INVESTIGATION ON TUNA FISHERIES ASSOCIATED WITH FISH AGGREGATING DEVICES (FADs) IN INDONESIA FMA 572 AND 573

The use of anchored fish aggregating devices (a-FADs) in the Indonesia-Indian Ocean has increased rapidly. Since 2004, the Government of Indonesia has issued various FAD related fisheries regulations; however, its implementation was difficult, largely due to the lack of such information. As an endeavor to improve the management of tuna fisheries associated with FADs in this area, an Indonesia–Australia research collaboration project conducted a port sampling program from November 2013 to December 2015 in three key fishing ports in the western Indonesia, i.e., Pelabuhanratu (West Jawa), Muara Padang (West Sumatera), and Bungus (West Sumatera). Data were collected through daily enumeration and interviews with skippers, which consisted of catch, trip duration, biological data, and number of FADs visited. These data were analyzed to estimate catch rate, success rate, and length frequency distribution. The success rate of hand line/trolling line (HL/TR) at Muara Padang showed much lower than that at Pelabuhanratu. This may be due to more a-FADs or higher density in the Padang region, competing with purse seine (PS) boats operating in the same area, than those in the Pelabuhanratu region. The species composition caught by HL/TR and PS associated a-FADs in Indonesian FMA 572 and 573 include skipjack (SKJ, Katsuwonus pelamis), yellowfin (YFT, Thunnus albacares) and bigeye tuna (BET, T. obesus). A large proportion of the SKJ, YFT and BET caught at both Indonesian FMA 572 and 573 were juvenile fish, below the reported length at maturity (L m ) for those species.


INTRODUCTION
Utilization of floating objects (logs, seaweed, etc.) as known attractors of fish has been a feature of artisanal and coastal fishing in the waters of the Western and Central Pacific Ocean (WCPO) for hundreds of years (Kakuma 2000, Morales-Nin et al. 2000. The 1970s marked the beginning of fishers deploying floating fish aggregating devices (FADs) in deep-water in the eastern Indonesian waters to attract and catch tunas. Anchored FADs (called as a-FADs) in waters as deep as 2000 -2500 m have since become a dominant practice for tuna fishing in Indonesia's archipelagic waters, including those in the western Indonesia. With the aim of increasing the efficiency of catch by purse seine (PS) boats in Prigi Bay, East Java (FMA 573), Indonesia's (RIMF) conducted a trial research with deployment of a-FADs in 1986. In 1992, with similar purposes, the RIMF deployed two a-FADs in Binuangen Cape waters (FMA 573) at the depth of 500-600 m for hand line (HL) and troll-line (TR) or HL/TR fisheries, two a-FADs in Cempi Bay, Nusa Tengara Barat (FMA 573), for PS fisheries, and two a-FADs in Semangka Bay, Lampung, and Pesisir Selatan waters in West Sumatera (FMA 572), for HL/TR fisheries (Linting et al., 1992). Unfortunately RIMF did not monitor the results of the a-FAD after deploying but obtained information that some fishers subsequently tried to develop similar a-FADs in the trial areas. Since the early 2000s, the use of a-FADs has been developed rapidly in the FMAs 572 and 573, including in the waters near Pacitan, East Java (FMA 573) (Nuraini et al., 2014). The a-FADs in deep-water have become an integral component of tuna fishing in Indonesian waters. Figure 1 shows

