Evaluating methods that classify fisheries stock status using only fisheries catch data

Abstract

Methods that use only fisheries catch records to determine the status of exploited fish populations have been used to draw important conclusions regarding the world's fisheries. The reliability of two such approaches is evaluated by simulating a range of fisheries development and overfishing scenarios. The success rate and bias of stock status classification by two catch-based methods is compared with those of two stock assessment methods that explicitly model population dynamics and use additional fishing effort data. On average the catch-based methods correctly classified the status of stocks in 31% and 34% of the cases considered. Two simple stock assessments successfully classified stock status in 57% and 59% of the cases. The catch-based methods and the surplus production stock assessment were negatively biased and on average provided overly pessimistic conclusions regarding stock status. Catch-based methods were more negatively biased on average than the stock assessment approaches.