18,771 16,678 12,604  Trichiurus lepturus 1,113 1,302 2,280 2,352 2,388 3,679 2,196 2,848 4,430 3,529 4,377 6,554 10,011 8,739 12,118 12,430 11,567 14,130 12,765 9,956 11,862 12,338 21,570 20,067 14,292 14,876 15,324 14,754 17,380 11,838 20,109 16,344 19,904 17,098 15,283 14,144 17,810 11,582 17,519 23,684 10,666 9,457 9,651 8,725 11,524 9,737 7,366 5,348 4,109 4,962 3,446 4,668 4,766 4,633 5,490 4,716 5,168 3,706  Todarodes pacificus 0 0 0 0 0 0 3,628 827 1,022 2,566 1,920 3,201 6,572 7,716 7,366 6,787 8,126 5,365 4,430 5,208 6,047 3,780 6,315 14,433 8,524 13,377 16,901 12,600 12,694 23,602 20,908 24,570 21,433 11,875 48,670 17,167 11,134 12,576 13,021 23,350 18,313 26,799 5,985 5,279 6,514 13,361 11,894 7,424 17,916 5,961 3,239 2,492 2,395 256 0 0 0 0  Saurida tumbil 1,113 1,302 2,012 2,075 2,746 8,994 8,973 9,004 6,390 8,502 10,827 9,222 10,240 14,093 10,772 15,328 17,204 15,642 15,993 15,624 15,661 15,761 13,442 13,167 11,706 14,733 16,760 17,996 13,863 18,177 15,771 10,975 12,107 10,956 9,067 7,993 9,230 6,503 7,666 4,468 3,449 2,892 3,749 3,445 2,639 4,051 4,085 1,916 1,568 1,628 1,712 1,188 864 975 1,230 1,281 1,024 818  Others 62,736 77,286 88,117 95,421 102,320 98,201 99,974 110,829 117,518 121,652 121,975 151,830 161,813 177,174 188,571 178,996 184,566 204,665 217,913 230,056 248,356 275,201 258,758 271,445 261,290 300,222 328,348 339,301 370,604 376,904 378,404 358,096 354,451 340,893 309,820 312,356 320,257 310,804 344,190 332,690 296,688 272,749 248,810 225,435 213,181 209,083 195,601 212,235 171,551 163,373 151,955 167,347 192,380 229,432 209,543 228,895 186,121 164,776  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  107  RECONSTRUCTION OF FISHERIES CATCHES FOR TOKELAU (1950-2009)1 Kyrstn Zylich, Sarah Harper, and Dirk Zeller  Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC, V6T 1Z4, Canada k.zylich@fisheries.ubc.ca; s.harper@fisheries.ubc.ca; d.zeller@fisheries.ubc.ca  ABSTRACT Total marine fisheries catches were estimated for Tokelau between 1950 and 2009. As there are no commercial fisheries in Tokelau, our estimate represents subsistence fisheries only. Subsistence catches were estimated using per capita consumption rates. These rates were either found in independent fisheries studies or estimated from dietary assessments. The subsistence catches total approximately 24,250 t over the 1950-2009 time period (399 t·year-1 in the 2000s), which is almost 4.3 times larger than the FAO reported landings. This report highlights the importance of accurate fisheries catch reporting, as it is required for proper fisheries management and, for small island countries such as Tokelau, it is essential for maintaining food security.  INTRODUCTION Tokelau, a territory of New Zealand, is comprised of three atolls: Fakaofo, Nukunonu, and Atafu. They are located approximately 300 miles north of Western Samoa at 8°-10° S and 171°-173° W (Figure 1). The land area is only 12.2 km2, with an Exclusive Economic Zone (EEZ) of 319,031 km2 (www.seaaroundus.org). The atolls are situated with Atafu in the northwest, Nukunonu in the middle, and Fakaofo in the southeast. Although each atoll consists of many small islands and islets, the majority of the population for each atoll is confined to one main island or islet (van Pel, 1958). The atolls are all closed, leaving no deep water passes for ships to enter the lagoons (van Pel, 1958). Atafu is the smallest atoll, both in land area (2.5 km2) and lagoon size (19 km2). Nukunonu is the largest with about 5.5 km2 of land area and 109 km2 for lagoon coverage. Fakaofo falls in the middle with 3 km2 of land area, 59 km2 of lagoon coverage (Ono and Addison, 2009).  Figure 1. Map of Tokelau showing the three atolls, Atafu, Nukunono, and Fakaofo, and its EEZ, as well as Swains Island, a previous fourth atoll known as Olohega.  The Tokelau chain does, geographically, also consist of a fourth atoll, now known as Swains Island (Clanton, 2008). It is located 100 miles south of Fakaofo (Bertram and Watters, 1984) (Figure 1). The island was originally settled in 1400 by the Tokelauans and was given the name Olohega (Bertram and Watters, 1984). However, the atoll was officially annexed to the United States in 1925, and now forms part of American Samoa (Bertram and Watters, 1984; Clanton, 2008). Although there are many different 1  Cite as: Zylich, K., Harper, S. and Zeller, D. (2011) Reconstruction of fisheries catches for Tokelau (1950-2009). pp. 107-117. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727]  stories of what occurred with the early foreign explorers who came across Olohega, it is consistent that the eventual annexation to the United States occurred because of the Jennings family. In 1856, an Englishman reportedly sold his perceived ―claim of ownership‖ to an American, Eli Hutchinson Jennings Sr. (Bertram and Watters, 1984; Clanton, 2008). Jennings and his wife ran a copra plantation using Tokelauan labour (Clanton, 2008). In 1917, Tokelauan workers submitted a claim to Western Samoa complaining about the working conditions on Olohega (Bertram and Watters, 1984). However, the matter was referred over to the United States as officials in Apia stated they had no authority over the Jennings family (Bertram and Watters, 1984). This led to questions of sovereignty of the island and seeing an opportunity to claim new land, the United States took possession. The island is still inhabited by a small Tokelauan population and many Tokelauans continue to consider it part of Tokelau, both historically and culturally (Bertram and Watters, 1984; Ono and Addison, 2009). Tokelau was under British colonial rule (starting in 1889) and was incorporated with the Gilbert and Ellice Islands Colony in 1916 (Bertram and Watters, 1984). The current three atolls of Tokelau were brought under New Zealand administration in 1926. However, it was not until the Tokelau Act of 1948 that New Zealand took formal control over the atolls (Adams et al., 1995; Townend, 2007). Although New Zealand has been pushing for Tokelau to become an independent self-governing country, Tokelau has resisted at every step. In 1974, the New Zealand Ministry of Foreign Affairs was said to begin the ―process of administrative decolonization‖, even though it was known that Tokelau had firmly rejected the idea of selfgovernment (Bertram and Watters, 1984). Since then, Tokelau voted in two referenda, first in 2006 and again in 2007, to determine whether or not they wish to become a self-governing nation. In both referenda they fell short of a two-thirds majority vote needed to change their political status (Hoëm, 2009). Despite the Tokelauans best efforts to resist the new social and economic structure that the New Zealand government was trying to implement, there have been some radical changes to their way of life. Although tradition is still observed on the surface, the same sense of community and cooperation is not there (Hooper, 1985). This disruption is the result of New Zealand‘s attempts to introduce new programs and social order, which are counter to the Tokelauans expressed feelings (Bertram and Watters, 1984). A major change which had wide-spread effects was the introduction of the New Zealand public service lines in 1976, which were staffed by Tokelauans (Bertram and Watters, 1984; Hooper, 1985). This took a lot of able-bodied men out of the village labour force. Before this happened, the activities of these men were coordinated by the village council. Now that they are being employed by an outside entity there is less cohesiveness between the village and council (Hooper, 1985). Although this change led to less fisher-hours per week, the new influx of cash allowed more modern fishing gear to be purchased. New aluminum dinghies (12-14 feet long with 15 to 25 horsepower outboard engines) have mostly replaced the traditional canoes used (some canoes with outboard motors are in use) (Hooper, 1985). These allow for more efficient fishing as it takes less time to manoeuvre the area and less human effort (Hooper, 1985). As a fairly isolated community, Tokelau has been and continues to be very dependent on subsistence fishing activities. Almost everyone participates in some type of fishing activity (Chapman et al., 2005). Even though ship transport has increased significantly over the years, there is still heavy reliance on locally caught fish in the diet. On Atafu, the people not only engage in subsistence fishing, but also rootcrop cultivation and fruit-tree harvesting (Ono and Addison, 2009). Previously, copra was a major commercial crop exported from the atolls and, at times, brought in decent revenues (Bertram and Watters, 1984), but not anymore (Ono and Addison, 2009). Tokelau is heavily subsidized by the New Zealand government, and beginning in 1976, many Tokelauans went to work in the public service lines with wages being paid by New Zealand (Hooper, 1985; Passfield, 1998). Tokelau does bring in a small amount of profit from the sale of handicrafts such as hats, bags, and wood carvings (Passfield, 1998; Ono and Addison, 2009). Although fishing is extremely important to Tokelauans, catch amounts are not recorded or monitored. The FAO FishStat database, which provides time series data on marine fisheries landings from 1950 to present, is based predominantly on the national statistical data supplied by its member countries. Therefore, the quality of these data depends on the accuracy of the country‘s collection methods. The objective of this study is to provide a complete time series estimate of the total marine fisheries catch of Tokelau from 1950-2009. Although there have been several studies in the past which have estimated Tokelau‘s fisheries catch for specific years, there has not been a comprehensive review of the trend over time.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  109  MATERIALS AND METHODS  Population  Estimates were only made for subsistence catches, as there are no fishing practices in Tokelau which would qualify as commercial fishing. These estimates were based on reports from several different researchers. These reports consisted 2,000 of a mixture of catch and 1,800 consumption data, as well as information about eating and fishing 1,600 habits, which were used to make 1,400 assumptions when there were no 1,200 data available. Available catch data 1,000 were used, along with human 800 population data, to estimate per 600 capita consumption rates. 400 Interpolations between data anchor 200 points were used to estimate the 0 catch rates for the entire study period 1950 1960 1970 1980 1990 2000 (1950-2009). A taxonomic breakdown was then applied, which Year included categories by species and family. Figure 2. Estimated population of Tokelau, 1950-2009.  Human population data Our estimates for total catch amounts were computed from per capita consumption rates combined with human population data. Population data were obtained from the population statistics historical demography website (www.populstat.info [accessed July 8, 2011]). For most years after 1978, no population data were available. This can be attributed to the fact that in 1975 the New Zealand Ministry of Foreign Affairs assumed control of the islands and shortly after that they reverted from annual population censuses to having one estimate every five years (Bertram and Watters, 1984). Data for the years 1991, 1996, and 2001 were obtained from official Tokelau census data2 as well as the year 2006 (Anon., 2006). The population for 2007 was obtained from Gillett (2009) as this also provided a critical anchor point for seafood consumption for the same year. Data for 2008 and 2009 were linearly interpolated between the 2007 population and a 2011 estimate from the World Factbook 3. For the remaining years when no data were available, a linear interpolation between years of known population was performed to give a complete time series of population data from 1950-2009 (Figure 2).  Subsistence fisheries Several reports indicate the absence of commercial fisheries operating in Tokelau (Dalzell et al., 1996; Passfield, 1998). All fishing activities are therefore deemed to be subsistence. Subsistence fishing in Tokelau includes fishing for personal consumption, as gifts to send overseas, and for trade within the community.  Table 1. Per capita consumption rates used to estimate total demand. Years Consumption Rate Source (kg/person/year) 1950 255 Assumption-baseda 1951-2006 Linear interpolation 2007 214 Gillett (2009) 2008 214 Carried forward 2009 214 Carried forward a  2007 estimate adjusted for increased domestically sourced fresh  Gillett (2009) estimated the 2007 total catch for fish in 1950. Tokelau to be 375 tonnes. This estimate was obtained by analyzing estimates from several independent studies completed between 1977 and 1998, with adjustments for changes in transportation availability and population (Gillett, 2009). This estimate accounts for increased transport between Tokelau and Apia, Samoa, and thus includes 125 tonnes of export (Gillett, 2009). Using these figures and the current population, Gillett (2009) estimated a consumption rate of 214 kg∙person-1∙year-1, which was used as the 2007, 2008, and 2009 anchor point (Table 1). 2 3  http://www.spc.int/prism/country/tk/stats/Social/Population/age_sex_.htm [accessed June 8, 2011] https://www.cia.gov/library/publications/the-world-factbook/geos/tl.html [accessed June 8, 2011]  For the early time period (1950-1980s), the only contact that Tokelau had with the outside world was from a trading ship, which three or four times a year brought basic items such as flour, sugar, rice, kerosene, and tobacco to the atolls (van Pel, 1958). It was therefore assumed that they were not, at that time, receiving the canned meats and fish that were available later in the time period (Passfield, 1998). Given the percentages of total animal protein obtained from canned meats and fish in 1998 (Passfield, 1998), it was assumed that this amount of animal protein was supplied in the early period by fresh seafood. In this computation we assumed that the densities of all protein products were the same. The result of this assumption was a consumption rate in 1950 that was 41 kilograms higher than in 2007, (i.e., 255 kg∙person-1∙year-1) (Table 1). We then linearly interpolated between the 1950 assumed and the 2007 reported consumption rates. Using these consumption rates along with the population data, the total demand for fresh fish was calculated for the 1950-2009 time period.  