|Bigeye tuna - Atlantic|
|Marine Resource Fact Sheet|
|Bigeye tuna in the Atlantic|
|FAO Names: en - Bigeye tuna, fr - Thon obèse(=Patudo), es - Patudo, ru - Тунец большеглазый|
|Considered a single stock: Yes Spatial Scale: Regional|
Management unit: Yes Reference year: 2014
Biological State and Trend
An assessment for bigeye tuna was conducted in July 2015 following a data preparation meeting in May 2015. The previous stock assessment was conducted for bigeye tuna in 2010 through a process that included a data preparatory meeting in April (SCRS/2010/011) and an assessment meeting in July (SCRS/2010/017). The last year fishery data used was 2014. Information including biology, fisheries, tagging, genetic studies and stock modeling can be found in the ICCAT special editions of the Bigeye Tuna Year Program (Anon. 2005a), the Second World Meeting on Bigeye Tuna (Anon. 2005b) and Chapter 2 of the ICCAT Manual.
Habitat and Biology
Climatic zone: Tropical; Temperate. Horizontal distribution: Oceanic. Vertical distribution: Pelagic.
Bigeye tuna are distributed throughout the Atlantic Ocean between 50ºN and 45ºS, but not in the Mediterranean Sea. This species swims at deeper depths than other tropical tuna species and exhibits extensive vertical movements. Similar to the results obtained in other oceans, pop-up tagging and sonic tracking studies conducted on adult fish in the Atlantic have revealed that they exhibit clear diurnal patterns: they are found much deeper during the daytime than at night. Spawning takes place in tropical waters when the environment is favorable. From nursery areas in tropical waters, juvenile fish tend to diffuse into temperate waters as they grow larger. Catch information from surface gears indicate that the Gulf of Guinea is a major nursery ground for this species. Dietary habits of bigeye tuna are varied and prey organisms like fish, mollusks, and crustaceans are found in their stomach contents. Bigeye tuna exhibit relatively fast growth: about 105 cm fork length at age three, 140 cm at age five and 163 cm at age seven. Bigeye tuna over 200 cm are relatively rare. Bigeye tuna become mature at about 3.5 years old. Young fish form schools mostly mixed with other tunas such as yellowfin tuna and skipjack. These schools are often associated with drifting objects, whale sharks and sea mounts. This association appears to weaken as bigeye tuna grow larger. Estimated natural mortality rates for juvenile fish, obtained from tagging data, were of a similar range as those applied in other oceans.
Jurisdictional distribution: Highly migratory
The geographical distribution of bigeye tuna is very wide and covers almost the entire Atlantic Ocean between 50ºN and 45ºS, but not in the Mediterranean Sea.
Considered a single stock: Yes
Various pieces of evidence, such as a lack of identified genetic heterogeneity, the time-area distribution of fish and movements of tagged fish, suggest an Atlantic-wide single stock for this species, which is currently accepted by the Committee. However, the possibility of other scenarios, such as north and south stocks, should not be disregarded.
The stock has been exploited by three major gears (longline, baitboat and purse seine fisheries) and by many countries throughout its range of distribution and ICCAT has detailed data on the fishery for this stock since the 1950s. Scientific sampling at landing ports for purse seine vessels of the EU and associated fleets have been conducted since 1980 to estimate bigeye tuna catches (Table 1
, Figure 1). The size of fish caught varies among fisheries: medium to large for the longline fishery, small to large for the directed baitboat fishery, and small for other baitboat and for purse seine fisheries.
