Spatial Distribution, Behaviour, and Biological Aspect of Albacore (Thunnus alalunga) Caught in Eastern Indian Ocean

This study highlighted the occurrence of the Indonesian tuna longline fishery targeting albacore (Thunnus alalunga) caught in the Eastern Indian ocean. The data used in this study based on the Research Institute for Tuna Fisheries (RITF) observer program in Benoa. This paper presents the current information on biological aspect (size distribution and length-weight relationship) and fish behavior (swimming layer and feeding periodicity) of albacore. Total albacore samples of 3,152 were taken from scientific observer data from 20102013. The study area of albacore was between 0-33°S and 75-131°E. Albacore length distributed from 70-196 cmFL (median=93 cmFL, mode=100 cmFL, mean=92.12 cmFL) and dominated at size of 95 cmFL. The highest percentage length of albacore was >90 cm (L50) occured in the area between (30-35°S and 80-95°E) and (10-15°S and 120-125°E). The length weight relationship was determined to be W=0.0045 FL1.8211 (W in kg, FL in cm). The expected season to catch the ALB was from April to July with the peak season in June and July. The swimming layer of albacore based on minilogger data were distributed from at 118 to 341 m depth and mostly catch at depth of 156 m with temperature degree 18°C. The feeding periodicity of albacore’s are start from 7:45am to 17:59 pm, mostly active at 10 am to 11 am. The majority of ALB caught by Indonesian longliner was mature condition and negative allometric growth. The ALB peak season was in June-July and the best time to catch was 10 am to 11 am at depth of 156 m.


Introduction
Albacore (ALB) is one of a major important commercial species of Indonesian tuna longline fisheries in Eastern Indian Ocean. The production of albacore (Thunnus alalunga) was the third-largest tuna after yellow fin tuna (T. albacares) and big eye tuna (T. obesus). Groups of tuna production reached up to 1,297,062 tons from 2004-2011. The total catch production consist of yellow fin tuna 69%, big eye tuna 24%, albacore 6% and southern blue fin tuna 1%. Most of which were exported to Japan (DGCF, 2012) Albacore (T. alalunga) is a temperate tuna species, widely distributed in temperate and tropical waters of all oceans. The main fisheries are in temperate waters. In the Atlantic, their geographic limits are from 45-50 O N and 30-40 O S, while in the Indian Ocean, their distribution ranges from 5 O N to 40 O S with adults occurring from 5 O N to 25 O S (ISSF, 2014).
In the Indian Ocean, albacore is caught almost exclusively under drifting longline (98%), with remaining catches recorded under purse seines and other gears (IOTC, 2007;Nishida and Tanaka, 2008). Catch of ALB by Indonesian longline fleets operating in Indian Ocean from 2004-2006 was estimated at 9,081 tons by IOTC, while 53.4% of which was landed at Benoa fishing port (Proctor et al., 2007).
The information about distribution and environment factor is important to determine CPUE and stock assessment, especially for migratory species (Lehodey, 2001). (Rochman et al., 2016) shows that environmental factor such as temperature, salinity, dissolve oxygen, and nutrition were greatly influenced ALB behaviour both in vertical and horizontal movement. Furthermore, (Rochman et al., 2016) indicated that ALB tend to stay at thermocline zone with at depths of 118-291 m with temperature ranged from 12.41-20.47 O C, dissolve oxygen from 3.24-4.68 ml.L -1 and the average salinity at 34.90 psu. The aims of this research is to recognize the current information on spatial distribution, effort, behaviour and biological aspect of albacore (T. alalunga) in Eastern Indian Ocean based by scientific observer program.

Materials and Methods
Data of ALB were obtained from on board observer program on commercial tuna longline fleets based at Benoa-Bali in period of March 2010-October 2013. Based on 25 fishing trips (2.338 fishing days), catch and effort data were collected. Data collected using fork length (cmFL) with level of accuracy 1 cm and setting recorded by global positioning system (GPS). The fishing effort (f) and CPUEs for ALB are used as additional information and calculated using the following formula, modified from De Metrio and Megalofonou (1998).
Where a' is the average number of hooks in longline per day (divided by the 1,000 hooks longline effort unit), d' is the number of fishing days per trip. CPUE=N/f (N is the number of fish caught) CPUE=B/f (B is the biomass of fish caught) Length (L) -weight (W) relationship of ALB was fitted to 269 speciments using fork length (cmFL). The parameter (a,b) of the power equation describing the length-weight relationship : Where a and b are constanta of the equation value.
Distribution of seasonal mean of length and seasonal mean of weight tested by One way ANOVA and followed by Post Hock Test (Scheffe Test) to recognized least significant different among month. In this study, minilogger was used to determine the depth of hook and hook timers to recognize the feeding periodicity. The type of hook timer was HT 600 from NKE Instrumentation. The hook timers were fitted into the one series with branch line and will be linked directly with main line. The type of minilogger was SP2T-1200 from NKE Instrumentation. Minilogger was fitted at the end of branch line and replaced the hook. (Figure 1) shows the position of minilogger and hook timer on tuna longline gear. Radio data pencil is used to transfer data from minilogger to WinMemo program in the computer. The feeding periodicity data which was obtained from hook timer then analyzed using the diagram.