Exports In order to calculate total subsistence catch, ‗exports‘ needed to be added to what was being locally eaten by the Tokelauans. Although there are no formal exports, the people of Tokelau do send frozen and dried fish to friends and family in Samoa as gifts (Hooper, 1985; Passfield, 1998; Gillett, 2009). Passfield (1998) calculated exports for Fakaofo, and when extrapolated to all of Tokelau, they were estimated to be 15.4 tonnes per year. It is also known that it was around 1980 when the Tokelauans began sending frozen clams to Western Samoa a few times a year (Hooper, 1985). Therefore, the exports were set to zero from 1950 to 1979, and then linearly interpolated between zero tonnes in 1979 and 15.4 t in Table 2. Estimated catch composition for the subsistence fisheries 1998. We interpolated again between the of Tokelau. Prior to 1986, adjustments were made to account for 1998 point and Gillett‘s (2009) point of 125 the complete absence of Katsuwonus pelamis until 1980. Percentages were linearly interpolated between 1980 and 1986. t for 2007-2009. Adding these values to (See ‗catch composition‘ in text). the dietary demand for each year gives the Taxa Catch (%) estimate for annual subsistence catches, 1950-1980 1986-2009 1950-2009. Acanthocybium solandri 6.02 5.06  Catch composition  Acanthuridae  Elagatis bipinnulata Selar crumenophthalmus  3.65 1.50 7.30 7.30 1.00 1.46 1.50 18.25 0.13 1.05 0.50 0.00 1.46 1.46 2.19 3.65 1.46 3.65 3.65 1.46 1.46 10.95 2.19 0.50 7.40 7.40 1.46  3.65 1.50 7.30 7.30 1.00 1.46 1.50 18.25 0.11 0.88 0.50 3.52 1.46 1.46 2.19 3.65 1.46 3.65 3.65 1.46 1.46 10.95 2.19 0.50 6.215 6.215 1.46  The Food and Agriculture Organization of other Carangidae the United Nations (FAO) reports the fish Carcharhinidae catch for Tokelau under two categories: Chaetodontidae ‗tuna-like fishes not elsewhere included‘ Coryphaena hippurus (nei) and ‗marine fishes nei‘. Also Cypselurus spp. according to the FAO data, up until 1990 Grammatorcynus bilineatus there were no tuna-like fish caught (except Gymnosarda unicolor for one year of non-zero catch data in Istiophorus platypterus 1956). However, it is known that tuna- Katsuwonus pelamis fishing has historically been a very Labridae important practice to the Tokelauans Lethrinidae (Gillett and Toloa, 1987; Ono and Addison, Lutjanidae 2009). In order to show a more complete Misc. invertebrates picture of the catch composition, Misc. marine crustaceans information was combined from several Misc. marine fishes Mugilidae different sources. Gillett and Toloa (1987) Octopus spp. gave us a basic composition by not only Panulirus spp. recording the species but also the Scaridae percentage breakdown of the tuna and Serranidae tuna-like fish. They also recorded the Sphyraena barracuda species information for the other pelagic Thunnus albacares fish, but only provided an overall Thunnus obesus percentage. Based on the household survey Tridacna spp. of Passfield (1998) and the observations of van Pel (1958), we broke down the inshore fish category into species and assigned percentages to both those and the species within the other pelagic category. Although marlins (Makaira spp.) have been known to be landed by Tokelauan fishers, they have not been included as a distinct taxon within the catch  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  111  composition. Gillett (1985) reports Tokelauan fishers occasionally catch billfish, with sailfish (Istiophorus platypterus) being caught most commonly, while marlins are landed far less frequently than they used to be. Table (2) shows the complete breakdown of the subsistence catch for 1986 and onwards. Prior to 1986, the percentages of the various tuna species were adjusted to account for skipjack tuna (Katsuwonus pelamis) being absent from the waters of Tokelau from 1950-1980 (Table 2). Skipjack tuna is subject to large fluctuations in abundance. For example, the Skipjack Survey and Assessment Programme which completed its surveys for Tokelau waters in 1978, spotted 71 schools in just 35 hours (Anon., 1983). By contrast, the United States Bureau of Commercial Fisheries spent approximately 47 hours of a research cruise looking for schools of skipjack in the waters of Tokelau and spotted none (Anon., 1983). A native Tokelauan communicated to the researchers of the Skipjack Survey that in a span of ten years they will probably see one or two extremely abundant years (Anon., 1983). Hooper (1985) described a span of 15-20 years with no skipjack sightings. This covers the period of 19501970. In 1971, skipjack tuna reappeared but only for a span of about a week, after which they disappeared again (Hooper, 1985). In 1986, skipjack was recorded as part of the catch during a 12 week study (Gillett and Toloa, 1987). Taking into account all of this information, these fluctuations were adjusted for by assigning zero percent of the tuna catch to skipjack from 1950-1980. We renormalized the percentages for the other tuna species for this time period (Table 2). Skipjack tuna percentages were then linearly interpolated from zero in 1980 to 3.52% in 1986. The values for the other tuna species were also adjusted, as the percentage of skipjack increased.  Baitfish Fisheries of Tokelau use handline and noose fishing techniques which require the use of baitfish. Baitfish fisheries in other parts of the South Pacific operate in parallel to the tuna pole-and-line fishery, often using the same vessels (Gillett, 2011). The baitfish required to catch a given amount of tuna varies depending on the type of bait species used, fishing method applied, and many other factors. Tokelau appears to be a unique case in all aspects of baitfish use. The method typically used in other countries does not apply for Tokelau due to topographical characteristics which prevent large vessels from entering the lagoon (Anon., 1983). Tokelauans use very traditional methods for tuna fishing which differ from the method that the tuna to baitfish ratio of 32:1 is based on (Gillett, 2011). Reports on Tokelauan use of baitfish suggest a range of species are used including flying fish (Cypselurus spp.), double-lined mackerel (Grammatorcynus bilineatus), garfish (Hemiramphus spp.), shortfin scad (Decapterus macrosoma), bigeye scad (Selar crumenophthalmus), and squirrelfish (Holocentridae) (Gillett, 1985). This variation in baitfish use may reflect the seasonal availability of certain species. Given these differences in baitfish use, we assume that the traditional methods used by the Tokelauans are more specific and efficient than the typical baitfish catch method for pole-and-line gear. Assuming that Tokelauan fishers use half as much baitfish as used for pole-and-line fishing, we modified Gillett‘s (2011) tuna to baitfish ratio to an adjusted 64:1. Tokelauans use baitfish to catch more than just tuna, therefore this ratio was applied to the following species which are all landed by the handline and noose methods: yellowfin tuna (Thunnus albacares), dogtooth tuna (Gymnosarda unicolor), wahoo (Acanthocybium solandri), rainbow runner (Elagatis bipinnulata), bigeye tuna (Thunnus obesus), indo-pacific sailfish (Istiophorus platypterus), sharks (Carcharhinidae), and barracuda (Sphyraena barracuda) (Gillett, 1985). Skipjack tuna was not included, as they are mainly caught with artificial lures (Gillett, 1985). Traditional methods made use of pearl-shell lures, however these have become rare and difficult to acquire (Hooper, 1985). Gillett (1985) found that Tokelauans are resourceful people and would even make use of a yellow, translucent handle of a screwdriver for a lure. Baitfish estimates for Tokelau were divided evenly between the six previously mentioned taxa of baitfish used. These estimates were not included in the subsistence catch composition and therefore the 3 taxa which are used solely for bait (Hemiramphus spp., Decapterus macrosoma, and Holocentridae) and not actually consumed in the diet are not listed in Table (2).  RESULTS The reconstructed total catch for the period 1950-2009 was estimated at 24,255 t. This equates to approximately 4.3 times the total landings reported by the FAO on behalf of Tokelau for the same time period (Figure 3). Average annual catches peaked in the 1960s at approximately 460 t·year-1, with a low of 368 t·year-1 in the 1990s. Catches in the recent period (2000s) are estimated to be 399 t·year-1. Exports  DISCUSSION  Total reconstructed  500 450 400 350 300 250 200 150 100 50 0  Supplied to FAO  1950  1960  1970  1980  1990  2000  Year Figure 3. Total reconstructed fisheries catches for Tokelau compared to data supplied to FAO, 1950-2009.  Catch (t)  The total reconstructed catch was dominated by flying fish (Cypselurus spp.) with an estimated 4,425 t over the time period, representing approximately 18% of the total catch (Figure 4). Families Scaridae and Carangidae, as well as the species Selar crumenophthalmus (bigeye scad), Thunnus albacares (yellowfin tuna), Thunnus obesus (bigeye tuna), and Acanthocybium solandri (wahoo) also made up large portions of the catch, with approximately 44 t∙year-1, 29 t∙year-1, 30 t∙year-1, 28 t∙year-1, 28 t∙year-1, and 23 t∙year-1, respectively. The category ‗other taxa‘ includes the remaining 21 taxonomic groups of the subsistence catch (as per Table 2), as well as the other three baitfish taxa, as they represented smaller portions of the total catch. Together, they represent roughly 37% of the total catch. Baitfish alone represented only 0.4% of the total catch, with approximately 90 t being used over the entire time period.  Catch (t)  totalled 1,091 t over the study period, representing 4.5% of the total catch. However, these were not commercial exports, but rather gifts mainly sent to Samoa.  Acanthocybium solandri Thunnus obesus 500 Scaridae Selar crumenophthalmus Thunnus albacares 450 other Carangidae 400 350 300 250 200 150 100 50 0 1950 1960 1970 1980 1990 2000  Year The total reconstructed catch for Figure 4. Taxonomic breakdown of total reconstructed catch Tokelau during the time period 1950for Tokelau. The grouping ‗other taxa‘ represents 24 taxa (as 2009 was over four times larger than listed in Table 2, plus 3 additional baitfish taxa). the total catch reported by the FAO on behalf of Tokelau. The main reason for this difference is that there is no system in place for monitoring fisheries catches in Tokelau (Anon., 1983). According to Bertram and Watters (1984), the reports of the New Zealand Ministry of Foreign Affairs, which took over control of the islands in 1974-1975, show the lack of interest the Ministry had in the territory. Large discrepancies exist between the reported and reconstructed tuna catches. The FAO data suggest that there were zero tuna or tuna-like fish caught from 1950-1989, whereas the total tuna and tuna-like fish estimated from the reconstruction for the years of 1950-1989 represented 68% of the estimated tuna and tuna-like catch for the entire 1950-2009 study period. Also, the data supplied to FAO only represent two categories: ‗tuna-like fish nei‘ and ‗marine fish nei‘. In contrast, our reconstruction accounts for 31 taxa. Furthermore, while there are no invertebrate species taken into account in the FAO data, our reconstructed estimate suggests that invertebrates represent 9.5% of the total reconstructed catch.  Since all of Tokelau‘s catches are subsistence and our estimates are based on consumption rates, annual catch rates are subject to fluctuations which correspond to fluctuations in the population. There is a general trend of population decline from its peak in the 1960s until present. Countering the effect of this decline is the increased transport between Tokelau and Samoa which has led to an increase in informal exports. In 1980, Tokelauans began sending frozen seafood by ship to friends and family as gifts. The decrease in dietary demand, in combination with increasing amounts being sent overseas, has resulted in a relatively constant overall trend in catches.