Comparative performance of catch-based and surplus production models on evaluating largehead hairtail (Trichiurus lepturus) fishery in the East China Sea

Abstract

Most of capture fisheries in China are considered as data-poor and lack effective stock assessments. Recently, the catch-based methods have been applied for evaluating main commercial fisheries in China seas. However, how the catch-based assessment models performed and comparison of assessment results with traditional assessment models have not been documented. In this study, we applied two catch-based methods, catch-based maximum sustainable yield (Catch-MSY) and Monte Carlo Catch-MSY-type model (CMSY), on the stock assessment of largehead hairtail fishery in the East China Sea (ECS), and compared the results with other two types of production models, i.e., traditional production models (TPMs), and Bayesian Schaefer model (BSM). Results showed the estimated MSYs are 68.6×104 t with standard deviation (SD) of 3.12×104 t, and 77.6×104 t with SD of 3.98×104 t, using CMSY and Catch-MSY, respectively. The estimated intrinsic rates of increase (r) are 0.27 year −1 with SD of 0.05 year −1, and 0.64 year −1 with SD of 0.11 year −1, using CMSY and Catch-MSY, respectively. Comparing the assessment results with production models, Catch-MSY produced higher r value and lower biomass at MSY level (BMSY) while CMSY produced similar r and BMSY estimates. CMSY and BSM produced similar MSY estimates while estimated MSY of Catch-MSY was similar with those using TPMs. The estimates of biological reference points (BRPs, i.e., B2012/BMSY and F2012/FMSY) using TPMs showed this fishery had not overfished in 2012 while the two catch-based methods and BSM showed the opposite result. The results of leave-one-out cross-validation (LOOCV) confirm that BSM is the best model and the two catch-based methods performed better than TPMs. This study addresses the reliability of catch-based models for data-poor fisheries and is useful for the sustainability of the largehead hairtail fishery in the ECS.