Study area
The study area of ALB can be described base on result of onboard observer program in 2010-2013. The ALB fishing area caught by tuna longline fisheries was between 0-33°S and 75-131°E. The ALB fishing area mostly conducted outside Indonesian Exclusive Economic Zone (EEZ) (Figure 2). The catch data including CPUEs, and size of ALB were georeferenced in 5° grids of latitude and longitude. Surfer 9 program is used to describe spatial distribution of CPUE and size.

Size distribution
Total of 3152 ALB samples were taken from scientific observer data ranged from 2010-2013. ALB length (cmFL) distributed from 70-196cmFL (median=93 cmFL, mode= 95cmFL, mean=92.12 cmFL) and dominated at size 95cmFL ( Figure 3 and   Farley et al. (2012) reported that in South West Pacific Ocean (fish ranged from 43-115cmFL and dominated at size 90-95cmFL). According to Ueyanagi (1969) and Wu and Kuo (1993), the length at the first maturity of Indian Ocean Albacore was 90 cmFL.
Seasonal mean-length distribution was significantly different among month (ANOVA test). According to the Scheffe post hock test, the largest difference occurred in June (mean=97.96 cmFL) and August (mean=79.74 cmFL) (scheffe test, P<0.001).
The highest mean-length distribution occurred in June and the lowest occurred in August. This might be due to seasonal movement caused by sea surface temperature (SST) that influenced reproductive behavior (Chen et al., 2005). reported that mature albacore concentrated in the Indian Ocean south of 10 O S but showed a widely distributed pattern. In north of 10 O S, mature albacore (about 22 kg) occurred all year. In the region between 10 O S and 30 O S, the albacore were mostly mature. The significant difference (P<0.001) in weight composition occurred between the spawning and non-spawning periods. More immature albacore occur in region between 10 O S and 30 O S during the non-spawning season. Many albacore in region south of 30 O S, are immature, with a mean weight of nearly 13 kg throughout the year.

Length and weight relationship
The data were taken from observer program involved 269 samples. The result of t-test showed the significant different with b value1.8211 (< 3) and a value 0.0045 (Figure 6.). It is assumed that the growth pattern of ALB was negative allometric, where the growth in length is faster than the growth in weight.
The growth in length is faster than the growth in weight. Test of b (slope) value was 1.8211. According to Zhu et al. (2008) the value of the parameter b was well within the normal range of 2.5 to 3.5. The value of b<2.5 or >3.5 are often derived from sample with narrow size range. So narrow size range (72-123 cmFL) and limited samples (269) may be contributed to the reason why b value = 1.8211 (<2.5) was low for ALB in Eastern Indian Ocean. The comparison of length-weight of ALB in another ocean is shown in Table 1 .
Length-weight relationship is an important input to the regional stock assessment as it is used to convert catches in weight into catch in number (Farley et al., 2012). It can also provide information on relative condition of fish (Farley et al., 2012).   The current study indicated that ALB caught in this study area are in relatively poor condition (compare to the previous study in Table 1). ALB samples in the current study were heavier on average for length compare to the previous study (slope b=1.8211). In addition, morphometrics may not be the best indicator of fish condition and measuring fat content may be better proxy in future studies (Farley et al., 2012).