MATERIALS AND METHODS
Fishery data were obtained through daily enumeration including interviews with skippers at the earliest opportunity right after their catches were unloaded, and also by direct observations and biological sampling during November 2013 to December 2015. Fishery information on numbers and positions of a-FADS were obtained from local port authorities, fisheries offices, fishing companies and fishing association representatives. The enumeration carried out at three key tuna fishery bases where the fishers fish in FMAs 572 and 573, i.e., Muara Padang in Padang City, Bungus Fishing Port in West Sumatera, and Pelabuhanratu Fishing Port in West Java. The primary tuna fisheries associated with a-FADs based at both Pelabuhanratu Fishing Port and Muara Padang included HL and TR. On occasion, the HL and TR boats from Muara Padang also unload their catch at Bungus Fishing Port as did PS boats from Fishing Port of Sibolga (North Sumatera). At the three key tuna fishery bases, it is generally found that one boat switches both HL and TR fishing gears subject to season, prevailing sea conditions and catch success, and such boat is registered as a kapal tonda (TR boat).
The types of data and information collected from interviews with boat skippers include technical aspects of a-FADs, number and positions of a-FADs, tuna fisheries associated with a-FADs, operational aspects of the fisheries using a-FADs includes a-FAD visit success rate and catch rate (nominal catch per unit effort or CPUE) as well. The types of data collected from biological sampling include catch composition and individual tuna species size (cm-fork length or cm-FL). The sampling was done on a subsample of the catch at time of catch unloading or at point of auction or sale. The tuna species include skipjack or SKJ (Katsuwonus pelamis), yellowfin tuna or YFT (Thunnus albacares) and bigeye tuna or BET (T. obesus). Data were first recorded onto hard-copy Landings and Biological Samplings Forms and later entered into a project specific database (Oracle/ Apex), FAD Fisheries Database.
The a-FAD visit success rate is defined as the number of successful a-FAD visits, with respect to fish caught or not caught during the visit, expressed as a percentage (%) of total number of a-FADs visited during the fishing trip. Catch rate (nominal catch per unit effort or CPUE) is calculated to indicate the productivity of each boat of fisheries associated with a-FAD/s using the following equation: Where; C = total catch of tuna (kg) E = effort (day-fishing).
Length distribution of each tuna species was compared with length at first matured (L m ) to investigate the tuna fishing activities are still applying the sustainable practices (comparing the proportion of tuna catch relative to its L m ).

Technical Aspects of Indonesian Anchored FADs
The common fish aggregating devices (FADs) deployed by tuna fisheries in Indonesia are anchored FADs (a-FADs). The Indonesian a-FADs have four key components: i) the surface float (bamboo raft, steel pontoon or polystyrene block), ii) the mainline to seafloor (polypropylene rope), iii) a subsurface attractor (nypa or coconut leaf), and iv) the anchor ( Figure 2). In general, the a-FAD surface floats are not equipped with navigation aids (no radio signal emitters or radar reflectors), but in some cases the surface floats are attached to a superstructure such as a flag to make the a-FAD more visible.
In the eastern Indonesia, the most sophisticated type of a-FAD, which is relatively common, is made of bamboo raft with a bungalow ('rakit') in which the fishers and/or caretakers of the a-FAD reside for weeks or even months. Supply of fresh food and water and other necessities for the persons staying at the rakit of a-FAD is made by fishing boats or carrier boats. The bamboo raft a-FADs with rakit have not yet extended to the western Indonesia, where a-FADs are commonly of the steel (pontoon) or polystyrene types. The pontoon is a steel cylinder of 2 -3 m length and approximately 0.8 m diameter, with generally one end conical. Previously, this was the most common type of a-FAD float in the western and eastern Indonesia. The a-FADs with polystyrene (commonly called 'gabus') are large cylinders or blocks of styrene foam, encased in cloth and often bound by rope and usedmotorcycle tires, and strengthened by a wooden frame. This type of a-FAD has replaced the pontoon as the most common a-FAD type, due to its lower cost.

