Comparison of joint CPUE standardization based on various scenarios for Pacific saury (Cololabis saira) in the Northwestern Pacific Ocean
To resolve different patterns in standardized catch per unit effort indices among members and to increase spatial and temporal coverage of catch and effort data, we standardized multiple joint CPUE indices with different assumptions. Three scenarios were established by generalized linear model (GLM) and generalized additive model (GAM) to compare standardized CPUE trends based on different fishing area (CA and EEZ) and different fishing effort (operation days and net hauls). Compared with GLM, GAM was found to be more suitable to conduct joint CPUE standardization for all scenarios, according to AIC, BIC and cross validation. The results showed the Pearson’s correlation of standardized CPUE indices with different types of fishing effort was 0.92, indicating that different fishing effort used in joint CPUE standardization would not significantly affect the standardized CPUE trend. However, there is a discrepancy between the trends of standardized CPUE indices in CA and EEZ. This maybe induced by the different historical patterns in exploitation between the different areas. In general, we recommended using operation days as fishing effort in joint CPUE standardization and separately conduct joint CPUE standardization in CA and EEZ.