Abstract
In the frame of the Ecosystem Approach to Fishery (EAF) management, impact of the tropical tuna purse-seine fishery on targeted and incidentally captured species should be investigated. They may include megafauna species, such as sharks, rays, cetaceans, turtles or sea birds, which often are emblematic and vulnerable species. Tropical tuna purse-seiners actively search, at the surface of the sea, for clues indicating the presence of tuna schools (e.g. birds, floating objects, whales, dolphins or whale sharks). In the eastern Atlantic and western Indian Oceans, the main two fishing modes are sets on free swimming tuna schools and schools associated to natural or artificial floating objects, thereafter called Fish Aggregating Device (FAD). However dedicated studies on fishing sets associated to whale sharks and cetaceans are still lacking. The aim of this thesis is therefore, using logbook and scientific onboard observer data, to investigate the spatio-temporal co-occurrences and/or interactions between whale sharks, cetaceans and the tuna purse-seine fishery within an ecosystem conservation perspective. This work underlines that the megafauna/ fishery co-occurrence occurs in specific spatio-temporal strata: i) Gabon to Angola (April–September), ii) the Mozambique Channel (June–September), and iii) East of Seychelles (April–September). As baleen whales and whale sharks are filter feeders, the co-occurrence with the purse-seine fishery was mostly linked to highly productive environments (i.e. using proxys including chlorophyll-a concentration). In addition fishing sets involving megafauna were relatively high before 2000 (up to 20% of the sets), but are nowadays less frequent (AC3 and 1.5% of the sets associated to baleen whales and whale sharks). The purse-seine fishery appears to have a relatively low impact on megafauna species with mortality rates of 1.4% for whale sharks and 5.6% for cetaceans. Whale shark satellite tagging also confirms these results on the longer term, but the low sample size precludes any final conclusion. While dolphins are present in fishing areas, very few interactions with the fishery was detected, which highlights the striking difference with the eastern Pacific Ocean where half the sets are associated to dolphin herds. In addition, the diversity of targeted and bycatch species captured under whale shark and baleen whale sets was also investigated. Whale sharks are principally associated to skipjack and yellowfin (of various sizes) tunas and baleen whales mostly to large yellowfin tuna. In addition, bycatch species associated to these two megafauna groups was relatively low and dominated by the silky shark, and bycatch diversity was close to the one found for free swimming tuna schools. Finally, real and/or simulated encircled megafauna conservation measures or fishing effort management measures (especially FAD moratoria including larger ones) were investigated. The first ones were found to have limited consequences on tuna catch and bycatch. Conversely FAD moratoria had limited impacts on the number of megafauna associated fishing sets, due to the fact that the main spatio-temporal strata of megafauna and FAD sets differ. However larger and longer moratoria could be beneficial for juvenile tuna and some bycatch species. Overall, this thesis has lead to increase the knowledge on megafauna/ fishery interactions, essential in the general framework of setting up an EAF in the tropical tune purse-seine fishery.