Total Numbers and Positions of a-FADs
It is difficult to estimate the total number and position of tuna a-FADs in Indonesia's FMAs 572and 573 due to the lack of effective systems of a-FAD registration and monitoring, and the desire of fishing companies and boat skippers to keep a-FAD position confidential. However, the current fisheries laws require the registration of a-FADs, and owners of a-FADs should provide information on the position and the use of their boats for each a-FAD installed in the Directorate General of Capture Fisheries in Jakarta. These laws have not yet been effectively implemented and complied with. In general, National, Provincial, Regency and District offices of Ministry of Marine Affairs and Fisheries (MMAF) were unable to provide information on numbers and locations of a-FADs. Port Authorities are primarily concerned with monitoring of boat traffic into and out of ports and boat activity in their ports, and do not in general maintain records of a-FAD locations. Some fishing companies, boat owners, and skippers interviewed for this study provided positional information for their a-FADs, whereas others were reluctant to do so because they want to keep their fishing locations confidential. Information obtained by this enumeration program, combined with that from other sources, suggested that the total number of a-FADs in FMAs 572 and 573 ranged from several hundreds to several thousands. Figure 3 shows the positional information of a-FADs in FMAs 572 and 573 based on some of the enumeration done for this study. The current Indonesian fisheries regulations for a-FADs require that a-FADs should be apart at a minimum of 10 nm. There is strong evidence to suggest that this requirement is not being complied with, and in many cases a-FADs are deployed significantly less than 10 nm apart. Achieving the effective enforcement of this regulation is undoubtedly one of the biggest challenges faced by Indonesia's management agencies, and requires improved education to fishing companies, fishing boat owners, and fishers about the proven benefits that are likely to come from a reduction in density of FADs in any given area (Cayré 1991;Marsac and Cayré 1998). The main fishing areas for HL and TR boats based at Pelabuhanratu Fishing Port are in the Indonesian EEZ waters, 20 -200 nm from port, but some of the a-FADs are located in the high seas, about 350 nm from Pelabuhanratu. The main fishing areas for HL and TR boats based at Muara Padang and Bungus are on the western side of the Mentawai Islands, approximately 70 -300 nm from Padang ( Figure 3).

Tuna Fisheries Associated with a-FADs
Fishers based at Muara Padang and Pelabuhanratu Fishing Port use multi-gear boats, commonly called 'kapal tonda', to catch tuna in the vicinity of the a-FADs. The kapal tonda (TR boats) normally use HL and TR, which switches gears depending on season, prevailing seas conditions, and catch success. The boats are commonly woodenhulls of size 6 -10 GT. The kapal tonda and their fishing methods originated in southern Sulawesi (Bugis fishermen). Fishers based at Bungus Fishing Port use TR and PS boats to catch tuna around a-FADs. The PS boats have two types: catcher boat

97-105
and carrier boat (called group purse seiner, consists of one catcher boat, two to three carrier boats and several light boats). Both the TR and PS boats are commonly of wooden-hull structure. The size of TR boats ranged 6-10 GT and the size of PS boats ranged 30 -150 GT.

Operational Aspects of the Fisheries Using a-FADs
The success rate of a-FAD visit The a-FADs visit success rate of boats of HL and TR fisheries based at Muara Padang during the 2013 -2014 survey periods was 34.5%, and that of the PS boats based at Bungus Fishing Port was 48.3%. The HL and TR boats based at Pelabuhanratu Fishing Port were significantly higher, at 87.7%, than those at Muara Padang and Bungus (Table 1). The significantly lower a-FAD visit success rate of HL/TR boats based at Muara Padang may be due to more a-FADs in the Padang region with a higher a-FAD density (see Figure  3) or due to competition with PS boats operating on similar fishing grounds or due to both. Widodo et al. (2016) noted that skippers of PL boats based at Kendari (Southeast Sulawesi) and Sorong (West Papua) often expressed their frustration that the deployed FADs were found 'empty of fish' after sets by PS boats. In the FMAs 713-717, the PS, PL, and HL/TR fleets have significant overlapping in their fishing grounds. Information provided by the PL skippers indicated that it normally takes at least 1 -2 weeks before fish re-aggregated at the FADs after a PS set. Limited data and information were obtained for PS boats during the enumeration (5 fishing trips surveyed only). The PS boats based at Bungus Fishing Port operate in a group, which consists of a PS catcher boat, 2 to 3 PS carrier boats, and several light boats. The pattern of fishing operations of the group is that the PS catcher boats have fishing trips for more than 6 months; catches are transferred directly to the PS carrier boats because the catcher boats generally do not have fish-holds of sufficient size to hold the catch; and one setting of a net per day or per night is normal. The amount of fish that is transported to Bungus Fishing Port by a PS carrier boat is mostly the catch from 3 to 4 sets by a catcher boat. The unloaded catch by PS carrier boats in the Bungus Fishing Port showed that five carrier boats surveyed during 2013 (3 boats) and 2014 (2 boats) were between 1,530 kg and 43,500 kg with an average of 15,406 kg per carrier boat per landing (Table 3). If these landings were from catches from 3 -4 sets of the PS net, the estimated average catch per set was 3,191 kg to 5,722 kg.

