Abstract
The European Research project " Catch, Effort, and eCOsystem impacts of FAD-fishing " (CECOFAD) set out to improve our understanding of the use of drifting fish-aggregating devices (DFADs) in tropical purse seine tuna fisheries in open ocean ecosystems. Data from unofficial technology information related to FAD-fishing were retrieved and the changes over time in systems used for positioning buoys at-sea (radio, satellite transmitters, echo sounder buoys) were quantified. The total number of DFADs deployed at sea in the Atlantic and Indian oceans over the last ten years was estimated from 2 different approaches, based on information provided by the French tuna association and extrapolated to the other purse seiner fleets. From data collected within the FAD National Management Plan, the relationship between the number of active DFADs and the catch per Spanish purse seiner (with or without the assistance of a supply vessel), was explored. Owing to the difficulty of gathering unofficial information, the standardization of CPUE was limited to juvenile bigeye caught using DFADs and to non-standard explanatory variables provided by the French fleet. The analysis was repeated for the combined French and Spanish fleets but with a smaller number of predictors because data for the Spanish fleet was missing. However the relevance of some non-traditional factors was highlighted, in particular the needs to collate the links between each supply vessel and its associated purse seiners and the number of active DFADs per vessel. As an alternative to CPUE, direct indices of tuna abundance through the use of echo sounder buoys attached to DFADs in the Spanish fleet were investigated. Due to the heterogeneity of echo sounder buoy information several criteria for cleaning datasets before performing an " alternate " Buoy-derived Abundance Index (BAI) were identified. Nominal measurements were then standardized using a GLMM approach. Behavioral models representing the continuous process of association and disassociation under FADs, were also identified as an alternative to commercial catch data for estimating abundance. Information collected by national FAD management plans was used to explore the rate of deployment of non-entangling DFADs and the apparent survival rate of released species of incidental catch. Time-area moratoria on DFADs, introduced regularly by tuna RFMOs, are limited to the protection of juvenile tuna and do not account for the potential impact on bycatch or associated megafauna (whales and whale sharks). A simple iterative " fishing-day " model, which included realistic scenarios in terms of difference in fishing strategies between the French and Spanish fleets, was developed to investigate the consequences on tropical tuna and bycatch of introducing wide area, six-month moratoria on DFAD sets. The simulations predicted a decrease in DFAD sets and an increase in free school sets, leading to an overall decrease in tuna catch of ~100 t/yr/boat in the Atlantic Ocean and 600–1800 t/yr/boat in the Indian Ocean. The bycatch for all groups considered (other bony fishes, billfishes, sharks and turtles) decreased, except for turtle and shark in the Atlantic Ocean. Because the fishing practices were modified, whale and whale shark associated sets increased slightly in the Indian Ocean. French GPS buoy trajectories were analyzed to detect the potential damage of lost DFADs on vulnerable coastal ecosystems. This showed that, for the period 2007-2013, around 10% of the trajectories of floating objects with GPS-buoys ended with a “beaching event” in the Atlantic and Indian Oceans, with significant portions of these beaching events occurring in areas with sensitive habitats, such as coral reefs. Bycatch data collected under the EU observer programs in the Indian, Pacific and the Atlantic Oceans, was used to evaluate the effect of FAD-fishing in terms of diversity. Regional differences were identified but the species diversity associated with the DFADs was found to be richer than for free school sets. One of the tasks of CECOFAD was to provide bridges between the various databases used in the project. The French versions of the Balbaya (logbooks) and ObsTuna (observer data) databases were linked using the Standard Data-Exchange Format used within the EU Data Collection Framework (DCF). This format allows a data aggregation level while respecting data confidentiality issues and should be considered as a good candidate for the exchange of data within the tropical tuna research community. Owing to the lack or imprecision of DFAD-fishing activity data, the definitions of variables required for evaluating DFAD-fishing activities that should be continuously recorded in logbooks were reviewed. On the basis on the most recent recommendations from tRFMOs the electronic fishing logbook data model (ERS) was extended to floating objects with the aim to incorporate the information required for measuring the DFAD-fishing effort and to produce indicators of the effect on the ecosystem of deploying floating objects. These extensions now include a new classification of floating objects and a detailed list of operations (including buoy activity) to be filled in by the skipper.