Application of non-equilibrium production models to the red shrimp ( Aristeus antennatus, Risso, 1816) fishery in the northwestern Mediterranean

Abstract

Non-equilibrium production models were used to provide guidance/information on the exploitation level of the main deep-water crustacean fishery in the western Mediterranean. Annual catch and effort (engine horsepower) data from 1957 to 2001 were used to compute a long data series of catch per unit effort (CPUE) as kg/HP. Furthermore, a sampling program on-board the trawl fleet in the Balearic Islands was carried out between 1991 and 2001 to estimate the annual CPUE in kg/h. Both CPUE data series were used to fit non-equilibrium production models. Process-error estimators were applied to the Fox model and observation-error estimators to the Schaefer model. Parameter estimates obtained from the short CPUE data series were unreliable and highly uncertain for both models, mainly due to insufficient contrast in the data between 1991 and 2001. The parameter triplet k, r, and q (i.e. the carrying capacity, intrinsic rate of biomass increase and catchability coefficient), as estimated with the CPUE (kg/HP) data series from 1957 to 2001, gave 1814 t, 0.327 per year and 0.054 for the Fox model and 3656 t, 0.226 per year and 0.019 for the Schaefer model. These results were used to calculate biological reference points. Maximum sustainable yield, Y MSY, was estimated at 218 t (Fox) and 210 t (Schaefer). Parameter estimates were less variable for the Schaefer model and these were therefore used for management recommendations. Currently, catches are 30% below the Y MSY, suggesting that red shrimp (RS) is slightly under-exploited. A precautionary fishing effort was estimated taking into account the uncertainty of the parameters. Consequently, a fishing effort corresponding to a biomass level of 60% of the carrying capacity, f 60% k , was adopted. At the current fishing effort, catches in 2002 remain at recent levels. Nevertheless, taking into account the precautionary approach in the fisheries, it is recommended that catches in 2003 should not exceed approximately 100 t, so that the long-term catch would correspond to f 60% k , which is estimated to be around 200 t. Despite the simplifying assumptions of production models, the correct use of these models can provide an insight into species/fishery dynamics and can be used to analyse the effect of changes in the level of fishing level on catches. It should be noted, however, that the perception of the exploitation level of RS presented in this study should be further validated. It is recommended that the sampling program to collect data on fishing effort (fishing hours) should be continued allowing the non-equilibrium production models, both with frequencist and Bayesian approaches, to be applied to a longer CPUE (kg/h) data series.