Résumé
We revised the existing framework for tuna CPUE standardisation in light of the increasing literature that advocates the use of mixed effects models to account for the characteristics of logbook data. We apply the framework on yellowfin tuna (YFT) from the Indian Ocean, caught by the purse seine EU fleet (Spain and France) from 1984 to 2015. We used a comprehensive list of candidate covariates, including non- conventional covariates, and run exploratory models to assess the contribution of each covariate. Due to the large number of covariates, the lasso – least absolute shrinkage and selection operator- method was applied for data mining and model selection purposes. The results are two standardised YFT CPUE time series for the period 1984-2015, one for large fish caught in free-school related sets, and one for mainly juveniles caught in floating object related sets. Issues on the usefulness of highly aggregated data (low resolution: annual and fleet wide) is discussed along with the need for more detailed information on the use of dFADs, preferably at the level of a fishing trip.