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  113  Fishing is an activity that almost everyone in Tokelau takes part in. Fishing occurs off-shore, on the outer reef, and in the lagoons (van Pel, 1958). Typically, it is the men who fish on the outer reef, with women and children taking part in gathering and fishing in the lagoon and on the reef (Chapman, 1987; Ono and Addison, 2009). Chapman et al. (2005) surveyed the atolls to obtain an estimate of the gender distribution for various fishing activities, the percentage of time using different fishing techniques, and the percentage of people on the island who take part in fishing. Of the 75% of households interviewed, 99% take part in fishing activities. An average of 12% of the fishing effort was performed by women, with most of this effort focused on reef gleaning and diving for clams. Women also participated, albeit to a lesser extent, in gillnetting and reef fishing (Chapman et al., 2005). Passfield (1998) found that each household spent approximately 14 person hours per week fishing. Although the Tokelauans have incorporated newer, modern fishing gear into their practices, there are some areas where tradition still dominates. The handmade wooden canoes are still required for handling yellowfin, skipjack, marlin, or large sharks, as well as for a traditional Tokelauan method of noosing pāla (Acanthocybium solandri) (Hooper, 1985). Many traditional fishing methods such as lama hahave (fishing with a scoop net and torch) and takiulu (luring fish into a prepared noose) are still used. Tokelauans also still use the inati system to distribute large catches (particularly of skipjack tuna) or catches of sacred ‗fish‘ (turtles, marlin, and sailfish), so that everyone receives an equal share (Hooper, 1985; Ono and Addison, 2009). Use of spears and spear-guns are not a favoured form of fishing, as it is believed that they result in fish being afraid whenever people enter the water (Hooper, 1985). However, spear-guns have become more widely used in recent years (Ono and Addison, 2009). The main methods of fishing are trolling, reef gleaning, reef fishing, mid-water fishing, gillnetting, diving, bottom fishing, and bait fishing (Chapman et al., 2005). Today, as well as in the past, trolling for Katsuwonus pelamis (skipjack tuna) and Thunnus albacares (yellowfin tuna) is the most important type of fishing to the Tokelauan people (Ono and Addison, 2009). There are no commercial fisheries in Tokelau. Although it is known that a portion of the catch is used for internal island trade or shipped to friends and family as gifts, these activities are not a commercial cash enterprise. The act of selling fish is viewed as offensive by the Tokelauan people (Hooper, 1985). Traditional systems make it almost impossible to sell fish, regardless of what the individual‘s views are. The people of Tokelau operate under the inati system, which essentially requires fishers to share large catches with the entire community. This takes away the motivation and possibility of catching fish to be sold for pers0nal financial gain (Passfield, 1998). Even if Tokelau abandoned these traditional systems, they do not have the resources to maintain a commercial fleet of their own. There are no ports for larger commercial vessels to dock and land large quantities of catch. The MV Tokelau, which transports goods and people to and from Tokelau, must anchor offshore and have its passengers and cargo transported to land by a low aluminum barge with an outboard motor (Townend, 2009). The atolls are closed and only have about 15 natural depressions and a few blasted channels which small vessels can pass through into the lagoon (Gillett and Toloa, 1987). Therefore, given the topography of the atolls, it would not be easy to maintain a commercial fleet for Tokelau. In 1980, the United Nations Development Programme (UNDP) attempted to develop the Tokelauan fishing industry by giving each atoll a 29-foot catamaran which would be used to start a small artisanal fishery or to simply help increase their current subsistence efforts (Hooper, 1985). However, use of the catamarans required a large amount of village cooperation and this was difficult to organize with a large proportion of the population working in the public service sector (Hooper, 1985). The catamarans also have high operating costs (e.g., fuel and maintenance) which would require fish to be sold to recoup costs, which is something the Tokelauans frown upon (Hooper, 1985). Thus, the catamarans ended up being used for shipping people and goods across the lagoon (Hooper, 1985). Another reason that commercial fishing would not be profitable in Tokelau is that fishing beyond the subsistence level would possibly put too much pressure on the already limited fish stocks which would likely be detrimental to resource levels (van Pel, 1958; Passfield, 1998). Overall it appears that there is little threat of overfishing in Tokelau. This can be attributed to the Tokelauans methods and views toward fishing. The Tokelauans are very aware of what is available in their waters and thus know what they can and cannot take. As well, appropriate measures would be taken if any declines were in fact observed (Hooper, 1985). The Tokelauans rely heavily on the fish, and know if overfished they will not have anything to eat for the next week, month, or year. Giant clams on the other hand, may be in jeopardy. According to an early report by van Pel (1958), Tridacna gigas were relatively abundant in the early period, whereas later on there is no mention of them. However, later reports mention two other clam species: Tridacna squamosa and Tridacna maxima (Passfield, 1998). According  to Hooper (1985), giant clams were regarded as an emergency resource that were only harvested when severe weather conditions prevented the catch of other fish. However, around 1980, the Tokelauans made use of the commercial freezers given to them by the UNDP and would load them up with frozen giant clams to be sent to friends and family in Western Samoa as gifts (Hooper, 1985). This occurred only a few times a year based on the shipping schedule (Hooper, 1985). Although this may have put pressure on stocks, it seems that the Tokelauans only had to change their technique and dive a little deeper if they wished to continue harvesting (Hooper, 1985). In a more recent study, there is mention of frequent harvesting of T. squamosa and T. maxima. However, it also stated that on Atafu there have been strict limitations put in place on the harvesting of these clams (Ono and Addison, 2009). Therefore, it seems that concern over these stocks has resulted in measures being put in place to prevent overharvesting. Although there have not been many studies conducted on the fishing activities of Tokelau, it can be seen from the data that has been collected that even something as simple as a household survey once every five years could provide important information (Zeller et al., 2006; Gillett, 2009). Although creel surveys (i.e., interviews with fishers and surveillance of their catch) are too costly for small countries to complete every year, a survey completed once every five years would provide important data which could be interpolated between collection years (Zeller et al., 2007). If data were collected on the eating habits of Tokelauans, how much they export and import, how much of their seafood consumption is fresh, what types of fish they are catching, or what kind of gear they are using, this could be used to derive more comprehensive estimates of their annual catches (Zeller et al., 2007). Regional (e.g., SPC) and international agencies (e.g., FAO) should consider facilitating and establishing such data collection and utilization approaches, and the required technical and financial resources. For a small island country such as Tokelau, where people rely on the ocean for their sustenance, monitoring of fisheries removals is fundamental to maintaining national food security.  ACKNOWLEDGMENTS We acknowledge the support of the Sea Around Us Project, a scientific collaboration between the University of British Columbia and the Pew Environment Group.  REFERENCES Adams, T., Richards, A., Dalzell, P. and Bell, L. (1995) Research on fisheries in the pacific islands region. Forum Fisheries Agency and South Pacific Commision, Nouméa, New Caledonia, 79 p. Anon. (1983) An assessment of the skipjack and baitfish resources of Tokelau. South Pacific Commission, Nouméa, New Caledonia, 46 p. Anon. (2006) Tokelau 2006 census of population and dwellings - tabular report. Statistics New Zealand and the Office of the Council for the Ongoing Government of Tokelau 67 p. Bertram, I.G. and Watters, R.F. (1984) New Zealand and its small island neighbours: A review of New Zealand policy toward the Cook Islands, Niue, Tokelau, Kiribati and Tuvalu. Institute of Policy Studies, Wellington, New Zealand, 574 p. Chapman, L., Des Rochers, K. and Pelasio, M. (2005) Survey of fishing activities in Tokelau. SPC Fisheries Newsletter 115: 36-40. Chapman, M. (1987) Women's fishing in oceania. Human Ecology 15: 267-288. Clanton, A. (2008) The men who would be king: Forgotten challenges to U.S. sovereignty. UCLA Pacific Basin Law Journal 26: 1-50. Dalzell, P., Adams, T.J.H. and Polunin, N.V.C. (1996) Coastal fisheries in the Pacific islands. p. 395-531 In Ansell, A.D., Gibson, R.N. and Barnes, M., (eds.), Oceanography and Marine Biology: An Annual Review. UCL Press, London, UK. Gillett, R. (1985) Traditional tuna fishing in Tokelau. South Pacific Commision, Nouméa, New Caledonia, 52 p. Gillett, R. (2009) Fisheries in the economies of the pacific island countries and territories. Asian Development Bank, Mandaluyong City, Philippines, 521 p. Gillett, R. (2011) Replacing purse seining with pole-and-line fishing in central and Western Pacific: Some aspects of the baitfish requirements. Marine Policy 35: 148-154. Gillett, R. and Toloa, F. (1987) The importance of small-scale tuna fishing: a Tokelau case study. p. 177-190 In Doulman, D.J., (ed.) Tuna Issues and Perspectives in the Pacific Islands Region. Pacific Islands Development Program, Honolulu.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  115  Hoëm, I. (2009) 'Getting out from under': Leadership, conflict resolution and Tokelau migration. p. 191-202 In Lee, H. and Francis, S.T., (eds.), Migration and Transnationalism: Pacific Perspectives. ANU E Press, Canberra, Australia. Hooper, A. (1985) Tokelau fishing in traditional and modern contexts. p. 9-38 In Ruddle, K. and Johannes, R.E., (eds.), The Traditional Knowledge and Management of Coastal Systems in Asia and the Pacific. UNESCO, Jakarta, Indonesia. Ono, R. and Addison, J. (2009) Ethnoecology and Tokelauan fishing lore from Atafu Atoll, Tokelau. SPC Traditional Marine Resource Management and Knowledge Information Bulletin 26: 3-22. Passfield, K. (1998) A report of a survey of the marine resources of Fakaofo Atoll, Tokelau. Fisheries and Environmental Resource Consultants Ltd 31 p. Townend, A. (2007) Tokelau's 2006 referendum on self-government. New Zealand Journal of Public and International Law 5: 121-168. Townend, A. (2009) From a Tokotoko: pacific travels with professor Angelo. Victoria University of Wellington Law Review 39: 575-582. van Pel, H. (1958) A survey of fisheries in the Tokelau islands. South Pacific Commision, Nouméa, New Caledonia, 18 p. Zeller, D., Booth, S., Craig, P. and Pauly, D. (2006) Reconstruction of coral reef fisheries catches in American Samoa, 1950–2002. Coral Reefs 25: 144-152. Zeller, D., Booth, S., Davis, G. and Pauly, D. (2007) Re-estimation of small-scale fishery catches for U.S. flag associated island areas. Fisheries Bulletin 105.  Appendix Table A1: FAO landings vs. total reconstructed catch (in tonnes) for Tokelau, 1950-2009. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009  FAO landings 0.25 0.25 0.25 0.25 0.25 0.25 0.50 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 0.25 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 150.00 150.00 150.00 200.00 200.00 200.00 220.00 231.00 191.00 200.00 200.00 200.00 200.00 200.00 200.00 200.00 160.00 150.00 140.00 130.00 120.00 110.00 100.00 90.00 80.00 70.00  Reconstructed catch 384.0 408.5 407.3 406.1 430.3 429.1 427.8 451.7 450.4 449.1 472.7 471.3 470.0 468.6 467.2 465.8 464.5 463.1 461.7 411.9 410.7 409.4 384.2 383.0 381.9 380.7 379.6 378.4 377.3 376.1 375.8 379.3 382.7 386.2 389.8 393.1 396.5 391.5 386.5 381.6 376.7 371.8 416.5 393.6 370.8 358.0 345.0 345.3 346.0 357.8 369.6 388.8 396.9 405.0 412.8 420.9 429.0 376.7 388.3 399.6  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  117  Appendix Table A2: Total reconstructed catch (in tonnes) for Tokelau (1950-2009) by major taxa. Others grouping includes 24 taxa. Year  Cypselurus  Scaridae  spp.  1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009  70.1 74.5 74.3 74.1 78.5 78.3 78.1 82.4 82.2 81.9 86.2 86.0 85.7 85.5 85.2 85.0 84.7 84.5 84.2 75.1 74.9 74.7 70.1 69.9 69.7 69.5 69.2 69.0 68.8 68.6 68.6 69.2 69.8 70.5 71.1 71.7 72.3 71.4 70.5 69.6 68.7 67.8 76.0 71.8 67.7 65.3 62.9 63.0 63.1 65.3 67.4 70.9 72.4 73.9 75.3 76.8 78.3 68.7 70.8 72.9  41.9 44.5 44.4 44.3 46.9 46.8 46.7 49.3 49.1 49.0 51.6 51.4 51.3 51.1 51.0 50.8 50.7 50.5 50.4 44.9 44.8 44.7 41.9 41.8 41.7 41.5 41.4 41.3 41.1 41.0 41.0 41.4 41.7 42.1 42.5 42.9 43.3 42.7 42.2 41.6 41.1 40.6 45.5 42.9 40.5 39.1 37.6 37.7 37.8 39.0 40.3 42.4 43.3 44.2 45.1 45.9 46.8 41.1 42.4 43.6  Selar crumenophthalmus 28.2 30.0 29.9 29.8 31.6 31.5 31.4 33.1 33.0 33.0 34.7 34.6 34.5 34.4 34.3 34.2 34.1 34.0 33.9 30.2 30.1 30.0 28.2 28.1 28.0 27.9 27.8 27.8 27.7 27.6 27.6 27.8 28.1 28.3 28.6 28.8 29.1 28.7 28.3 28.0 27.6 27.3 30.5 28.9 27.2 26.2 25.3 25.3 25.4 26.2 27.1 28.5 29.1 29.7 30.3 30.9 31.5 27.6 28.5 29.3  Other Carangidae 27.9 29.7 29.6 29.5 31.3 31.2 31.1 32.8 32.8 32.7 34.4 34.3 34.2 34.1 34.0 33.9 33.8 33.7 33.6 29.9 29.9 29.8 27.9 27.9 27.8 27.7 27.6 27.5 27.4 27.3 27.3 27.6 27.8 28.1 28.4 28.6 28.8 28.5 28.1 27.8 27.4 27.0 30.3 28.6 27.0 26.0 25.1 25.1 25.2 26.0 26.9 28.3 28.9 29.5 30.0 30.6 31.2 27.4 28.2 29.1  Thunnus albacares  Thunnus obesus  Acanthocybium solandri  Others  28.3 30.1 30.0 29.9 31.7 31.6 31.5 33.3 33.2 33.1 34.8 34.7 34.6 34.5 34.4 34.3 34.2 34.1 34.0 30.4 30.3 30.2 28.3 28.2 28.1 28.1 28.0 27.9 27.8 27.7 27.7 27.2 26.7 26.2 25.7 25.1 24.6 24.3 23.9 23.6 23.3 23.0 25.8 24.4 23.0 22.2 21.4 21.4 21.4 22.2 22.9 24.1 24.6 25.1 25.6 26.1 26.6 23.3 24.0 24.8  28.3 30.1 30.0 29.9 31.7 31.6 31.5 33.3 33.2 33.1 34.8 34.7 34.6 34.5 34.4 34.3 34.2 34.1 34.0 30.4 30.3 30.2 28.3 28.2 28.1 28.1 28.0 27.9 27.8 27.7 27.7 27.2 26.7 26.2 25.7 25.1 24.6 24.3 23.9 23.6 23.3 23.0 25.8 24.4 23.0 22.2 21.4 21.4 21.4 22.2 22.9 24.1 24.6 25.1 25.6 26.1 26.6 23.3 24.0 24.8  23.0 24.5 24.4 24.4 25.8 25.7 25.7 27.1 27.0 26.9 28.4 28.3 28.2 28.1 28.0 28.0 27.9 27.8 27.7 24.7 24.6 24.6 23.1 23.0 22.9 22.8 22.8 22.7 22.6 22.6 22.5 22.2 21.7 21.3 20.9 20.4 20.0 19.7 19.5 19.2 19.0 18.7 21.0 19.8 18.7 18.1 17.4 17.4 17.4 18.0 18.6 19.6 20.0 20.4 20.8 21.2 21.6 19.0 19.6 20.2  136.3 145.0 144.6 144.2 152.8 152.3 151.9 160.4 159.9 159.4 167.8 167.3 166.8 166.4 165.9 165.4 164.9 164.4 163.9 146.2 145.8 145.4 136.4 136.0 135.6 135.2 134.8 134.3 133.9 133.5 133.4 136.8 140.1 143.5 147.0 150.4 153.9 151.9 150.0 148.1 146.2 144.3 161.6 152.7 143.9 138.9 133.9 134.0 134.3 138.8 143.4 150.9 154.0 157.1 160.2 163.3 166.5 146.2 150.7 155.1  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  119  RECONSTRUCTING MARINE FISHERIES CATCHES FOR THE KINGDOM OF TONGA: 1950-20071 Patricia Sun, Sarah Harper, Shawn Booth and Dirk Zeller  Sea Around Us Project, Fisheries Centre, University of British Columbia, 2202 Main Mall, Vancouver, BC, V6T 1Z4, Canada p.sun@fisheries.ubc.ca; s.harper@fisheries.ubc.ca; s.booth@fisheries.ubc.ca; d.zeller@fisheries.ubc.ca  ABSTRACT Total marine fisheries catches were estimated for Tonga from 1950-2007 by reconstructing past catches and accounting for all fisheries sectors, including commercial and non-commercial components. Our catch reconstruction for Tonga estimated the total catch to be over 3.5 times larger than the landings reported to FAO over the study period. For recent period (2000s), this discrepancy has decreased with total catches averaging 5,600 t∙year-1, compared to 3,300 t∙year-1 reported by FAO on behalf of Tonga. The majority of catches that were unaccounted for in the reported landings were from the subsistence sector, which represented 70% of the total reconstructed catch. This illustrates the substantial under-representation of small-scale subsistence fisheries in the official statistics for Tonga. Small-scale fisheries play a key role in the socioeconomic framework and food security of Pacific Island countries such as Tonga and this needs greater recognition and better accounting.  