CPUE standardization of Neon flying squid caught by the Chinese squid jigging fishery up to 2024 in the Northwest Pacific Ocean

The neon flying squid (Ommastrephes bartramii) population in the Northwest Pacific Ocean comprises two main seasonal cohorts: the winter-spring cohort and the autumn cohort. We used three approaches including Generalized Additive Models (GAM), Random Forest (RF), and Gradient Boosting Machine (GBM) to standard the catch per unit effort (CPUE) for these two cohorts and for the combined stock during 2005–2024 of Chinese squid jigging fishery data. All models incorporated spatial, temporal, and environmental variables. Model performance metrics and residual diagnostics indicated acceptable fits for all models. The resulting standardized CPUE showed highly consistent interannual trends among the three methods for each cohort, as evidenced by high pairwise correlations (r > 0.94). Specifically, the standardized CPUE for the winter-spring cohort peaked during 2005–2007, declined thereafter, and has since remained at relatively low levels. In contrast, the index for the autumn cohort showed an increasing trend from 2012 onwards, reached its highest point around 2019–2020, and has exhibited a recent declining trend. The resulting standardized CPUE indices, validated through multi-model comparison, are provided to support future stock assessment and management of this species.