The major baitboat fisheries are located in Ghana, Senegal, the Canary Islands, Madeira and the Azores. The tropical purse seine fleets operate in the Gulf of Guinea and off Senegal in the East Atlantic and off Venezuela in the West Atlantic. In the eastern Atlantic, these fleets are comprised of vessels flying flags of Ghana, EU-France, EU-Spain and others which are mostly managed by EC companies. In the western Atlantic the Venezuelan fleet dominates the purse-seine catch of bigeye tuna. While bigeye tuna is now a primary target species for most of the longline and some baitboat fisheries, this species has always been of secondary importance for the other surface fisheries. In the surface fishery, unlike yellowfin tuna, bigeye tuna are mostly caught while fishing on floating objects such as logs or man-made fish aggregating devices (FADs). During 2009, landings in weight of bigeye tuna caught by the longline fleets of Japan and Chinese Taipei, and the purse seine and baitboat fleets of the EU and Ghana represented 75 % of the total bigeye tuna catch.
The total annual Task I catch (Table 1, Figure 2) increased up to the mid-1970s reaching 60,000 t and fluctuated over the next 15 years. In 1991, catch surpassed 95,000 t and continued to increase, reaching a historic high of about 133,000 t in 1994. Reported and estimated catch has been declining since then and fell below 100,000 t in 2001. This gradual decline in catch has continued, although with some fluctuations from year to year, until the most recent year of data 2009. The preliminary estimate for 2009 is 86,011 t, the highest value in the last five years. This estimate includes preliminary estimates made for a few fleets that have not yet provided data to ICCAT.
After the historic high catch in 1994, all major fisheries exhibited a decline of catch while the relative share by each fishery in total catch remained relatively constant. These reductions in catch are related to declines in fishing fleet size (longline) as well as decline in CPUE (longline and baitboat). The number of active purse seiners declined by more than half from 1994 until 2006, but then increased since 2007 as some vessels returned from the Indian Ocean to the Atlantic. The number of purse seiners operating in 2009 and 2010 was similar to the number operating in 2003-04 (Figure 3).
IUU longline catches were estimated from Japanese import statistics but the estimates are considered uncertain. These estimates indicate a peak in unreported catches of 25,000 t in 1998 and a quick reduction thereafter The Committee expressed concern that historical catches from illegal, unreported and unregulated (IUU) longliners that fly flags of convenience from the Atlantic might have been poorly estimated. The magnitude of this problem has not yet been quantified, because available statistical data collection mechanisms are insufficient to provide alternative means to calculate unreported catch.
Significant catches of small bigeye tuna continue to be channeled to local West African markets and sold as “faux poissons” in ways that make their monitoring and official reporting challenging. Monitoring of such catches has progressed in some countries but there is still a need for a coordinated approach that will allow ICCAT to properly account for these catches and thus increase the quality of the basic catch data available for assessments.
Mean average weight of bigeye tuna decreased prior to 1998 but has been relative stable, at around 10 kg during the last decade (Figure 4). This weight, however, is quite different according to the fishing gear, around 62 kg for longliners, 7 kg for bait boats, and 4kg for purse seiners. In the last ten years all longline fleets have shown increases in mean weight of bigeye tuna caught, with the average longline-caught fish increasing from 40 kg to 60 kg between 1999 and 2009. During the same period purse seine-caught bigeye tuna had weights between 3 kg and 4 kg, with the exception of 2009 when the average weight was 4.5 kg. Bigeye tuna caught since 2004 in free schools are significantly larger than in previous years. Since FAD catches began being identified separately in 1991 by EU and associated purse seine fleets, the majority (75%-80%) of bigeye tuna are caught in sets associated with FADs. Similarly baitboat-caught bigeye tuna weighted between 6 and 10 kg over the same period, showing greater interannual variability in fish weight than longline or purse seine caught fish.