INTRODUCTION The Kingdom of Tonga is located between 15˚23.5˚S and 173˚- 177˚W in the South Pacific Ocean (Figure 1). Tonga consists of approximately 170 islands, of which 37 are inhabited (Zann 1994; Anon., 2010a), and has an Exclusive Economic Zone (EEZ) of approximately 665,000 km2 (www.seaaroundus.org). The islands of Tonga are clustered into three groups: Tongatapu, Ha‘apai, and Vava‘u, which have a combined land area of about 747 km2 (Malm, 2009). About 70% of the population of Tonga resides on Tongatapu, which is also the location of the capital city, Nuku‘alofa. The islands of Ha‘apai and Vava‘u are less developed, have smaller populations and rely more on subsistence fishing and farming (Evans et al., 2003). Prior to the Tongan constitution of 1875, fishing rights to nearshore marine areas were under control of community chiefs and belonged only to coastal people (Kronen et al., 2003). After 1875, there was an abolishment of exclusive fishing rights to particular marine areas and all people had the right to fish or gather marine resources (Kronen et al., 2003). This system is still in place today, with the exception of fish fences, live rock extraction, 1  Figure 1. Map of Tonga and its Exclusive Economic Zone.  Cite as: Sun, P., Harper, S., Booth, S. and Zeller, D. (2011) Reconstructing marine fisheries catches for the Kingdom of Tonga: 1950-  2007. pp. 119-130. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727].  aquaculture, and marine protected areas (Kronen et al., 2003). Tongan fisheries include both commercial and non-commercial sectors. The small-scale fisheries target reef-lagoon, pelagic and deep-slope species. While the majority of small-scale fishery catches are from reef areas, Tonga‘s deep-slope fishery has been considered one of most successful in the Pacific Island region (Dalzell et al., 1996). Traditionally, Tongans fished only inside the reef (Bell et al., 1994); however, in the early 1980s a deep-water fishery began, initially using small boats, but eventually expanding to include larger vessels (Mead, 1979; 1981). A large-scale commercial fishery for offshore pelagics started in the 1980s with one vessel, but grew to 25 vessels by the 2000s (Likiliki et al., 2005). A commercial lobster fishery began in the late 1960s and a sea cucumber (bêche-de-mer) fishery was established in the 1990s. Neither of these invertebrate fisheries expanded substantially. The open ocean fisheries were conducted mainly by men, while reef collecting activities (reef gleaning) involved mostly women and children (Malm, 2009). Expansion of Tonga‘s commercial fisheries has been relatively unstable due to large fluctuations in stocks. For example, mullet stocks (Mugilidae) in Tongatapu have faced several major collapses, beginning in the 1960s when monofilament nets were introduced (SPC, 1988). Deepwater snapper (Lutjanidae) stocks on offshore slopes and seamounts have also been substantially depleted (Zann, 1994). The lobster fishery has gone through various boom and bust cycles since it began in the 1960s (Bell et al., 1994) and the bêche-demer fishery, which began in 1990, was deemed overfished by 1996 and has been under a ten-year moratorium since 1998 (Bell et al., 1994). The commercial fisheries sector of Tonga makes up around 3% of the country‘s GDP (Gillet, 2009); however, subsistence farming and fishing are practiced by the majority of people (Zann, 1994). Though there has been a shift in recent years towards agriculture and imported foods in Tongatapu, fishing remains an important source of protein in the diet for many on the outer islands of Ha‘apai and Vava‘u. Methods for subsistence fishing vary depending on the equipment that is readily available, and many fishers employ a combination of techniques. The most popular gear is the handline, followed by spear fishing and gillnetting (Kronen and Bender, 2007). Women and children use spears, traps, and may also participate in some types of group fishing. Men typically fish with spears, hooks, nets, and traps. The objective of this study is to provide a comprehensive estimate of Tonga‘s total marine fisheries catches by accounting for commercial (industrial and artisanal) and non-commercial (subsistence) fisheries sectors. The Food and Agriculture Organization of the United Nations (FAO) reports marine fisheries landings, as supplied by each member country, in the form of time-series data, extending from 1950 to present, and can be found in the FAO FishStat database (www.fao.org/fishery/statistics/en). These data are based on national statistical data from each country, and can be subject to omissions depending on the method of statistical collection and the quality of data transfer. As such, much of the FAO landings data include only the commercial fisheries sector. Given the importance of fishing for subsistence in the coastal communities of a country such as Tonga, and its common under-representation in catch statistics (e.g., Zeller et al., 2006a), we will estimate this unreported fisheries component to improve our understanding of Tonga‘s use of marine fisheries resources using a catch reconstruction approach (Zeller et al., 2007). This study looked at all subsectors within Tonga‘s fisheries, including large-scale (i.e., offshore fishing), artisanal (i.e., reef, deep-slope, and near-shore fishing), and subsistence fishing in order to estimate total marine fisheries catches by Tonga from 1950-2010.  METHODS Tongan fisheries focus on targeting reef and lagoon, deep-slope and pelagic species. There are both commercial and non-commercial components to the reef and lagoon fisheries, whereas the deep-slope and pelagic fisheries are predominantly commercial. Estimates of Tonga‘s commercial landings were obtained from FAO, whereas subsistence catch data and taxonomic information on small-scale fisheries were derived from household surveys, independent studies and from gray literature sources. Subsistence fisheries estimates were only available for some years; therefore, subsistence catch data were combined with human population data to derive per capita subsistence catch rates, which were then expanded to cover the entire study period. Our efforts at reconstructing Tonga‘s fisheries catches focused mainly on estimating catches for the small-scale sector and improving the taxonomic resolution of all sectors.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  121  Human Population data 120  Population (x 103 )  Human population data for 1950 were obtained from populstat (www.populstat.info/), for 1956, 1966, 1976, 1986 and 1996 from the Statistic Department of Tonga (Fifita, 1996), and for 2006 from the Secretariat of the Pacific Community (Anon., 2010b). Linear interpolations were used to estimate the population between years of known data. Data from 1976 onwards provided a population breakdown for the main island (Tongatapu) and the outer islands (Ha‘apai, Vava‘u, and others). Prior to 1976 we extrapolated the ratio back to 1950 and applied it to the total population to derive the Tongatapu and outer Island populations for the earlier time period (Figure 2).  100  Outer islands  80 60  40  Tongatapu  20  0 1950  1960  1970  1980  1990  2000  Year Figure 2. Tonga‘s population for the main island (Tongatapu) and outer islands combined (Ha‘apai, Vava‘u and others), 1950-2007. There has been a migration to the main island, Tongatapu, from the outer islands in the recent period. In the mid-2000s approximately 70% of the population resided on Tongatapu, whereas in the 1950s just over 50% resided on the main island.  Subsistence fisheries Most inhabitants of the outer island groups of Ha‘apai and Vava‘u participate in fisheries on some level. On the main island group, Tongatapu, there has been a shift in dietary preferences from fish as a major source of protein to mutton flaps, chicken pieces, and corned beef, which have become cheap alternatives (Finau et al., 1994; Gillett, 2009). Our estimates for subsistence fishing reflect a decrease on islands with increased urbanisation (e.g., Tongatapu) due to wider availability of other food alternatives. This is supported by Kronen et al. (2003), who found a decrease in fresh fish consumption with increased urbanization. Reliable estimates for subsistence fisheries catches in the early time period were not readily available. Kent (1980) estimated that local fisheries production in 1977 (i.e., 1,100 t) supplied less than half of the populations seafood demand. Assuming the remaining demand was met through subsistence catch, we derived an estimated subsistence amount of 1,300 t for 1977 by assuming that artisanal catches supplied approximately 45% of the overall demand. An estimate by Dalzell et al. (1994), for the mid 1990s was 933 t. Both of these seemed low when considering that more recent estimates were more than double. The estimation techniques in these earlier accounts were not adequately described, and they seemed unrealistically low when converted to per capita subsistence catch rates (i.e., 10-14 kg∙person-1∙year-1). Furthermore, the FAO country profile for Tonga (www.fao.org/fishery/countrysector/FI-CP_TO/en [accessed June 2010]) suggests that these earlier subsistence estimates were in fact low. Therefore, we used Gillet and Lightfoot‘s (2002) estimate for 1997 of 2,863 t as the best estimate of subsistence catches. This 1997 subsistence estimate was converted to per capita subsistence catch rates for the main island and the outer islands using the proportion of subsistence catches from the main and outer islands (40% and 60% respectively) as presented by Lovell and Palaki (2002). The resulting per capita rates were: 17 kg∙person-1∙year-1 for the main island and 56 kg∙person-1∙year-1 for the outer islands for 1997. Halapua (1981 in Bell et al., 1994) estimated that in the Ha‘apai island group (i.e., outer islands), subsistence reeflagoon fisheries constitute 70% of the total annual catch. We applied this breakdown to the 1997 catch for the outer islands using our estimated subsistence catch as 70% of the total catches to derive our artisanal catch amount (i.e., 56 kg∙person-1∙year-1 accounted for 70% of total catch rate for outer islands being approximately 80 kg∙person-1∙year-1). We assumed that in 1950 catches on the outer Islands were all subsistence and used our 1997 per capita rate for total catches of 80 kg∙person-1∙year-1 (i.e. domestic supply) as the 1950 subsistence catch rate. A linear interpolation was used to derive a time series from 1950 to 1997, and this trend was carried forward from 1997 to 2007.  We further assumed that the per capita subsistence catch rate for the main island (Tongatapu) in 1950 was the same as the outer islands in the late 1990s (56 kg∙person -1∙year-1), as commercial fisheries and food imports were less developed in the early time period and the entire population would have relied more heavily but not exclusively on subsistence catches.  Table 1. Comparison of tuna catches presented in FAO Fishstat and in Likilili et al. (2005) for the Tongan Longline fishery, which included albacore, bigeye, black marlin, skipjack, swordfish, tuna-like, and yellowfin catches. Year FAO Likiliki et al. (2005) 1997 690 214 1998 870 193 1999 1129 327 2000 1269 931 2001 1836 1988 2002 1804 1647 2003 1126 1308 2004 517 373  In the later time period, as commercial fisheries developed, mainly around Tongatapu, reliance on subsistence fisheries diminished as the population derived more of their animal protein supply from commercial fisheries and from alternative non-seafood sources including imported food items. To derive a complete time series, we carried the 1950 rate (56 kg∙person-1∙year-1) forward unaltered to 1970, when the reef and lagoon fishery of Tongatapu started to show signs of overexploitation. From the 1970 anchor point of 56 kg∙person -1∙year-1 to the 1997 estimate of 17 kg∙person-1∙year-1 we interpolated linearly. From 1997 onward, the rate was decreased by 1% year -1, which is the same decrease observed in the subsistence rate for the outer islands. When combined with commercial catches for the domestic market (i.e., artisanal catch), these subsistence catch estimates resulted in an average per capita fish consumption rate of 80 kg∙person-1∙year-1 in 1950, decreasing to 38 kg∙person-1∙year-1 by 2007.  