Catch Composition
The enumeration results in Pelabuhanratu Fishing Port showed that at least nine species were identified from about 180,751 kg of sampled fish that were caught by the HL and TR boats, with adult (big) yellowfin tuna (a-YFT, Thunnus albacares) (49.2%) and skipjack tuna (SKJ, Katsuwonus pelamis) (32.9%) making up a large proportion of the catch. Common dolphin fish (DOL, Coryphaena hippurus) and striped marlin (MLS, Kajikia audax) were key bycatch species ( Figure 4A). During 2013-2015, the enumeration results at Muara Padang showed that at least 10 species from 136,563 kg of sampled fish were caught by the HL/TR boats. The highest catch proportion among the overall catch during the study period was SKJ at 37.7% by volume, followed by the mixed juvenile and adult yellowfin tuna catch (j-a-YFT, Thunnus albacares) at 24.7% and mixed juvenile and adult bigeye tuna catch (j-a-BET, Thunnus obesus) at 13.1%. Mixed juvenile yellowfin and bigeye tuna catch (j-YFT-BET) also comprised a significant portion of the total volume (8.0%). Black marlin (BLM, Makaira indica) and various sharks (Carcharhinus spp.) only made up of 0.1% and 0.01% of the catch by volume, respectively ( Figure 4B). Catch composition of the 77,030 kg landed fish by PS carrier boats at Bungus was SKJ at 29.9 %, j-YFT at 19.5 %, j-BET at 4.9 %, neritic tuna or tongkol (BLT-FRI, Euthynnus affinis, Auxis rochei, and A.thazard) at 27.3%, and scads at 18.8% ( Figure 4C).

97-105
Fish Size The result of the port sampling program at Pelabuhanratu showed that the average size (cm fork length or cm-FL) of SKJ caught by HL and TR in 2013, 2014, and 2015.3 cm, and 42.1 cm, respectively. The results in the three study areas showed that the average sizes of most YFT and BET caught by HL and TR vessels and PS vessels were below the reported Lengths of Maturity (L m ): 103 cm FL for YFT (Mardlijah et al., 2012) and 102 -105 cm FL for bigeye tuna (Schaefer et al., 2005). The majority of SKJ size landed by HL and TL vessels at Muara Padang were also below their reported L m of 40 -42 cm FL (Tandog-Edralin et al. 1990). In contrast, the average size of SKJ caught by PS vessels based at Bungus Fishing Port and by HL and TL vessels based at Pelabuhanratu were larger than the L m . The average size of most SKJ caught by HL, TR and PS boats in Muara Padang and Bungus was that of juvenile fish, and YFT and BET caught by HL, TR and PS boats in the three landing locations were also small (Table 4).

