An empirical Bayesian approach for estimating fleet- and vessel-level bycatch rates in fisheries with effort heterogeneity and limited data: a prospective tool for measuring bycatch mitigation performance

Abstract

Abstract Minimizing fishing-induced mortality on bycatch and endangered, threatened or protected species is a necessity for fisheries managers. Estimating individual vessel bycatch rates by dividing the amount of bycatch by effort (nominal rate) can be biased, as it does not consider effort heterogeneity within the fleet and ignores prior knowledge of fleet bycatch rates. We develop an empirical Bayesian approach for estimating individual vessel and fleet bycatch rates that: (i) considers effort heterogeneity among vessels and; (ii) pools data from similar vessels for more accurate estimation. The proposed standardized bycatch rate of a vessel is, therefore, the weighted average of the pool rate and nominal rate of the vessel; where the weights are functions of the vessel’s fishing effort and a constant estimated from the model. We apply this inference method to the estimation of seabird bycatch rates in the component of the Australian Eastern Tuna and Billfish Fishery targeting yellowfin tuna. We illustrate the capability of the method for providing fishery managers with insights on fleet-wide bycatch mitigation performance and the identification of outperforming and underperforming vessels. This method can also be used by fishery managers to develop fleet-wide performance measures or quantitative evaluation standards.