Commercial fisheries data Offshore pelagics: Catches for tuna presented in the FAO data were comparable to estimates presented in Likiliki et al. (2005) for the tuna longline fishery. For some years, the FAO provide more comprehensive estimates (Table 1). We therefore assumed that the FAO data for offshore pelagics (tuna and billfishes) best represented total catches for the industrial pelagics fishery. We assumed that catches of pelagic species for the domestic market by the small-scale sector, were not included in the landings presented by the FAO and were therefore estimated as a component of the artisanal catch (see taxonomic breakdown section). Deep slope fishery: The deep slope fishery began in the 1980s and mainly targets snapper (Lutjanidae) and grouper (Serranidae). This fishery was developed to alleviate pressure on the reef and lagoon fish stocks, which by the 1980s were already under pressure from overexploitation (Ministry of Fisheries, 2007). Initially, these catches would have supplied the domestic market; however, once their value as an export commodity was established, most of the catch went to foreign markets (Ministry of Fisheries, 2007). Catches for the deep slope fishery were obtained from Bell et al. (1994) for the period 1986-1992, from the Ministry of Fisheries (2007) for the period 1992-2005, and from FAO Fishstat for 2006 and 2007. Estimates for the 1980-1987 period were derived by linear interpolation from assumed zero catches in 1980 to the first available anchor point in 1987. We assumed that these estimates included catches for both domestic and foreign (export) markets, although the majority would have been for export, and thus were not included in our calculations of total domestic supply (see Subsistence section). Artisanal fishery: A report by Bell et al. (1994) presented estimates for the artisanal fishery of Tongatapu for 1987 and 1993. The 1993 estimate was based on catches from the two major landing sites, expanded to account for all of Tongatapu. These estimates were made over a 10-month period (March-December), so catches were also expanded (using the monthly average over the 10-months) to represent catches for January and February. The 1987 and 1993 artisanal catch estimates were 823 t and 386 t, respectively. These included mullet (Mugilidae) catches, which were subsequently removed from the artisanal catch estimate and treated separately. We estimated artisanal catches for 1950 using our Tongatapu subsistence catch estimate for 1950 and the assumption that in 1950 subsistence catches made up 70% of the total catch, while artisanal catches made up the remaining 30%. This assumption was based on our estimated subsistence-artisanal breakdown for the outer islands in the 1990s. On the outer islands, we assumed that in 1950 there was no artisanal sector and that all catches were subsistence. For the later time period, we estimated artisanal catches using our subsistence estimate and the assumption that, in the 1990s on the  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  Table 2. Species composition of the reef fisheries catches in Tonga applied to both artisanal and subsistence catches. Source: Ministry of Fisheries Database, Inshore Fisheries Statistics (Bell et al., 1994). Family Species Tongan Name Catch (%) Acanthuridae Acanthurus spp. Pone 8.71 Acanthuridae Acanthurus lineatus Ponetuhi 0.50 Acanthuridae A. triostegus Manini 0.25 Acanthuridae Naso spp. Ume lei 0.54 Acanthuridae N. unicornis Ume 8.90 Balistidae Pseudobalistes fuscus Humu 1.53 Belonidae Tylosurus crocodilis Haku 0.10 Carangidae Caranx spp. Lupo 0.33 Diodontidae Diodon spp. Sokisoki 0.32 Ephippidae Platax pinnatus Sifisifi 0.07 Fistulariidae Fistularia spp. Totao 0.09 Gerreidae Gerres spp. Matu 2.01 Haemulidae Plectorhinchus spp. Fotu'a 0.97 Holocentridae Ostichthys spp. Ta'a 2.40 Holocentridae Sargocentron spp. Telekihi 0.19 Holocentridae Myripristis spp. Malau 0.70 Kyphosidae Kyphosus cinerascens Nue 0.50 Labridae Cheilinus undulatus Tangafa 0.13 Leiognathidae Leiognathus spp. Sipesipa 0.05 Labridae Thalassoma spp. Meai 0.87 Labridae Cheilinus spp. Lalafi 0.67 Lethrinidae Lethrinus atkinsoni Hoputu 1.78 Lethrinidae Lethrinus harak Tanutanu 11.53 Lethrinidae Lethrinus nebulosus Koango, Liki 2.58 Lethrinidae Lethrinus spp. Manga 1.33 Lutjanidae Lutjanus bohar Fangamea 0.07 Lutjanidae Lutjanus kasmira Fate 1.21 Lethrinidae Gymnocranius spp. Mu 1.23 Monocanthidae Aluterus spp. Papae 0.05 Mullidae Mulloidichthys spp. Vete 1.40 Mullidae Parupeneus spp. Tukuleia 3.82 Muraenidae Gymnothorax spp. Toke 0.14 Pomacentridae Abudefduf septemfasciatus Tukuku moana 0.68 Priacanthidae Priacanthus spp. Mataheheva 0.40 Scaridae Leptoscarus vaigiensis Ufu 7.67 Scaridae Scarus spp. Olomea, Pose 18.31 Scaridae Bolbometopon muricatum Menenga 0.22 Serranidae Epinephelus spp. Ngatala 5.84 Siganidae Siganus argenteus Ma'ava 6.11 Siganidae S. chrysospilos Pongongo 0.63 Siganidae S. spinus O 4.31 Sphyraenidae Sphyraena barracuda Ono 0.15 Sphyraenidae Sphyraena spp. Hapatu 0.45 Terapontidae Therapon jarbua Kavakava 0.27  123  outer islands, artisanal catches represented 30% of the total catch. Artisanal catches for 1950, 1987 and 1993 for both main and outer islands were then converted to per capita artisanal catch rates and linear interpolations were done to derive a complete time series from 1950-1993. From 1993-2007 we, conservatively, held the artisanal catch rate constant. Mullet: Traditionally, mullet were the most sought after species by the domestic market (Bell et al., 1994). Zann et al. (1994) estimated that in the 1960s, mullet represented 40% of all domestically marketed species. However, heavy exploitation of mullet stocks in the 1950s and 1960s lead to a substantial decline in mullet landings, which likely began in the 1970s (Zann et al., 1984) and continued to decline dramatically, with a near collapse in the 1990s (Bell et al., 1994).  FAO data do not present mullet catches as a separate category, but they may be accounted for as ‗miscellaneous marine fishes‘. Independent reports were obtained, which described the state of the mullet fishery and presented catches for some years (Bell et al., 1994; Kimura and Fa‘anunu, 1995). Bell et al. (1994) present mullet landings for Tongatapu of 140 t in 1987 and 3.2 t in 1993. We estimated that from 1950-1970, mullet represented 40% of artisanal catches for Tongatapu, which translates into a 1950 catch of almost 250 t and a 1970 catch amount of 364 t. A linear interpolation was done to estimate catches for the time period, 1970 to 1987 and 1987 to 1993. From 1994 onward we carried forward the 1993 catch of 3.2 t unaltered.  Taxonomic breakdown for artisanal and subsistence fisheries Taxonomic composition for the reef, deep-slope, and coastal pelagic fisheries are presented in Dalzell et al. (1996) but only to the family level. Mead (1980) provided a similar breakdown for deep-slope and pelagic fisheries, which included both family and species details, but only for 1979. Bell et al. (1994) provided taxonomic information to the species level for the reef fishery for 1987 and 1993, for the deep-slope fishery (1986-1992), for the mullet fishery and for pelagic species. We used the species composition in Bell et al. (1994) as it provided the most comprehensive taxonomic breakdown by sector and species. Mullet catches  were composed of Mugil cephalus (70%), Valamugil seheli (15%) and Liza spp. (15%). The species breakdown for the deep-slope fishery was Etelis coruscans (25.44%), Pristipomoides filamentosus (22.35%), Epinethelus septemfasciatus (13.98%), Etelis carbunculus (6.07%), Lethrinus chrysostomus (5.74%), Pristipomoides flavipinnis (3.90 %), Epinephelus morrhua (3.71%) and others (18.81%). The remainder of the artisanal catch was broken into reef-lagoon species (70%), sharks and rays (0.15%), other oceanic pelagics (20.1%), small pelagics (8.25%), and ‗miscellaneous marine fishes‘ (2.5%). To each of these categories, we applied the species breakdown presented in Bell et al. (1994). Reef fishery catches were assigned to 49 taxa from 25 families (Table 2). Catches of oceanic pelagics were composed of barracudas (Sphyraena spp.; 32%), marlin (Makaira nigricans, M. indica and Tetrapturus audax; 40%) and dolphinfish (Coryphaena hippurus; 28%). Small pelagics consisted of 11 taxa from 5 families (Table 3). Tuna catches by the artisanal fishery were mainly skipjack (Katsuwonus pelamis; 76%), with smaller amounts of yellowfin tuna (Thunnus albacares; 20%) and little tuna (Euthynnus affinis; 3%). The taxonomic breakdown for sharks was derived from species presented in Bell et al. (1994). We assumed: grey reef shark (Carcharhinus amblyrhynchos; 20%), silvertip shark (C. albimarginatus; 20%), black-tip reef shark (C. melanopterus; 20%), great white shark (Carcharodon carcharias; 5%), hammerhead (Sphyrna lewini; 20%), mako shark (Isurus oxyrinchus; 5%) and tiger shark (Galeocerdo cuvieri; 10%). No species breakdown was available for ray catches. Thus 1/3 of the shark and ray allocation was assigned to rays Table 3. Species composition of artisanal catches of small pelagics.  Catch (t x 103 )  Reef gleaning, which targets mainly Source: Bell et al. (1994) invertebrates, is known to be a Species Family Catch (%) significant contributor to subsistence Stolephorus devisi Engraulidae 20.0 fishing and is conducted mainly by Atherinomorus lacunosus Atherininae 20.0 women (Chapman, 1987; Malm, Spratelloides delicatulus Clupeidae 13.0 2009). While invertebrate fisheries Selar crumenopthalmus Carangidae 11.0 are likely of equal importance as Atule mate Carangidae 11.0 finfish in the subsistence sector Hypoatherina ovalaua Atherinidae 9.2 (Adams and Dalzell, 1994), Sardinella sirm Clupeidae 7.0 quantitative information on the Herklotsichthys punctatus Clupeidae 6.0 magnitude of invertebrate extractions Spratelloides gracilis Clupeidae 2.0 Carangidae 0.4 for subsistence purposes was not Scomberoides spp. Leiognathidae 0.4 readily available (Bell et al., 1994). Gazza minuta Therefore, we derived the breakdown of invertebrate to finfish catches for the subsistence fisheries using an estimate given by Malm (2009) for the amount of invertebrates caught by women on the outer islands of Tonga in 1975 (11 kg whole weight∙household-1∙week-1), since almost all invertebrate fishing was conducted by women. We took 30% of the 11 kg∙household-1∙week-1 to estimate the meat portion of the invertebrate catch (60-70% shell weight [Kunatuba and Uwate, 1993 in Malm, 2009]) and assuming a household size of approximately 5 persons. We expanded the repored catch rate to cover 52 weeks (whole year). The resulting per capita invertebrate catch rate was 34 kg∙person-1∙year-1. Total reconstructed 7 We then compared the per capita catch rate for invertebrates to the 6 total per capita subsistence catch rate for the outer islands in 1975 5 (67 kg∙person-1∙year-1) and used the 4 resulting ratio to estimate the finfish component (50%) of the 3 as supplied to FAO artisanal catch. We then applied 2 this breakdown (50% invertebrates and 50% finfish) to subsistence 1 catches throughout the study period. We used the artisanal 0 sector, reef-species breakdown for 1950 1960 1970 1980 1990 2000 the finfish component of the subsistence catch and estimated a Year Figure 3. Total reconstructed catch for Tonga 1950-2007 compared to breakdown for the invertebrate total landings as supplied to the FAO. catch of 80% molluscs and 20% crustaceans.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  125  RESULTS Total reconstructed catch Our estimate of total marine fisheries catches for Tonga, 1950-2007, which included total commercial catches and subsistence catch estimates, was over 323,000 t (Figure 3). This estimate of total marine fisheries catches for Tonga is over 3.5 times larger than the 94,753 t that were reported by the FAO on behalf of Tonga, with major discrepancies prior to the mid-late 1980s. For the more recent years, total reconstructed catches were on average 2 times larger than reported landings. The reconstruction also suggests that total catches peaked in the 1970s and 1980s and has declined since. This pattern is in contrast to the reported data which suggest catches increased until the early 2000 (Figure 3). Our data suggest that the subsistence sector catches 3 times the amount of fish that the commercial sector lands, and accounts for 70% of total marine fisheries extractions (Figure 4). The time series of subsistence catches shows an increasing trend until the early 1970s, followed by a decrease throughout the remainder of the study period (Figure 4).  Mullet catches totalled approximately 250 t in 1950, increased to a peak of approximately 360 t∙year-1 in the late 1960s and then steadily decreased, with a dramatic decline in the late 1980s (Figure 4). We estimated that catches of mullet in the recent period (2000s) were about 3 t∙year-1.  Large Pelagics  7  Deep-slope  Mullet  6  Catch (t x 103 )  Catches for the artisanal sector (mullet and deep slope catches excluded) were estimated to be 54,309 t over the 1950-2007 time period (Figure 4). Of this total, 54% were from the main island of Tongatapu, while the remaining 46% were from the outer islands.  5  4  Artisanal  3 2  Subsistence  1 0 1950  1960  1970  1980  1990  2000  Year  Figure 4. Total marine fisheries catches for Tonga, 1950-2007. Catches include mullet and deep-slope and large pelagic fisheries in addition to estimates of artisanal and subsistence sectors.  The deep slope fishery caught an estimated 5,176 t, mainly of snapper and grouper, between 1980 and 2007 (Figure 4). The deep-slope fishery represented approximately 5% of all commercial sector catches. Large tuna and billfish catches totalled 19,496 t from 1967-2007. Prior to 1967, catches of tuna and billfish were not reported by the FAO as a separate category.  