Swimming layer
Majority of Indonesian longline fleet based at Benoa port Bali are using middle longline type with 12 hooks and depth between 100-350 m. We used minilogger to recognize the depth of hook. We are assumed that the depth of hook number 1 as equal as number 12 and the depth of hook number 2 as equal as number 11 and so on. The recorded of minilogger data showed that the average of deepest hook type was 341.52 m depth and the average of shallowest hook type was 117.83 m depth (Barata et al., 2011a). The recorded data from minilogger showed that the deepest hook was 340 m and the shallowest hook was 117 m and the majority of ALB were caught at a depth of 156 m with fishing pole position no. 2 and 11 ( Figure 9).
The data showed that ALB were caught at temperature range from 14-22°C and caught more at 18°C and a depth of 156 m. Swimming layer is one of the important factor to obtain maximum catches, especially on tuna longline effort. The research showed that ALB majority were taken at a depth of 156 m with fishing no.2 and 11 (Figure 9.). Nugraha and Triharyuni (2009) mentioned that ALB in Indian Ocean were caught in depth of 150-199.9 m. Chavance (2005), stated that ALB on the West Coast of New Caledonia were caught at a depth of 100-410 m. The depth of hook determined by the length of bouy line, branch line, main line, number of hooks and curvature coefficient among branch. This condition caused the difference of depth measurement (swimming layer) of ALB.
The swimming layer of ALB mostly located in the surface layer to mid layer, but there were some of ALB spreaded into deep layer. The big fish of ALB were in deep layer with a fewer in number, while the smaller ALB were in surface layer with a large number of fish. Barata et al. (2011a) reported that ALB dominated 64% with size >100 cmFL at depth from 85 to 124.74 m. Chen et al. (2005), showed that the distribution of immature ALB were caught at surface layer, while mature ALB were caught at mid layer. Tuna longliner which is operated based in Benoa, and are divided into 3 types. There are surface longline, mid longline and deep longline (Barata et al., 2011a). According to Irianto et al. (2013), the surface longline type consist of 5 hooks among buoy which was operated at depth of 100 to 175 m, the mid longline type consist of 12 hooks among buoy which was operated at depth of 125 to 350 m, and the deep longline type consist of 18 hooks among buoy which was operated at depth of 150 to 450 m. According to various study above, we can concluded that the assimilation of surface longline type and mid longline type was the suitable way to catch ALB in their habitat.

Feeding periodicity
The feeding periodicity depends on the distribution of food and the environmental condition. The water pollution can caused the change of feeding periodicity (Effendie, 2002). The intensity of ALB to search of food was in the morning until evening and 1 time periodically in 24 hours. This indicated that ALB was included in diurnal fish.
Based on the onboard observation using hook timer, the intensity of the feeding habit of ALB conducted between 07.45am-17.59pm with the highest frequency at 10.00-11.00am. The average of feeding periodicity at 11.44am (Figure 10.). The activities of diurnal fish have fast movement, active, and migrate within a large area (Effendie, 2002). Gunarso (1998), stated that ALB search for food during the day, while at night quite active to hunt preys. Musyl et al. (2003) stated that the amount of colour pigment in ALB vision will be influenced to feed habit. The limitation of colour pigment caused that the feeding habit only consentrated in clear water. The ability of sunlight to penetrate the water layer also influenced the ability of fish vision to search of food. In the musky water or in the lower intensity of sunlight, the ability of fish vision to the object in water could be reduced.
Generally, all types of tuna can adjust both the ability of the senses of sight during the day and evening, depending on the setting of cone and rod cells function contained in the fish retina (Masuma et al., 2001). The types of fish which were active during the day, generally have cone that arranged in a rectangular shape. This kind of fish use their visión intensively and active to hunt preys. Tuna fishes were not selective to search for preferred food (Barata et al., 2011b).

Catch number and Catch Per Unit of Effort (CPUE)
The data of catch number and catch per unit of effort (CPUEs) of ALB caught by tuna longline fleets tabulated by year in (Table 2). ALB CPUEs ranged between 0.688-1.953 fish/1000 hooks. CPUEs for ALB were highest in 2012 and 2013 (1.272 and 1.953 fish.1000 hooks -1 ). The overall moving average of CPUE showed that ALB catches generally increase from November to (June and July) and decrease in August to November (Figure 7). The highest catch of ALB was observed in the area between ( Levesque (2010), the CPUE value was caused by several factors such as reproductive behavior and feeding behavior. In the open-ocean environment, the availability of food is often limited to specific areas of oceanic convergence (currents, and sea mounts or ridge), which creates productive fishing conditions at certain times of year.
The overall moving average of CPUEs (kg.effort -1 or number.effort -1 ) showed that ALB catches increased in (June and July) and decreased in August to November (Figure 7). It is in accordance with Lee et al. (1999) which was stated that monthly CPUEs distribution of ALB during October-November are lower than the others month. This is also consistent with research conducted by Research Institute for Tuna Fisheries about Indices of Fishing Season (IFS) of ALB with the case study in Cilacap fishing port, Central Java. The study, mentioned that the tuna fishing season expected to be about 4 months (April-July) with the peak season in May (RITF, 2013).

Conclusion
The majority of ALB in Eastern Indian Ocean were caught in mature condition with the average length 92 cmFL. The highest distribution of mature ALB was found at (30-35 O S and 80-95 O E) and (10-15 O S and 120-125 O E). The length-weight relationship of ALB were negative allometric, where the growth in length is faster than the growth in weight. The majority of ALB were caught at depth of 156 m with average temperature 18 O C. The intensity of the feeding habit ALB conducted between 07.45 am-17.59 pm with the highest frequency at 10.00-11.00 am. This study suggests that the best time to cacth ALB in June-July at the depth hook of 156 m. The catch of ALB recommended during the day at 07.45 am-17.59 pm.