Development and testing of a Bayesian population model for the bigeye thresher shark, Alopias superciliosus, in an area subset of the western North Pacific

Abstract

A Bayesian population modelling tool integrating separable virtual population analysis, per-recruit models and age-structured demographic analysis was developed for the bigeye thresher Alopias superciliosus (Lowe) population in an area subset of the western North Pacific. The mortality rates for years 1989-2016 were estimated, various biological reference points and associated risks of decline were also estimated, and alternative harvest strategies for the stock were evaluated. Estimates of the posterior mean of fishing mortality for bigeye thresher shark suggest fishing pressure has been high in recent years (2011-2016). The estimated population growth rate (lambda) (without fishing) obtained from age-structured demographic model was relatively low (lambda = 1.01 per year; 95% confidence intervals of 1.00 and 1.03 per year). Risk analyses revealed that only low levels of fishing pressure (10% of the current fishing pressure) over a wide range of ages could maintain a relatively low risk of population decline for bigeye threshers. Sensitivity testing indicated that the model is robust to prior specification. The developed framework could be used as an assessment tool to evaluate the risk of decline for other widely distributed pelagic shark species where insufficient catch and effort data are available.