DISCUSSION Our total reconstructed catch estimate for Tongan marine fisheries for the period 1950-2010, was 3.5 times larger than the data reported to FAO on behalf of Tonga suggest. For more recent periods, this discrepancy declined to two times, i.e., 2,300 t∙year-1. Subsistence catches accounted for the largest unreported component of our reconstruction, representing 84% (3,200 t∙year -1 ) and 48% (2,600 t∙year-1) of the total catch for Tonga in 1950 and 2007, respectively, which is in line with the regional estimate for the South Pacific Islands of 80% (Zann and Vuki, 2000). Artisanal catches were estimated using independent data and were approximately 16% (623 t∙year -1) and 24% (1,317 t∙year-1) of the total catch for 1950 and 2007, respectively. Artisanal catch estimates likely did not include the portion of the catch that was sold directly to restaurants, hotels or at the roadside, as catch estimates from Bell et al. (1994) were  based on surveys from fish market sales. Our estimate of artisanal catches may, therefore, have been on the low side. The remaining catch was from the large-scale pelagic fishery, which was zero in 1950 and 28% of the total catch in 2007. Artisanal fisheries (i.e., small-scale commercial) are the main supplier of seafood to the domestic commercial market in Tonga. In 1950, we assumed that the outer islands had no commercially marketed seafood, and all seafood was sourced through subsistence fishing. However, our estimates of artisanal catches on the outer islands for the recent period were substantial, compared to catches on the main island of Tongatapu. This is likely due to the more recent transition on the outer islands towards a cash-based economy. More fish are likely sold, whereas previously, catches on the outer islands were mainly subsistence. Our results suggested that overall, the main island of Tongatapu landed 54% of the artisanal catch, while the outer islands landed the remaining 46%. Mead (1987) provided a similar estimate, with over half of the commercial catch being landed in Tongatapu. The average per capita fish consumption rate was 80 kg∙person-1∙year-1 in 1950, decreasing to 38 kg∙person-1∙year-1 by 2007 when the commercial production for the domestic market (i.e., artisanal catch) was combined with the subsistence catch estimate. This is well within the range presented by Gillet and Lightfoot (2002) of between 14-102 kg∙person-1∙year-1 and for the recent period, close to the range given by Anon. (1993) of 20-50 kg∙person-1∙year-1. Lower consumption rates were presented in the literature (e.g., 23 kg∙person-1∙year-1 derived from Kimura and Fa'anunu [1995]); however, these were likely based solely on commercial production. On the main island, there has been a shift in diet preference away from seafood toward cheaper, imported animal protein alternatives such as mutton flaps and corned beef (Finau et al., 1994). The reason for this shift may have been partly driven by a decline in reef-lagoon stocks due to overexploitation around Tongatapu, in addition to the increasing cost of local seafood. The proportionately smaller artisanal catches on the main island as compared to the outer islands may be the result of this decline in availability and accessibility of marine products on Tongatapu. The large-scale commercial fisheries sector of Tonga expanded substantially in the 1980s, when profitable commercial tuna longline operations began (Bell et al., 1994). Most of the tuna catch is exported, making this fishery the highest contributor to fisheries-based revenue for Tonga. The domestic commercial tuna fishery expanded in the 1980s from one vessel to 33 by 2003 (Likiliki et al., 2005). According to Bell et al. (1994), tuna stocks in Tonga did not seem depleted at that time, and could support an expansion, if this fishery were given sufficient resources. Annual tuna catches consist of 70% albacore (Thunnus alalunga), 20% yellowfin (Thunnus albacares) and 10% big eye (Thunnus obesus; Likiliki et al., 2005). Severe declines in 2003 and 2004 catches due to the effects of El Nino resulted in all foreign vessels leaving Tonga in 2005 (Likiliki et al., 2005; Gillett 2009). Mullet is an important food fish in the Tongan diet (Kimura and Fa‘anunu, 1995). Mullet catches began to decline in the 1970s, followed by dramatic declines in the 1980s and early 1990s due to the use of fish fences (Bell et al., 1994; Kimura and Fa‘anunu, 1995). Both of these sources agree that Mugil cephalus, which made up 70-75% of the mullet catch in the 1980s, was on the verge of local extinction by the 1990s. This study illustrated the importance of the small-scale fisheries sector to Tonga, and the magnitude of under-representation of subsistence catches. Most Tongans, fish more to meet their daily food needs than for commercial purposes (Kronen, 2004); however, this sector is often overlooked in fisheries management. The subsistence sector accounts for a substantial portion of total marine fisheries removals and draws mainly from nearshore, reef resources. Gillet (2009) estimated that the contribution by the subsistence sector to Tonga‘s GDP to be roughly 3%, while the commercial sector contributes roughly 6%. Reef fisheries provide a crucial source of animal protein to the Tongan people, particularly the inhabitants of the outer islands where subsistence fisheries dominate. Thus, reef resources and subsistence fisheries are crucial to national food security. Our estimate of domestic seafood supply, based on artisanal and subsistence catches, ranged from 80 kg∙person-1∙year-1 in 1950 to 38 kg∙person-1∙year-1, in 2007, which is within the range of 14-102 kg∙person-1∙year-1 given in Gillet and Lightfoot (2002), but less than the 102 kg∙person-1∙year-1 presented by Finau et al. (1994). Our estimate of domestic seafood supply was on average higher than some previous estimates (Kent, 1980; Anon., 1993; Kunatuba and Uwate, 1983 in Kimura and Fa‘anunu, 1995); however, their lower consumption estimates were likely based only on  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  127  commercial production. There is agreement, however, that the majority (60-70%) of domestic production is from reef and lagoon resources (Bell et al., 1994; Zann and Vuki, 2000). The health of the Tongan people relies heavily on the health of the reef ecosystem. Although parts of Tonga have switched to a cashed-based economy which relies less on subsistence fishing, alternative economic systems such as bartering still exist on the outer islands and should be considered in resource management decisions and policy development, particularly in terms of food security. For countries such as Tonga, who remain closely tied to the sea for their basic needs, proper accounting of marine fisheries extractions is paramount to their socio-economic stability.  ACKNOWLEDGEMENTS We acknowledge the support of the Sea Around Us Project, a collaboration between the University of British Columbia and the Pew Environment Group.  REFERENCES Anon. (1993). Tonga National Tourism Plan / prepared by Nicholas Clark and Associates for Government of Tonga and Asian Development Bank. Nicholas Clark and Associates, Nuku'alofa. Available at: http://www.adb.org/Documents/Reports/Consultant/24244-TON/24244-08-TON-TACR.pdf [accessed May 21, 2010]. Anon. (2010a) CIA world fact book. Available from: www.cia.gov/library/publications/the-worldfactbook/geos/tn.html [accessed: March 9, 2010] Anon. (2010b). SPC Statistics and Demography. Available from: http://www.spc.int/sdp/index.php?option=com_docman&task=cat_view&gid=28&Itemid=42 [accessed March 30, 2010] Adams, T.J.H. and P.J. Dalzell. 1994. Artisanal fishing. Pacific Islands Biodiversity Issues. Bell, L., Fa‘anunu, U., and Koloa, T. (1994) Fisheries resource profiles: Kingdom of Tonga. Forum Fisheries Agency. 197 pages Chapman, M. (1987) Women‘s fishing in Oceana. Human Ecology 15 (3): 267-288. Evans, M., Sinclair, R.C., Fusimalohi, C., Viliami, L.A., and Freeman, M. (2003). Consumption of traditional versus imported foods in Tonga: implications for programs designed to reduce diet-related non-communicable diseases in developing countries. Ecology of Food and Nutrition. 42(2):153-176. FAO (2002) Available from: www.fao.org/fi/fcp/en/TON/profile.htm [accessed: July 24, 2007] Fifita, V. (1996). Statistics department Tonga. Available from: http://www.pmo.gov.to/tongastats/Population-Census1996.html [accessed March 30, 2010]. Finau, T.L., Udagawa, K., and Nakajo, N. (1994) Fish and meat consumption of Tongan people. Fisheries Research Bulletin of Tonga. 1:29-36. Gillett, R. (2009) Fisheries in the economies of the Pacific island countries and territories. Andaluyong City, Philippines: Asian Development Bank. Gillett, R. and C. Lightfoot. (2002) The contribution of fisheries to the economies of Pacific island countries. Pacific studies series. Asian Development Bank. Forum Fisheries Agency and World Bank, Manila. Kent, G. (1980). The politics of Pacific Island Fisheries. Westview Press: Boulder ,USA. 191 p.Kimura, T., and Fa'anunu, U. (1995) Biological survey and resource management of mullet in Tonga. Inshore Fish. Mgmt. South Pacific Commission. pp.317-322. Kronen, M. (2004) Fishing for fortunes? A socio-economic assessment of Tonga‘s artisanal fisheries. Fisheries Research 70: 121-134. Kronen, M., Clua, E., McArdle, B., and Labrosse, P. (2003) Use and status of marine resources – a complex system of dependencies between man and nature: case studies from Tonga and Fiji, South Pacific. Noumea, New Caledonia: Secretariat of the Pacific Community. Likiliki, P.M., Matoto S.V., Fa'anunu, U. (2005) Tonga tuna fisheries status report. Western and Central Pacific Fisheries Commission. Lovell, E. R. and Palaki, A. (2002) National coral reef status report Tonga. pp 317-343. In Kulbicki, M. a 485 p. Malm, T. (2009) Women of the coral gardens: the significance of marine gathering in Tonga. SPC Traditional Marine Resource Management and Knowledge Information Bulletin #25. pp. 2-15. Mead, P.D. (1987). Deep Sea Fisheries Development Project. Report of the third visit to Tonga. South Pacific Commission, Noumea, New Caledonia. 49 pages. Minister of Fisheries. 1997. Report of the Minister for Fisheries. Government of Tonga. Ministry of Fisheries. 2007. Tonga national snapper & grouper fisheries management plan. SPC (1988) Twentieth regional technical meeting on fisheries: country statement – Tonga. WP.23.  Wilkinson, W. (1973). Fisheries developments in Tonga. South Pacific Islands Fisheries Newsletter No. 9, July 1973, South Pacific Commission , Noumea, pages 38-43, (Fish Lib, SPC Lib). Zann, L.P. (1994) The status of coral reefs in South Western Pacific islands. Marine Pollution Bulletin. 29: 52-61. Zann, L.P., Vuki, V.C. (2000) The Status and Management of Subsistence Fisheries in the South Pacific. Ocean Yearbook No. 14. The International Ocean Institute, The University of Chicago Press. Zeller, D., Booth, S., Craig, P. and Pauly, D. (2006a) Reconstruction of coral reef fisheries catches in American Samoa, 1950-2002. Coral Reefs 25: 144-152. Zeller, D., Booth, S., Davis, G. and Pauly, D. (2007) Re-estimation of small-scale fisheries catches for U.S. flag island areas in the Western Pacific: The last 50 years. Fishery Bulletin 105: 266-277. Zeller, D., Booth, S. and Pauly, D. (2006b) Fisheries contributions to GDP: Underestimating small‐scale fisheries in the Pacific. Marine Resource Economics 21: 355‐374.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  129  Appendix Table A1. FAO landings vs. total reconstructed catch (in tonnes) for Tonga, 1950-2007, in metric tonnes. Year FAO landings Total reconstructed catch 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  200 200 200 200 300 300 300 300 300 300 400 400 400 400 400 500 500 581 548 555 444 533 532 626 753 928 1,038 1,216 1,210 2,000 1,994 2,093 2,229 2,365 2,502 2,690 2,952 2,724 2,692 2,664 1,616 1,915 2,217 2,282 2,435 2,530 2,826 2,763 3,937 4,029 3,545 4,332 4,493 4,107 1,414 1,632 2,041 1,927  3,830 3,944 4,056 4,167 4,278 4,387 4,495 4,647 4,798 4,947 5,094 5,240 5,385 5,528 5,670 5,811 5,950 6,110 6,156 6,240 6,306 6,297 6,294 6,282 6,273 6,259 6,233 6,160 6,065 5,988 6,328 6,247 6,267 6,218 6,133 6,070 6,008 5,608 5,425 5,245 5,031 4,890 4,767 4,645 4,648 4,596 4,598 4,570 4,724 4,930 5,054 5,499 5,518 4,815 4,307 4,468 4,597 4,599  Appendix Table A2. Total reconstructed catch (in tonnes) for Tonga by major taxa, 1950-2007. Others grouping includes 81 taxa. Year 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007  Misc. molluscs 1,283 1,317 1,351 1,384 1,417 1,450 1,482 1,530 1,577 1,623 1,670 1,715 1,761 1,806 1,851 1,895 1,939 1,964 1,988 2,011 2,035 2,029 2,022 2,013 2,002 1,991 1,994 2,003 1,975 1,946 1,917 1,887 1,856 1,825 1,793 1,760 1,727 1,694 1,660 1,625 1,591 1,556 1,520 1,432 1,376 1,319 1,261 1,203 1,179 1,154 1,130 1,129 1,136 1,138 1,139 1,120 1,111 1,104  Scarus spp.  342 352 363 373 384 394 405 419 433 447 461 474 488 501 515 528 541 549 556 564 571 574 576 577 578 579 580 575 571 566 560 555 550 544 538 531 525 518 506 493 480 466 451 436 428 421 413 406 404 402 400 398 396 394 392 390 388 386  Misc. crustaceans 321 329 337 345 353 361 369 381 392 404 415 426 437 448 459 470 481 487 493 498 504 502 500 498 495 492 488 480 472 463 454 446 437 427 418 409 399 389 379 369 359 349 339 329 318 308 297 286 284 282 279 277 275 272 270 268 265 263  Lethrinus harak  Naso unicornis  Acanthurus  215 222 229 235 242 248 255 264 273 281 290 299 307 316 324 332 341 345 350 355 360 361 362 363 364 365 365 362 359 356 353 350 346 342 339 335 331 326 319 311 302 293 284 274 270 265 260 256 254 253 252 250 249 248 247 245 244 243  166 171 176 181 187 192 197 204 210 217 224 230 237 244 250 256 263 267 270 274 278 279 280 281 281 282 282 280 277 275 272 270 267 264 261 258 255 252 246 240 233 226 219 212 208 205 201 197 196 195 194 193 192 191 190 189 188 188  162 168 173 178 183 188 192 199 206 213 219 226 232 238 245 251 257 261 265 268 272 273 274 275 275 276 276 274 271 269 267 264 261 259 256 253 250 246 241 235 228 222 215 207 204 200 197 193 192 191 190 189 188 187 186 185 184 184  spp.  