Résumé
The PROCEAN (PRoduction Catch / Effort ANalysis) model is a multi-fleet biomass dynamic model developed in a Bayesian framework to conduct stock assessments based on catch and effort time series data. PROCEAN was applied to the eastern component of the Atlantic skipjack (Katsuwonus pelamis) to assess the current status of the stock and provide reference points to fishery managers. Data included 8 time series of catch and effort data spanning from 1969 to 2006 and mostly standardised based on generalized linear modelling techniques. Informative priors and constraints for some parameters were used to allow the model to converge. Model fits were generally good but indicated increasing trends in the residuals for some fishing fleets. Results suggested that the eastern Atlantic skipjack stock would be underexploited, with a biomass in 2006 equal to 1.8 times the biomass at the maximum sustainable yield (MSY) and a fishing mortality in 2006 equal to 30% of the fishing mortality at MSY (FMSY). A sensitivity analysis performed to account for uncertainty in some input parameters showed that FMSY estimates were robust to parameter changes while MSY estimates were sensitive to the prior distribution for MSY. Based on the different model runs, the MSY was estimated to be comprised between 152,000 and 185,000 t. The use of random walks on catchability was useful for tracking changes in catchability and associated increasing fishing power for some baitboat fishing fleets targeting skipjack.