Performance of a finite mixture model in CPUE standardization for a longline fishery with target change

Abstract

Standardized catch per unit effort (CPUE) is frequently used to represent annual fish-stock abundance trends. In multispecies fisheries, the catchability coefficient differs depending on the target strategy. Consideration of the target strategy in CPUE standardization is therefore crucial in improving the accuracy and precision of annual CPUE estimates. This study focuses on the potential of a finite mixture model (FMM), where target species and their annual trends are estimated simultaneously in a single model. The eight candidates of CPUE standardization methods—including the FMM—were used to evaluate the methodological performance as to whether they could identify the target strategy. A numerical simulation was undertaken using simplified data that mimic a Japanese longline fishery considering the target change over years, area effect, and multispecies abundance trends. Our simulation study indicated that the FMM outperformed other methods in the various scenarios, and therefore it is concluded that the FMM is an effective and robust methodology for clarifying the target strategy in the modeling of CPUE standardization.