Leptoscarus vaigiensis 143 147 152 156 161 165 169 175 181 187 193 199 204 210 215 221 226 230 233 236 239 240 241 242 242 243 243 241 239 237 235 232 230 228 225 222 220 217 212 206 201 195 189 182 179 176 173 170 169 168 167 167 166 165 164 163 162 162  Thunnus alalunga 0 0 106 143 135 174 206 252 242 195 152 171 199 231 343 379 431 493 616 801 862 1,268 1,189 611 182 283 414 390  Others 1,199 1,239 1,278 1,317 1,356 1,394 1,431 1,483 1,534 1,584 1,633 1,682 1,730 1,778 1,825 1,871 1,917 1,980 2,005 2,031 2,055 2,058 2,060 2,060 2,059 2,058 2,054 2,035 2,014 1,993 2,392 2,405 2,320 2,298 2,329 2,317 2,317 2,566 2,302 2,218 1,959 1,984 1,735 1,677 1,690 1,693 1,774 1,811 1,927 2,027 2,118 2,483 2,381 2,340 2,264 2,362 2,536 2,604  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  131  RECONSTRUCTION OF MARINE FISHERIES CATCHES FOR TUVALU (1950-2009)1 Kendyl Crawforda, Sarah Harperb and Dirk Zellerb a Department  of Marine and Environmental Science, Hampton University Hampton, Virginia 23668, USA b Sea Around Us Project, Fisheries Centre, University of British Columbia 2202 Main Mall, Vancouver, V6T 1Z4, Canada kendyl.crawleycrawford@my.hamptonu.edu; s.harper@fisheries.ubc.ca; d.zeller@fisheries.ubc.ca  ABSTRACT Tuvalu‘s total marine fisheries catches within its EEZ were reconstructed for the years 1950 to 2009. This reconstruction accounts for officially un- and underreported catches of artisanal and subsistence fishery sectors as well as the baitfish used in the pole-and-line tuna fishery. FAO data were used in combination with data from fish markets, regional reports and consumption data. Total reconstructed catches were estimated to be 69,631 t over the six decades, which is approximately 5 times larger than the amount reported by the FAO on behalf of Tuvalu (12,241 t). Total catches increased from 813 t·year-1 in 1950 to 1,607 t·year-1 by 2009. The majority of total catches were from the subsistence sector (87%). This investigation reveals the need for an improvement in the accounting of marine fishes catches by all fisheries sectors. Due to the heavy rates of fish consumption in Tuvalu, reliable estimation of catches and resulting resource management decisions will play a role in Tuvalu‘s future food security.  INTRODUCTION Tuvalu is an archipelago in the South Pacific consisting of nine atolls; Nanumea, Niutao, Nui, Vaitupu, Nukufetau, Nukulaelae, Niulakita and Funafuti. Tuvalu is located at 8° 31' S, 179° 13' E, approximately halfway between Australia and Hawaii in the south central Pacific (Figure 1). The country‘s total land area of 26 km2 is tiny in comparison to its nearly 752,000 km2 Exclusive Economic Zone (EEZ) (www.seaaroundus.org). Tuvalu is critically vulnerable to sea level rise due to its low-lying topography, with most of the country less than 3m above sea level (Connell, 2003; Sauni and Fay-Sauni, 2005; Rayfuse, 2011; Stephen, 2011). The continental shelf off Tuvalu is minimal; there are patch and fringing barrier reefs immediately surrounded by 1,000 m depths. The inner lagoons provide the only significant shallow water areas (Sauni and Fay-Sauni, 2005).  Figure 1: Map of Tuvalu and its Exclusive Economic Zone.  Formerly a British colony known as the Ellice Islands, Tuvalu gained its independence in October 1978. The country has been politically stable and its economy has grown from an initial Tuvalu Trust Fund investment of $ 27 million Australian Dollars (AU$) in 1987 to AU$ 66 million in 2002 (Gemenne and Shen, 2009). As of 2002, the Gross Domestic Product (GDP) of Tuvalu was AU$ 26.9 million, of which 8.2% was from the artisanal fishing industry (Gillett, 2009). The Tuvaluan economy is considered traditional and predominantly non-cash (Sauni and Fay-Sauni, 2005). Marine products, wages and 1  Cite as: Crawford, K., Harper, S. and Zeller, D. (2011) Reconstruction of marine fisheries catches for Tuvalu (1950-2009). pp. 131143. In: Harper, S. and Zeller, D. (eds.) Fisheries catch reconstructions: Islands, Part II. Fisheries Centre Research Reports 19(4). Fisheries Centre, University of British Columbia [ISSN 1198-6727]  remittances sent by family members working overseas - many on foreign fishing vessels - are considered to be the most important sources of income (Sauni and Fay-Sauni, 2005). In the past ten years, annual revenue from foreign fishing fleets has varied, providing between 5.5% and 36.7% of total government revenue and grants (Gillett, 2009). In 1999, Tuvalu received USD$ 5.9 million in foreign fishing access fees. This amount comprised 42.6% of its GDP (Gillett and Lightfoot, 2002). The revenue from foreign tuna fishing is approximately 14.6% of the total value of the fish caught (Gillett, 2009). With Tuvalu receiving such a comparatively small percentage of the value of its fish, the government continues in their attempt to develop commercial fisheries in its vast EEZ, and stimulate economic growth. However, the development of larger-scale commercial fisheries in Tuvalu is hindered by high costs, difficulty raising funds and a lack of infrastructure required for fleet maintenance and operations, processing, internal distribution and export (Gillett, 2002). The largest component of Tuvalu‘s fishing activities is subsistence, i.e., for direct consumption. More than 80% of domestic coastal catch in Tuvalu is produced by subsistence fishing (Gillett, 2010). Dalzell (1996) estimated that subsistence and artisanal fisheries make up 87% and 13%, respectively, of Tuvaluan coastal fisheries production. A recent Household Income and Expenditure Survey revealed that fishing contributes to 8% of personal income, after wages and remittances sent from overseas (Anon., 2006). Data on the artisanal sector is incomplete due to the high occurrence of informal bartering (Lambeth, 2000; Poulasi, 2008; WCPFC, 2009, 2010). Fishermen or their wives sell their catch either from home, the roadside using handcarts and ice chests or in small markets (Lambeth, 2000; Gay, 2010). As a result, catch data for the subsistence and artisanal fisheries are largely unknown. As more Tuvaluans try to make the change to working for an income, the artisanal sector is assumed to grow (Sokimi and Chapman, 2005). The heavy dependence on fish for animal protein is evident through the unique Tuvaluan word ‗miti’ which describes a craving specifically for fish. In 2004 and 2005, the urban per capita fish consumption in Tuvalu was 68.8 kg·person-1∙year-1 with 97% of that amount being fresh fish. For rural areas, the per capita consumption was 147.4 kg∙person-1·year-1 (99% being fresh fish) (Gillett, 2009). Tuvalu‘s seafood consumption rate is among the highest in the world (Gillett and Lightfoot, 2002; Gillett et al., 2001). The island communities of Tuvalu are distinguished not only for their fishing ability, but also their rich knowledge of their environment (Gay, 2010). The close relationship between Tuvaluans and the ocean is readily apparent through their dependence on fish for food security. The island of Niutao, one of the main islands of Tuvalu, has one of the highest population to reef area densities in the region, with 246 people per km2 of reef (Adams et al., 1996). At the same time, on the atoll of Funafuti, there was a lower population to reef area density of 165 people per km 2 of reef (Adams et al., 1996). However, Funafuti is now home to approximately 5,000 people, 47% of the country‘s population (Sauni et al., 2008). Even though the urban island has set up management measures such as the 33 km2 Funafuti conservation area, it is highly urbanized and as such faces problems such as sewage treatment and waste disposal. There is heavy fishing pressure from subsistence needs on this island with many people fishing after work and on the weekends (Sauni et al., 2008). It is estimated that 93% of households on Funafuti eat fresh fish that they catch and 70% of households eat fresh invertebrates that they catch (Sauni et al., 2008). Fish catch rates have increased when compared to estimates from previous years (Sauni and Fay-Sauni, 2005). Thus there is concern for the sustainability of inshore resources in light of the increased fishing pressure and population growth on Funafuti (Adams et al., 1996; Gillett, 2002; Sauni and Fay-Sauni, 2005; Aylesworth and Campbell, 2009). Fishing in Tuvalu uses a range of techniques including pole-and-line, trolling and reef gleaning, which are used to collect finfish, bivalves, crustaceans and other invertebrates in nearshore and offshore Tuvaluan waters. The fishing roles on Tuvalu, like many other Pacific Islands are divided by gender, with women mainly reef gleaning at low tide, and processing, and men fishing both inshore and offshore. The introduction of outboard engines on canoes in the 1960s and 1970s has made fishing much easier, and consequently, women have felt less of a need for their auxiliary fishing activities (Lambeth, 2000). However, when men are unable to fish because of the weather, women‘s collecting activities are vital (Chapman, 1987). While fishing techniques vary among the different islands, the main gears used are gillnetting, handlining, castnetting, pole-and-line and spearfishing. Handlines are used to catch demersal fish on the reef (Sauni and Fay-Sauni, 2005). On the outer reefs, spears, handlines, scoop nets and deep bottom methods such as deep-bottom droplining are used (Chapman and Cusack, 1990; Sauni et al., 2008). As of 1991, there were an estimated 200 motorized and 500 non-motorized vessels, most less than 10 m (Gillett, 2003). On the most populated island of Funafuti, there are 10 to 20 commercial vessels (4-5  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  133  m) engaged in trolling for mainly skipjack and yellowfin tuna (Gillett, 2003) and some line fishing for reef species (Gillett, 2002). Fisheries in Tuvalu are dominated taxonomically by skipjack (Katsuwonus pelamis) and yellowfin tuna (Thunnus albacares), which in 1978 made up half of the total fish catch (Gillett et al., 2001). A more recent estimate suggests skipjack and yellowfin represent approximately 75% of total fish landings (Gillett, 2002). Other pelagic fish such as flying fish (Cypselurus spp.) also represent a substantial part of the catch. Flying fish, usually caught at night with the use of scoop nets and lights (Gillett, 2002), are commonly used as baitfish in the tuna pole-and-line fishery (Aylesworth and Campbell, 2009). Fish inhabiting the lagoon and reef habitats such as red snapper (Lutjanus gibbus) are also caught but make up a lesser portion of the catch. Bycatch is generally consumed, bartered or given away as a part of the fishery. In the early 1980s, the government established the National Fishing Corporation of Tuvalu (NAFICOT) tasked with the goal of developing industrial fisheries (Sokimi and Chapman, 2005; Gillett, 2011a). One of the roles of NAFICOT was to manage national fishing vessels. In 1982, Japan donated a pole-and-line vessel, in 1989, Japan donated 7 additional vessels (6 launches and 1 extension vessel), and in 2004, Korea donated two longliners. Community fishing centers (CFC) were developed in the 1990s with foreign aid to provide an income opportunity for fishers and also to redistribute the excess supply of fish to the urban population center of Funafuti through NAFICOT. The CFCs provide salting, drying and at times cold storage facilities. However, ice is not usually present in many of the CFCs or on local fishing vessels because of a lack of transportation infrastructure and water availability (Aylesworth and Campbell, 2009). Presently, most CFCs have fallen into disrepair and rely heavily on government subsidies. In 2009, NAFICOT went bankrupt when the government decided to cease financial support (Gay, 2010). This same year, a joint venture was established between Tuvalu and Taiwan; the first purse seiner flying the Tuvaluan flag, the FV Taumoana began fishing in August of that year in Tuvalu, FSM and Kiribati, landing a total of 4,877 t of tuna, most likely skipjack and yellowfin (WCPFC, 2010). The purpose of this study is to reconstruct total marine fisheries catches by Tuvalu within its EEZ between 1950 and 2010 by accounting for all fishing sectors, as a baseline for the assessment of food security and resource availability. During the completion of this study, it was necessary to make assumptions in order to fill large gaps in data availability because the official reported data includes neither taxonomic specificity nor quantitative detail, especially for the pre-1980 time period.  MATERIALS AND METHODS  Population (x 103)  The Tuvalu Fisheries Department collects offshore commercial catch data for the national fleet. In addition, the department actively collects monthly reports from various fish markets reporting sales and purchasing information. In 2010, Tuvalu started a national catch database for their inshore fisheries (T. Poulasi, pers. comm., Tuvalu Fisheries Department). Unfortunately, data from 2010 was not yet available at the time of this study. Catch 12 data other than for the national fleet for all years prior to 2010 10 were not collected by the Vaitupu national Fisheries Department Nukufetau 8 (T. Poulasi, pers. comm., Tuvalu All Others Niutao Fisheries Department). The 6 artisanal data collected from fish market reports are limited, as 4 they encompass only the small amount of catch sold within the Funafuti markets (WCPFC, 2009), 2 whereas much catch is sold informally. Data on subsistence 0 fishery catches have not been 1950 1960 1970 1980 1990 2000 collected at all. The national data Year are considered insufficient for the evaluation and monitoring of Figure 2. Human population of Tuvalu by major islands, 1950fishing activities (Chapman, 2009. 2004). However, recent attempts  have been made to improve data collection. To provide the best picture of Tuvaluan fisheries, data from FAO FishStat, regional reports and independent assessments were accessed and used to develop data anchor points for the estimation of total catches. Interpolations between anchor points were used to derive a complete time series, using a catch reconstruction approach developed by Zeller et al. (2006; 2007).  