Development and testing of single- and multi-species models for fisheries management under climate change

Abstract

Despite the large body of research suggesting that the population dynamics of fishes are affected by climate and species interactions, the majority of tactical fisheries management use single-species population dynamics models that assume individual populations are independent of one another and their environment. Multi-species management strategies that account for shifting productivity due to climate variability and species interactions have been proposed as an alternative to single-species approaches. However, research is needed to evaluate the performance of multi-species management strategies relative to single-species approaches while accounting for the feedback between management and fish populations through continued data collection and assessment. Here we expand on an existing climate-enhanced multi-species statistical catch-at-age modelling framework used in the Bering Sea Alaska, to add important features for application to four groundfish fisheries in the Gulf of Alaska. We further evaluate the performance of single- and multi-species management strategies using management strategy evaluation approaches. The first chapter develops the climate-linked multi-species statistical catch-at-age models for groundfish in the Gulf of Alaska. The second chapter uses the multi-species model as an operating model to assess the performance of single-species harvest control rules under predation scenarios. The third chapter compares the performance of single- and multi-species management strategies when the multi-species operating model is projected under alternative climate projections. Overall, we found single-species management strategies are robust to mis-specification in predator-prey and recruitment dynamics. Multi-species management strategies can outperform single-species strategies in terms of catch, trading off performance in catch variability and conservation performance metrics, when the multi-species estimation model is correctly specified. However, when multi-species estimation models are mis-specified, they may no longer outperform single-species approaches and care should be taken when selecting the multi-species harvest control rule. Our work addresses key questions regarding the performance of Ecosystem-Based Management compared to traditional single species management under climate change and the utility of management strategies that include ecosystem considerations such as multispecies interactions.