Assessment of the impact of the PNG purse seine fishery on tuna stocks, with special focus on the impact of fish aggregation devices (FADs)

The PNG Tagging Project is a joint research project being implemented by the Oceanic Fisheries Programme (OFP) of the Secretariat of the Pacific Community (SPC) and the PNG National Fisheries Authority (NFA). Its major objectives are:
1. To obtain information on the large-scale movement of tuna in, and from, the PNG EEZ. This information is important for understanding the relationship of PNG stocks with those of adjacent areas. Movement rates are particularly important for assessing the potential for interaction between fisheries operating in different areas. The comparison of tagged fish movements from the Bismarck Sea (an area of major anchored FAD deployment) that will result from this project with tagged fish movements from the same area in the early 1990s (before extensive anchored FAD deployment) will provide important new information on the meso- to large-scale effects on tuna movement of large anchored FAD arrays.
2. To obtain information on current exploitation rates of tuna in the PNG EEZ.
Information on local exploitation rates is important for understanding the impact of fishing at the EEZ scale. In particular, it allows estimation of the extent to which current catch levels may reduce the standing stock of tuna and the catch-per-unit-effort of the fisheries, a phenomenon commonly known as "local depletion".
3. To obtain information on the dynamics of tuna associations with FADs, in particular species-specific information on residence times, vertical and horizontal movements and FAD interactions. This information is required for a better understanding of the effects of FADs on tuna stocks and their vulnerability to fishing, and for the design of appropriate management measures.
4. To obtain data that will contribute to regional tuna stock assessments. Conventional tagging data are an important component of tuna stock assessments, providing quasifishery-independent information on exploitation rates, natural mortality, movements and other parameters.
5. To obtain information on the trophic status of free-swimming schools of tuna, and tunas associated with FADs, other floating objects and seamounts. This information is required for the general understanding of the ecosystem impacts of FADs compared to other types of tuna aggregations.
6. To characterize the variability and extent of catches of by-catch species from purse seine catches in PNG. NFA runs an observer programme with high coverage rates, which offers the opportunity to document by-catch levels and their variability in purse seine sets on anchored FADs and other set types.
These objectives are being pursued through a tagging programme, and associated data collection activities in PNG waters. Funding support for the project has been generously provided by the PNG National Fisheries Authority, New Zealand Agency for International Development, Australian Centre for International Agricultural Research, European Commission 8th European Development Fund (through the PROCFish Project) and the Global Environment Facility (through the Pacific Oceanic Fisheries Management Project).

This progress report presents the results of the second of two three-month cruises by the chartered pole-and-line tagging vessel Soltai 6, owned and operated by Soltai Fishing and Processing Ltd, a Solomon Islands-based company. The report of the first three-month cruise, undertaken during August-November 2006, is available at
http://spc.int/oceanfish/Html/TAG/index.htm.

The operational objectives of this second and final cruise were:
o To tag and release 15,000 tuna (i.e. half the project target of 30,000 tuna) using
conventional tuna tags, with an ideal species composition of skipjack 60%; yellowfin
30%, and bigeye 10%.
o To increase the spatial distribution of tag releases already achieved during Cruise 1
throughout PNG waters:
o To tag and release 200 plus tuna using electronic archival tags, with a priority on bigeye and yellowfin tuna;
o To undertake sonic tagging and deployment of FAD monitors using methodology
developed during Cruise 1 in 2006;
o To train scientific staff, including two full-time PNG biological technicians, on tagging and sampling methods, including archival/sonic tagging procedures and data
management;
o To undertake biological sampling (length, sex, stomach contents and tissue samples) according to an experimental design in order to obtain information on the trophic status of tunas in different school associations.
Additional activities related to tag recovery were undertaken separately to the activities of the tagging vessel and are reported in section 7 of this report.

Summary of results
The Cruise (and the second charter) began with the departure of the Soltai 6 from Noro, Solomon Islands, on February 19th 2007. The vessel had been recommisioned during the previous week, after commercially fishing for the intervening three month period since the conclusion of Cruise 1, with tagging gear and the portable office relocated on the vessel, and routine maintenance undertaken. Tagging operations proper began in PNG on February 20th, with a productive fortnight in the Solomon Sea before spending six weeks fishing most parts of the Bismarck Sea and adjacent areas, then returning to Noro on May 20th via the Solomon Sea and waters east of Bougainville.

