Comparing three indices of catch per unit effort using Bayesian geostatistics

In assessing a fish stock, indices based on catch per unit effort (CPUE) are frequently used. Estimates of three indices of catch per unit effort were compared here (CPUE1, CPUE2 and CPUE3), considering the fitting of two models: (i) a bivariate geostatistical model for catch and effort; (ii) a bivariate model where catch and effort were considered spatially independent. For comparing the estimates of the three indices after the fitting of the two models, catch and effort data were simulated in different scenarios. The simulation study showed that, in general, the estimates of CPUE1 expressed by the ratio of the means of catch and effort, present better results for different scenarios and that the estimates from (i) are better than (ii), mainly when there is a correlation between catch and effort and an additional spatial correlation.