One of the most important biases to consider in tagging capture-recapture data for stock assessment studies is the proportion of reported tags among the actual recaptures, i.e., the tag-reporting rate. In this study, we used the model developed by Kimura (1976) and adapted in a Bayesian framework by Carruthers et al. (2015) to estimate the reporting rates for thirteen Atlantic Ocean tuna fleets using coincidental tagging data and catch data disaggregated by species, school-type (Fish Aggregating Device and Free Swimming Schools), location and time. The method was applied on recaptures and tag seeding experiments conducted during the Atlantic Ocean Tropical Tuna Tagging Program (AOTTP) of the ICCAT (International Commission for the Conservation of Atlantic Tunas). Tag seeding consists of secretly planting tags on fish by observers onboard fishing vessels to estimate how many tags are found during later stages (landing, processing, etc.) and reported to scientific authorities. Our results showed that the tag-reporting rate was as large as 84.70% (80.58-88.39%) for the European Union purse seiner fleet (Spain and France) but decreased for several surface fleets from 72.79% (67.49-77.77%) for the Spanish baitboats (operating off Senegal or in Canary islands) to 22.83% (15.26-31.24%) for the Ghanaian mixed purse seiner-baitboat fishery. Overall, we conclude that given the very low reporting rate for several important fleets operating in the Atlantic Ocean, it is crucial to account for the reporting rate estimates to avoid highly biased results in future stock assessment using tagging data.
Plan de classement
Sciences fondamentales / Techniques d'analyse et de recherche [020]
;
Ressources halieutiques [040]