Discussion
Among the two types of FADs, i.e., drifting and anchored FADs in the Pacific region, a-FADs have been used since the late 1970s by both industrial and artisanal sectors. Fonteneau et al. (2013) noted that tuna FADs were first introduced in the Indian Ocean in the early 90s. Among the two, only a-FADs have been used in the Indonesian tuna fisheries where drifting FADs (d-FADs) were not used even by PS boats. Developments in Indonesian a-FADs, including region specific designs, were first detailed by Subani et al. (1989). Monintja (1993) described ten different types of a-FADs in Indonesia. In general, a-FADs in Indonesia use biodegradable material such as nypa or coconut leaf as a subsurface attractor of a-FADs. Dagorn et al. (2012) mentioned that currently most FADs deployed are made of synthetic materials such as nylon ropes or small pelagic fishing nets. Synthetic materials were not recommended because of its contribution to the marine litter and other potential negative impacts on the ecosystem, such as FAD beaching (Maufroy et al., 2015, Zudaire et al., 2018. Anderson et al. (1996) defined at least 10 benefits of FADs and three of them are related to the increase of fishery production, reduced fuel consumption and reduced pressure on reef resources, but the use of FADs including a-FAD results in a lot of juvenile tuna catch at a size below the reported L m . The exact reasons for the association of juvenile tuna and FADs are still not known (Dagorn et al., 2007). However, Castro et al. (2002) stated that FADs provide protection to juvenile tuna while also providing them with a good food supply. The floating objects are also believed primarily for protection from predators, as a source of food availability and increasing survival of eggs, larvae, and juveniles (Gooding & Magnuson, 1967). Juvenile bigeye may also use FADs as a meeting point to develop larger schools (Fréon et al., 2000). Majority of YFT and BET landed by HL/TR and PS in Pelabuhanratu, Muara Padang and Bungus were at juvenile stage, and the majority of SKJ landed by HL/ TL vessels at Muara Padang were also at juvenile stage. Harvesting juvenile fish before they reach their socially optimal size generated growth overfishing (Diekert, 2012).
The significantly lower a-FAD visit success rate of HL/TR boats based at Muara Padang may be due to more a-FADs in the Padang with a higher a-FAD density, competition with PS boats operating on similar fishing grounds, or both. Natsir et al. (2017) said lower value of technical efficiency of fishing gear associated with FADs could be resulted by the increasing number of FADs that can result in a higher a-FAD density in certain areas or waters.

CONCLUSIONS
FADs employed by tuna fisheries in Indonesia are a-FADs with biodegradable attractors such as nypa or coconut leaf. Two principal tuna fisheries associated with a-FADs in Indonesian FMA 572 and 573 were i) HL and TR, and ii) PS. The lack of effective registration and monitoring system of a-FADs and confidentiality issues of fishing companies and boat skippers made it hard to estimate the total number and position of tuna a-FADs in Indonesian FMAs 572 and 573. Higher a-FADs density in certain fishing grounds and fishing competition among PS and HL and TR have resulted in a lower a-FAD visit success rate for HL/TR boats in the fishing grounds. Large portions of small-sized (< L m ) SKJ, YFT and BET were caught by HL/TR and PS boats associated with a-FADs. Therefore, further studies to determine the maximum proportion of small-sized tuna catch need to be done to ensure the sustainability of tuna fisheries. The use of a-FADs in tuna fisheries is considered to be an important part in their fishing operation. The use of a-FADs with biodegradable materials such as nypa and coconut leaf in Indonesia is considered to be environment-friendly a-FADs, and should be maintained. As the Government has issued regulations for the use of a-FADs, an increased awareness of all stakeholders is required for the effectiveness management of a-FADs in Indonesia. Further research on the a-FAD fisheries is required to determine 'realistic' and effective a-FAD management options and to identify the likely impacts of a-FADbased management measures on both industrial scale and small-scale fishers, e.g., restrictions on a-FAD numbers by region, regulated a-FAD sharing within and between gear-types, seasonal closures, etc. There is a need to address the question of whether a free school (i.e. FAD-free) tuna fishing by 'one by one fishing gears' (hand-line or troll-line) and by purse seine is likely to achieve the operational efficiencies, sufficient catch, and sustainable incomes for the communities and industries associated with those gears.