During Cruise 2, a total of 39,064 tuna were tagged with conventional yellow tags of two sizes, 212 with archival tags and 160 with acoustic tags. Archival and acoustic-tagged tunas were also conventionally tagged. Further details of these releases are given in the following sections. 68% of releases were anchored FAD-associated fish during Cruise 2, compared with 80% during Cruise 1.

As at 30th June 2007, 603 tag recoveries had been received from the Cruise 2 releases for an overall recovery rate of 1.5%, consistent with the much higher proportion of releases away from the intensively fished Bismarck Sea. Over 20,000 releases (>50%) were made in the Solomon Sea and in waters east of Bougainville. Cruise 1 recoveries stood at 4,071 (18.2%), with 4,675 recoveries overall (7.6%) for the combined releases (61,7649) on the two cruises.

Tag recovery arrangements are working and survival of fish following tagging is assumed to be good.

Tuna catches in PNG have averaged about 300,000 tonnes per year over the past five years, producing on average 20% of the total western and central Pacific purse-seine catch. Catch has been steadily increasing with time and in 2006 the catch exceeded 400,000 tonnes. The purse seine fishery in PNG is strongly dependent on floating objects, in particular anchored fish aggregation devices (FADs). Therefore, a significant proportion of the catch consists of juvenile yellowfin and bigeye tuna, which aggregate in large quantities around FADs. Recent regional stock assessments have raised concerns about the long-term viability of these stocks. Priorities for fisheries management in PNG are to better understand: the impacts of these large catches on local tuna stocks; and the impacts of FADs on the fishery, on the target species and on non-target species.

The overall goal of this project is to improve management of the tuna fishery in PNG by the collection of new information on tuna population dynamics and fishery impacts (including FADs). The specific objectives of the project are
To obtain information on the dynamics of bigeye, yellowfin and skipjack stocks in PNG waters, including information on local exploitation rates.
To obtain information on the dynamics of tuna associations with FADs, in particular species-specific information on residence times, vertical and horizontal movements and FAD interactions.
To characterize the variability and extent of catches of non-target species from purse seine catches in PNG.
Development of management advice and recommendations for the PNG tuna fishery.

The primary method applied by the project is the tag and release of the target tuna species (yellowfin, skipjack and bigeye) in PNG waters. Four periods of vessel charter, involving a commercial pole-and-line vessel were commissioned to achieve this objective. The Bismarck Sea, where a majority of FADs are deployed and where much of the PNG domestic catch originates from, was designated as the primary operational area.

During the charter periods (August-November 2006; February-May 2007, November 2007, February-April 2008), 102400 tunas were tagged and released over a wide area of the PNG EEZ and neighbouring sections of the Solomon Islands EEZ. The numbers of bigeye tagged were smaller than expected, due to the relatively low vulnerability and/or abundance of the species during the tagging period. 318 of the releases were archival tags and 222 sonic tags deployed in association with receivers attached to FAD arrays. 72 % of the releases were on anchored FADs, 17% on unassociated (free) schools and the remainder on a variety of other school types.

The analysis and modelling of tag returns with reliable information on date, place and method of capture will inform two of the main objectives of the project, with a recapture rate of close to 20%, spread over multiple years, anticipated. To ensure recaptured fish are returned with necessary information, publicity campaigns covering the main unloading, processing and transhipment points through the entire western and central Pacific Ocean (and in some cases beyond) are necessary, reinforced by rewards for the return of tags. Posters publicizing the project and the reward payments have been prepared in 13 languages and widely distributed. Contact persons have also been appointed in key locations to coordinate the recovery of tags.

As at the 30th of April 2008, 10431 tags, mostly with good information, had been returned to SPC, representing an overall recovery rate of 10.1%, in line with expectations of a final return rate of 20%. Recovery rates for yellowfin and bigeye tuna were slightly higher than those for skipjack, as a result of the often longer residence times around FADs resulting in increased vulnerability. 90% of the returns were of fish FAD-associated at release, with the majority of returns for all three species after 40 days at liberty. The average distance moved with time at liberty is increasing with time, as the tagged fish dispersed, after most initial recoveries were close to the point of release. As with previous experiments, skipjack appears to be more mobile then the other two species, with some recoveries already several thousand nm from the point of release.

PNG officers have been trained in all aspects of tuna tagging methods, biological sampling and preliminary data analysis, and have also participated with great success in the archival and sonic tagging. Analysis of PNG observer data, the only reliable source of information data on non-target species in the purse seine fishery, has commenced, well ahead of schedule, to address the third project objective, i.e. characterization of catches of non-target species. Development of management advice and recommendations will occur after the majority of the tagging data are available, and is programmed for mid-2009.