|Figure1 Geographical distribution of the bigeye tuna catch by major gears and decade. The maps are scaled to the maximum catch observed during 1960-2013. |
|Figure2 Bigeye estimated and reported catches for all the Atlantic stock (t). The value for 2014 represents preliminary estimates because some countries have yet to provide data for this year or are under revision. |
|Figure3 Changes over time in the carrying capacity (corrected by time at sea) of purse seiners and baitboats operating in the eastern Atlantic (1971-2009) and in number of boats for the European purse seiners (value estimated for 2010). |
|Figure4 Trend of mean weight for bigeye by major fisheries (1975-2014) based on the catch-at-size data (upper figure) and for European purse seiners separated between free schools (fsc) and FAD associated schools (1991- 2014) (lower figure). |
The 2014 stock assessment was conducted using 3 separate suites of models, namely a Production Model (SCRS/2015/073), a Statistical catch-at-age model Stock Synthesis (SCRS/2015/126) and VPA. In general, data availability has continued to improve, notably with the addition of relative abundance indices for an increasing number of fleets. There are still missing data on detailed fishing and fish size from certain fleets. In addition, there are a number of data gaps on the activities of IUU fleets (e.g., size, location and total catch). All these problems forced the committee to assume catch-at-size for an important part of the overall catch.
During the 2015 Bigeye Tuna Data Preparatory Meeting, a number of alternative relative abundance indices were presented. At that meeting, the Group reviewed those estimates for suitability as indices of relative abundance to use in different stock assessment models. In some cases, the Group recommended that some modifications or additional analyses be conducted prior to Bigeye Tuna Stock Assessment meeting. The Group requested the development of indices of abundance utilizing the purse seine catch and effort data for potential use in sensitivity runs. (Figure 5).
In order to evaluate the robustness of the procedure used to give advice in 2010, a new composite index was generated using the same methodology and an ASPIC run was conducted with a similar set up as that used in 2010 (which is referred as a continuity case) using the latest catch data up to 2014. To compare both assessments, the 2010 assessment was projected (i.e. hindcast) using the catch data from 2010 to 2014. This allows comparing changes in the perception of the stock solely resulting from the addition or update of the data sets used to fit the production model used to provide the main advice about stock status in 2010. This new run only differs from the one in 2010 in that the catch estimates contain additional years of data (2010-2015), and that the combined index of abundance has been estimated with indices that were presented/agreed during the 2015 preparatory meeting. There were big differences between the 2015 continuity run and the 2010 assessment and projection, which were due to the large difference in the 2010 and 2015 composite indexes. In addition, it was difficult to recreate the CPUE combined series when the CPC’s CPUEs were updated in a different manner from last assessment. Using combined indices, when individual indices show conflicting trends, will result in average/intermediate biomass/harvest estimates that differ from those estimated when fitting to individual indices. Therefore, indices should be evaluated separately or jointly within the stock assessment using appropriate diagnostics.
In 2015, to maintain continuity with the approach used to develop the previous advice for Atlantic bigeye tuna, results from non-equilibrium production models were used to provide the status of the resource; these included runs 1, 2, and 3, which used different individual CPUE indices. Those results were complemented with the results of an integrated statistical stock assessment model (SS3), which can account for changes in selectivity. Although VPA models also account for changes in selectivity, given that VPA results were uncertain in regards to absolute size of the stock and showed convergence problems, the VPA model results were not used to develop the management advice.
The stock biomass estimated from the three production model runs show a decline since the beginning of the time series in the 1950s (Figure 6). Corresponding with a sharp increase of fishing mortality and catch in the 1990s and a peak of fishing mortality by the end of the 1990s, biomass showed a sharp decrease during the same time period. From the late 1990s, the biomass and fishing mortality trajectories of the 3 runs are different. While biomass increased and fishing mortality decreased in run 3; biomass continued to decrease at a lower rate in run 1 and 2 and fishing mortality showed a general increasing trend in Run 2 (except the last 3 years when decreased) and was somewhat stable in run 1. The three runs show similar trajectories of increasing F and decreasing B towards the red area of the Kobe plot (F>FMSY and B<BMSY) until the end of the 1990s, but Run 1 and Run 2 estimate that on average the stock still remains in the red area since 2000; while Run 3 estimates a recovery towards the green area since mid-2000s (Figure 7). The current MSY estimated using the three production model runs ranges from 66,030 t to 86,830 t. The combined phase plots of three cases are shown in Figure 8. MSY is estimated to be from 66,030 t to 86,830 t which is lower (Run 1) and larger (Run 2 and 3) than the 2014 catch (68,390 t).