Human Population Data Population data were obtained in order to convert available per capita consumption rates into an estimate of overall demand. National population data were obtained from Populstat (www.populstat.info/ [date accessed: 20 July 2011]) prior to 1997, from the 2008 Biannual Statistical Report (Anon., 2008) for years 2002 to 2007 and from Index Mundi (www.indexmundi.com [date accessed: 20 July 2011]) for 2009. A linear interpolation was used between years of known population data in order to obtain a complete time series. Population data by island were obtained for Funafuti, Niutao, Nukufetau and Vaitupu from Populstat for 1985, 1987 and 1996, a Household Income and Expenditure Survey for 2005 (Anon., 2006) and City Population (www.citypopulation.de [date accessed: 20 July 2011]), for 1979, 1991 and 2002. Over the past 50 years, migration to Funafuti from the other islands has resulted in nearly 50% of the current total population residing on Funafuti, driven mainly by the availability of government jobs (Connell, 2003). The earliest population data found for Funafuti was for 1963 obtained from Populstat. The percentage of the total population which resided in Funafuti in 1963 was applied back in time to 1950. From 1964 to 2005, the population of Funafuti was determined through interpolation of data points. For Niutao, Nukufetau, Vaitupu and all others, the earliest year of data was 1979. Therefore, we calculated the proportion that each of these islands represented and applied these same proportions back to 1950. After 1979, interpolation was done between the island population data. From 2005 to 2010, all island populations were calculated based on 2005 percentages (Figure 2).  Fishing in Tuvalu  Catch (t x 103)  4.5 The FAO FishStat database was used as the official catch data 4.0 for Tuvalu (Figure 3). However, 3.5 Total FAO FAO data are presented by FAO 3.0 area and do not delineate the amount taken within EEZ 2.5 areas. This report aims to 2.0 reconstruct the catches taken 1.5 by Tuvaluans within their waters; therefore, it was Allocated to EEZ 1.0 necessary to disaggregate the 0.5 catch taken in Tuvalu‘s EEZ from that taken outside. A 0.0 significant component of 1950 1960 1970 1980 1990 2000 fishing recorded in the FAO Year data occurred outside of Tuvalu‘s EEZ. For example, in Figure 3. Catches presented by the FAO on behalf of Tuavlu and 1982, Tuvalu fished in a allocation of FAO data to the Tuvaluan EEZ based on assigning 90% of partnership with the Ika reported tuna catches to be taken outside EEZ waters, 1950-2009. Corporation in Fijan waters (SPC, 1994). Tuvalu also fished in Fiji and Solomon Islands in 1987 and 1988 with a peak catch of 1,091 t from the Solomon Islands in 1988 (SPC, 1994; Sauni et al., 2008). In 2009, Tuvalu in a joint venture with the Fong Haur fishing company of Taiwan fished in the Federated States of Micronesia and in Kiribati (WCPFC, 2010). Due to the large proportion of fishing known to occur outside of Tuvalu‘s EEZ (SPC, 1994; Apinelu, 2004; Sauni et al., 2008; WCPFC, 2010), we assumed 90% of the FAO tuna catch (bigeye (Thunnus obesus), skipjack (Katsuwonus pelamis), yellowfin (Thunnus albacares) and tuna-like) was taken outside of Tuvalu‘s waters (Figure 3). Remaining FAO landings (ie., marine fishes nei) were assumed to have been from small-scale fisheries within Tuvalu‘s EEZ.  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  135  Small-scale fisheries FAO FishStat presents landings of miscellaneous marine fishes from 1950 to 2009. Prior to 1982, FAO landings for Tuvalu are less than 0.5 t·year-1 and landings for tuna and tuna-like fishes during this period are zero. Catch amounts or rates from national or independent sources were not readily available for either the artisanal or subsistence sectors; therefore per capita consumption rates were used to determine the overall fresh fish demand for Tuvalu (Gillett and Lightfoot, 2002; Sauni et al., 2008; Gillett, 2009). This overall demand was compared to the reported supply to determine the magnitude of underreporting. For 2004, national fresh fish consumption in Tuvalu was calculated using consumption rates for the islands of Funafuti, Nukufetau, Niutao and Vaitupu (Sauni et al., 2008) (Table 1). For the remaining five islands not represented individually, here referred to as ‗all others‘, the average across the four individual islands (i.e., 151.0 kg·person-1·year-1) was used (Sauni et al., 2008). To derive a nation-wide consumption rate, a 2004 weighted average for all islands was calculated (145 kg·person-1·year-1). This rate was used for all of Tuvalu for 2004 and was carried forward, unaltered to 2010 (Gillett and Lightfoot, 2002; Sauni et al., 2008). The 1950 fresh fish consumption rate was assumed to be Table 1. Fresh fish consumption rates per similar to the rate for the other islands in 2004 (see All island for 2004 (Sauni et al., 2008). others Table 1) with the addition of 5 kg to account for Island Fresh fish consumption limited imports of protein alternatives available in the later rate (kg/person/year)a period. The resulting per capita consumption rate for 1950 Funafuti 135.0 was 155.6 kg·person-1·year-1. A linear interpolation between Nukufetau 117.8 these two anchor points between the 1950 and the 2004 Niutao 185.3 per capita rates was used to complete the time series. The Vaitupu 162.5 consumption rates were then combined with the human All Others 150.6 population data to estimate overall demand of fresh fish. a -1 -1  weighted average for all islands 145 kg∙person ∙year .  To determine the proportion of the total demand supplied by each sector, Dalzell‘s (1996) breakdown of 13% artisanal and 87% subsistence was used for 1996 to 2009. In 1950, we assumed 100% of the catch to be subsistence and interpolated to 87% subsistence in 1996. The annual artisanal production estimates were then checked to ensure that they exceeded the artisanal landing amounts given in annual reports by the Western and Central Pacific Fisheries Commission (Poulasi, 2008; WCPFC, 2009, 2010). This was necessary as the WCPFC reports reflect only the small fraction of catch landed and received by the CFCs. Fishers normally sell their catch directly to consumers; these transactions are not documented (Poulasi, 2008).  Invertebrates The people of Tuvalu also consume invertebrates as part of Table 2. Invertebrate consumption rates per their diet. However, unlike for fresh fish, there is no local island 2004. commercial fishery for invertebrates (Sauni et al., 2008). Island Invertebrate consumption Thus, an annual subsistence invertebrate demand was rate (kg/person/year) calculated with invertebrate consumption data from the Funafuti 4.6 Secretariat of the Pacific Community Coastal Fisheries Nukufetau 5.6 Program (www.spc.int/coastfish/ [date accessed: 15 July Niutao 3.6 2011]) (Table 2). Similar to the fresh fish consumption data, Vaitupu 0.9 invertebrate consumption data were available for Funafuti, All Others 3.7 Nukufetau, Niutao and Vaitupu. A national average per capita invertebrate consumption rate was calculated to represent the remaining islands. To determine the annual invertebrate demand, the same methodology was used as in determining total demand of fresh fish (see Small-scale fisheries). The weighted average consumption rate for invertebrates (3.8 kg·person-1·year-1) was used for 2004 and all years following. For 1950, the national average was used with an adjustment of 2 kg to account for the limited availability of alternate protein sources at that time (i.e., 5.8 kg·person-1·year-1). An interpolation was done between the 1950 (5.8 kg·person-1·year-1) and 2004 (3.8 kg∙person-1·year-1) anchor points and the 2004 rate was carried forward unaltered to 2010.  Taxonomic breakdown A substantial portion of both the subsistence and artisanal catch consists of tuna (Gillett et al., 2001). A thorough review of the scientific and grey literature provided numerous estimates of the tuna component  of the Tuvaluan catch. Sauni (2008) presents annual tuna landings data for the 1970s of 350 t·year-1. This amount, however, is viewed to be slightly overestimated (Eginton and Mead, 1978). Gillett (2001) reports that 50% of fish sold in Funafuti is tuna. The Pacific Island Fisheries Regional and Country Information report suggested that 75% of all fish landings are ocean species, mainly skipjack and yellowfin tuna (Gillett, 2002). Based on these information sources, an estimate of 300 t was used as an anchor point for 1975. This represents 32.5% of the total demand for 1975. We assumed that tuna consumption was similar in 1950, and allocated 32.5% of the total 1950 catch to tuna. The species composition for the tuna was derived from yearly reports provided by the Tuvaluan government to the Western and Central Pacific Fisheries Commission (Poulasi, 2008; WCPFC, 2009, 2010). The proportion of skipjack to yellowfin tuna represented in the catch were calculated by averaging annual catch data derived from fish market sales for the years 2003 to 2008. To account for other pelagic species, a small portion of the tuna catch (arbitrarily assigned as 10%) was allotted to miscellaneous large pelagic. The remaining 90% were assigned as 54% skipjack and 36% yellowfin tuna (Table 3). Catches were also broken down taxonomically according to fishing sector, either subsistence or artisanal. For the artisanal sector, species compositions given by Dalzell (1996) were used to taxonomically disaggregate all families aside from Scombridae (Appendix A1). The Etelinae subfamily (Family Lutjanidae) was divided into Etelis spp. and Pristipomoides spp. with each genus receiving half of the Etelinae percentage, to improve taxonomic resolution. Scombridae was excluded, because the tuna component of the artisanal fishery had already been calculated. The invertebrate and fresh fish demands were kept separate for the subsistence sector. The taxonomic breakdown for the subsistence fishery was determined through fish (Appendix Table A2) and invertebrate (Appendix Table A3) catch compositions available in a national report (Sauni et al., 2008). Species catch composition data were available for the islands of Funafuti, Niutao, Nukufetau and Vaitupu for several different habitats such as lagoon, outer reef, sheltered coastal reef and intertidal reef flat. Some species groups that composed a very small percentage of the total catch were grouped and represented at the family level. The invertebrate catch compositions were applied to the invertebrate catch and the fish compositions applied to the fish catch derived from the demand estimates. Table 3. Taxonomic breakdown of tuna for Tuvalu. Source: (Anon., 1984) Group Taxon Common name Scombridae Scombridae Miscellaneous large pelagics  Katsuwonus pelamis Thunnus albacares Misc. pelagics  Skipjack tuna Yellowfin tuna Misc. large pelagics  Proportion of catch (%) 54 36 10  Baitfish Associated with the pole-and-line fishery for tuna is the use of baitfish, rarely accounted for in fisheries statistics. Baitfish fisheries often operate in parallel to the skipjack pole-and-line fisheries, utilizing the same vessels. Tuvalu‘s baitfish resources are extremely limited (Gillett, 2011b). In the past, Tuvalu‘s only pole-and-line fishing vessel, Te Tautai, was forced to fish outside of the Tuvaluan EEZ because of the low availability of baitfish (Gentle, 1991).In addition, baitfish resources are more variable around atolls than high islands (Anon., 1984), and atolls like Tuvalu have been less reliable sources of bait. A regional poleand-line ratio of tuna to baitfish was presented by Gillett (2011b) as 32:1. The dominant species used as baitfish in Tuvalu are Spratelloides delicatulus, Archamia lineolata, Bregmaceros spp. and Atherinomorous lacunoaa (Anon., 1984). The amount of baitfish used was calculated by taking the estimated tuna catch and applying the 32:1 tuna to baitfish ratio provided by Gillett (2011b) for the region and the taxonomic composition derived from the Tuna Programme (Anon., 1984) was applied (Table 4). Table 4. Taxonomic breakdown of baitfish for Tuvalu. Source: (Anon., 1984) Group Taxon Common name Proportion of catch (%) Clupeidae Spratelloides delicatulus Delicate round herring 92 Atherinidae Atherinomorus lacunosus Hardyhead silverside 2 Misc. marine fishes Misc. marine fishes Misc. marine fishes 6  Bêche-de-mer As sea cucumber are not a part of the Tuvaluan diet, they have traditionally received little interest by fisheries managers (Belhadjali, 1997). However, a small bêche-de-mer export industry has developed in  Fisheries catch reconstructions: Islands, Part II. Harper and Zeller  137  Catch (t)  Tuvalu and has become the 1800 island‘s main fishery export item since the late 1970s (Gay, 2010). 1600 The industry began after the 1400 Fisheries Department became the 1200 recipient of UN Development Total Reconstructed 1000 Program funding in 1978. The first 800 export occurred in 1979, 1.8 t of bêche-de-mer sold to Fiji 600 (Belhadjali, 1997). With widely 400 Reported FAO varying production amounts, 200 within EEZ export continued from 1980 to 0 1983 (Belhadjali, 1997). 1950 1960 1970 1980 1990 2000 Remaining stagnant for a decade, export restarted from 1993 Year through 1995 supplying Singapore Figure 4. Total reconstructed catch from 1950-2009 for Tuvalu and Fiji. In 2007, harvest in compared to the subset of FAO data assigned to Tuvalu EEZ. Nukufetau, Nukulaelae and Funafuti began once again with exports to Hong Kong. Only the islands of Funafuti and Nukufetau have suitable habitats for the most profitable sea cucumbers, therefore the sustainability of the bêche-de-mer fishery is of concern (Gay, 2010). The Fisheries Department does not require data on bêche-de-mer catch or exports to be submitted (Gay, 2010). Export weights, nevertheless, were found for all years Table 5. Bêche-de-mer catch from prior to 2007. Gillett (2009) provided a bêche-de-mer export value of 1950-2009. AU$5000 for 2007. From 2007 to 2009, that same export value was Year Catch (t) used. The monetary value was converted to