Performance of Catch-Only Models at Providing Management Guidance for Unassessed Data-Limited Fish Stocks

The majority of the world’s fisheries, particularly in developing countries, are unassessed with little information about fish population status. Several methods based only on fisheries catch data along with basic assumptions about population resilience and fishing effort dynamics have been developed to assess stock status (B/B_msy). Research has shown that an ensemble of these methods provides more accurate status estimates than individual models. However, we do not know if these models can provide reliable and robust management advice, nor how well they perform after implementing management strategies that control total catch or fishing effort. We use management strategy evaluation to determine whether an ensemble of these catch-only models can reliably inform harvest control rules to deliver management objectives. We also investigate the value to management performance of collecting additional data sources: more accurate catch data, fish length composition data, and fishing effort data. To contrast stocks with different life histories, management histories and data constraints, we focus on six species from the west coast of the U.S., Canada, and the Eastern Tropical Pacific Seascape. Our work outlines important trade-offs when using catch-only models to inform management and provides recommendations on which monitoring programmes to improve or develop given limited resources.