The integrated model, SS3, was run with twelve different configurations to characterize uncertainty in model parameters. SS3 Model results indicate that fishing mortality increased steadily since the beginning of the fishery, rapidly increased by the end of the 1990s, fluctuating around the level corresponding to FMSY in the 2000s, then increased sharply at the end of the 2000s where F>FMSY in 2011, and decreased in the latest three years despite being kept at levels higher than FMSY in 7 out of 12 scenarios. With regards to biomass, it decreased constantly since the beginning of the time series and fell below and remained below BMSY levels since 2010. The current MSY estimated using the 12 SS runs ranges from 80,889 t to 102,268 t.
Most of the SS runs give a similar view compared to the ASPIC runs regarding the historical evolution of the relative trends in biomass and fishing mortality. Both assessment models (ASPIC and SS3) suggest that biomass decreased in the period investigated, with the exception of run 3 of ASPIC where a recovery is observed since 2005. For fishing mortality, both assessment models show that F increased sharply by late 90s, then fluctuated to reach a similar level of the late 1990s in 2004/2005 and increased again in 2011 to decrease the last three years. The range of MSY values estimated by SS3, however, is larger than those estimated by ASPIC.
|Figure5 CPUE Series agreed at the data preparatory meeting as potential proxies for stock abundance; points are the standardised values, lines the prediction from a GAM fitted to all the indices with year as a smooth term and index as a factor (red) and by index individually (blue). |
|Figure6 Time series of stock biomass, harvest rate and catch by assessment scenario |
|Figure7 Time series of stock biomass and harvest rate relative to MSY benchmarks; lines are medians and ribbons inter-quartiles |
|Figure8 Current status (2014) of bigeye tuna based on ASPIC. Graph combines results for the 3 runs considered. The clouds of points depict the bootstrap estimates of uncertainty for the most recent year (purple = Japan LL run, brown = US LL run, blue= Chinese-Taipei LL run). The median point estimate for each models results are shown in open (cyan) circles. The marginal density plots shown above and to the right of the main graph reflect the frequency distribution of the bootstrap estimates of each model with respect to relative biomass (top) and relative fishing mortality (right). The red lines represent the benchmark levels (ratios equal to 1.0). |
Overall Assessment Results
It is noteworthy that the modeled probabilities of the stock achieving levels consistent with the Convention objective at the end of the projection time period in 2028 are 29% for a future constant catch at the current TAC level of 85,000 t, and 41% probability at current levels of 70,000 t. Higher probabilities of rebuilding require longer timeframes and/or larger reduction of current catches. For instance, 49% probability of rebuilding would be achieved by 2028 with a constant catch of 65,000 t and 58% of probability with catches of 60,000 t, (Table 2).
It needs to be noted that projections made by the Committee assume that future constant catches represent the total removals from the stock, and not just the reported catches and the current selectivity pattern is maintained. ICCAT established a TAC of 85,000 t for 2010 onwards through Rec. 09-01, and Rec. 11-01. Note that because this TAC does not affect all countries that can land bigeye tuna, in theory the total catch removed from the stock could exceed 85,000 t which will worsen the prospect of rebuilding at current TAC levels. Furthermore, any future changes in selectivity due to changes in the ratios of relative mortality exerted by the different fleets – such as an increase in the relative mortality of small fish – will change and add to the uncertainty of these projections.
|Table 2 Estimated probabilities of the Atlantic bigeye tuna stock being below FMSY (overfishing not occurring), above BMSY (not overfished) and above BMSY and below FMSY (green zone) in a given year for catch level ('000 t), based upon the 2015 assessment outcomes. |
|Figure10 Combined Kobe phase plot of non equilibrium production model and integrated stock assessment model. The combined plot was developed by giving equal weighting between production models and integrated assessment model results. Within each model type equal weighting was given to different runs. |
Management unit: Yes
Effects of current regulations
During the period 2005-2008 an overall TAC was set at 90,000 t. The TAC was later lowered (Rec. 09-01 and later modified by Rec. 14-01) to 85,000 t. Estimates of reported catch for 2002-2014 have been always lower than 85,000 t with the exception 2011 where it was close to the TAC. Note, however, that catches for 2012-2014 are still under revision. The current TAC did not result in the stock achieving levels consistent with the Convention objectives.
