User menu
Blue marlin is one the bycatch species caught by tuna longline and gillnet fleets in the Indic Ocean. Unique stock in the Indic Ocean is assumed to the most probable hypothesis. The status of the blue marlin stock is unknown and the available data is limited to catch and catch rates. Biomass dynamic models are one of the alternatives to assess the stock status in such poor data scenario. In this paper the blue marlin is assessed by using Bayesian state-space models (Fox and Schaefer types) calculated based on estimated total catches and standardized catch rates of Japan. Informative and non-informative priors were used. Likelihood function was based on log-normal density distributions. Monte Carlo Markov Chains are used to calculate the posterior sample. Three chains starting with different parameters estimations were calculated. The first 30000 samples of each chain were discarded (burnin), and the next 50000 samples were sliced resulting in a final sample with size equal to 1000. Convergence of the chains was assessed using Gelman-Rubin diagnostics. Schaeffer type models converged, but all Fox models did not converge. Overall the production models fitted with observational error only are biased, while the state-space models are not. Nevertheless, because there are many parameters, and because the data on blue marlin are not that informative, the uncertain on the estimations were very high and the solutions were sensitive to the choices concerning the priors. State-space model needs to be further tested before using it in situations that the data is not informative as is the blue marlin case.