Concern over the catch of small bigeye tuna partially led to the establishment of spatial closures to surface fishing gear in the Gulf of Guinea (Recs. 04-01, 08-01, 11-01 and 14-01). The Committee examined trends in average bigeye tuna catches by areas as a broad indicator of the effects of such closures as well as changes in juvenile bigeye and yellowfin catches due to the moratorium. The efficacy of the area-time closure (moratorium) agreed in Rec. 14-01 was evaluated by examining fine-scale (1ox1o) skipjack, yellowfin, and bigeye catch by month distributions from the European and associated purse seine fleet FAD fishery and the Ghanaian purse seine and baitboat fishery. After reviewing this information, the Committee concluded that the moratorium has not been effective at reducing the mortality of juvenile bigeye tuna, and any reduction in yellowfin tuna mortality was minimal, largely due to the redistribution of effort into areas adjacent to the moratorium area (for more details see response to Commission 19.1).
The Atlantic bigeye tuna stock was estimated to be overfished and overfishing was occurring in 2014. Projections indicate that catches at the current TAC level of 85,000 t will have around 30% of probability to recover the population to a level that is consistent with the Convention objectives by 2028. Therefore, the Committee recommends the Commission to reduce the TAC to a level that would allow the recovery of the stock with high probability and in as short period as possible in accordance with the principles of Recommendation 11-13.
The Commission should be aware that increased harvests on FADs could have had negative consequences for the productivity of bigeye tuna fisheries (e.g. reduced yield at MSY and increased SSB required to produce MSY) and, therefore, should the Commission wish to increase longterm sustainable yield, the Committee continues to recommend that effective measures be found to reduce FAD-related and other fishing mortality of small bigeye tunas.
|ATLANTIC BIGEYE TUNA SUMMARY |
|Maximum Sustainable Yield ||78,824 t (67,725-85,009 t)1 |
|Current (2014) Yield ||72,585 t2 |
|Relative Biomass (B2014/BMSY) ||0.67 (0.48-1.20)1 |
|Relative Fishing Mortality |
|1.28 (0.62-1.85)1 |
|Stock Status || |
|Conservation & management measures in effect: || |
− Total allowable catch for 2012-2015 is set at 85,000 t for Contracting Parties and Cooperating non-Contracting Parties, Entities or Fishing Entities.
− Be restricted to the number of their vessels notified to ICCAT in 2005 as fishing for bigeye tuna.
− Specific limits of number of longline boats; China (45), Chinese Taipei (75), Philippines (11), Korea (14), EU (269) and Japan (245).
− Specific limits of number of purse seine boats; Panama (3), EU (34) and Ghana (13).
− No fishing with natural or artificial floating objects during January and February in the area encompassed by the African coast, 10º S, 5ºE and 5ºW.
Combined results of non-equilibrium production model and statistical integrated assessment models. Median and 10 and 90 % percentile in brackets.2
Reports for 2014 reflect most recent data but should be considered provisional.
Source of information
Report of the Standing Committee on Research and Statistics (SCRS). “Bigeye, Executive Summary.” Madrid, Spain 28 September - 2 October 2015. http://www.iccat.int/Documents/SCRS/ExecSum/BET_ENG.pdf
Report of the 2015 ICCAT Bigeye Tuna Stock Assessment Session. Madrid, Spain 13 – 17 July 2015.