Ecosystem-based fisheries management forestalls climate-driven collapse

Marine ecosystems face an uncertain future due to climate change, with predictions of large-scale ecosystem reorganization already being observed. Marine capture fisheries are particularly vulnerable because marine organisms are sensitive to even small shifts in ocean conditions. Currently, climate-adaptive measures are largely absent from fisheries management. Ecosystem-based fishery management (EBFM) is a more holistic approach that incorporates ecosystem information to manage multiple species, offering potential climate resilience. This study investigates the performance of EBFM policies in the Eastern Bering Sea, a highly productive system supporting the largest US fishery (walleye pollock) and a significant Pacific cod fishery. These fisheries operate under well-established EBFM policies, notably a 2-million-ton annual combined groundfish catch limit aimed at preserving ecosystem function. The study's goal is to evaluate whether EBFM reduces future risks of fishery declines, alters thermal tipping points, and improves harvest stability under climate change, especially considering the recent unprecedented warming and sea ice loss in the region.

Existing literature highlights the widespread evidence of climate change impacts on marine systems, yet the implementation of adaptive strategies remains limited. Several studies predict substantial ecosystem reorganization due to climate change and its effect on marine fisheries. Assessments point toward an ecosystem management (EM) approach, ranging from single-species management with ecosystem context to full ecosystem-based management (EBM). EBFM, a middle ground, expands adaptive management by using ecosystem information to manage multiple species. While intuitively beneficial, few studies have demonstrated EBFM's performance under climate change.

This research employed scenario analyses and management strategy evaluation (MSE) using coupled climate-enhanced multispecies assessment and fishery management models. The study utilized six high-resolution downscaled projections of oceanographic and lower trophic level conditions in the Bering Sea, driven by three global general circulation models (GCMs) under RCP 4.5 and RCP 8.5 scenarios. These projections, along with a persistence scenario, were used to drive a climate-enhanced multispecies stock assessment model (CEATTLE) for walleye pollock, Pacific cod, and arrowtooth flounder. CEATTLE incorporates temperature-specific growth and predation parameters, and climate-informed spawner-recruit relationships. The CEATTLE model outputs (ABC) were then fed into a socio-economic model (ATTACH) to simulate annual harvest recommendations, accounting for the 2 MT cap and other management regulations. The MSE evaluated three harvest scenarios: no harvest, a sloping harvest control rule (HCR), and the 2 MT cap scenario, under different climate projections. Risk assessments were conducted by analyzing the probability of declines (10%, 50%, 80%) in biomass and catch relative to a persistence scenario. Threshold analysis identified tipping points using general additive models, linking catch changes to bottom temperature.

The study found that EBFM policies, specifically the 2 MT cap, effectively ameliorated near-term climate change impacts on the Eastern Bering Sea fisheries, particularly for pollock. The 2 MT cap stabilized pollock catch until mid-century, mitigating climate-driven declines. However, these benefits diminished after mid-century as climate-driven declines exceeded the adaptive capacity of current management. Under RCP 8.5, end-of-century pollock and Pacific cod fisheries collapsed in >70% and >35% of simulations, respectively. A summer bottom temperature of 2.1–2.3 °C was identified as a tipping point for rapid declines in gadid biomass and catch. Multiyear warm stanzas exceeding this threshold became frequent in projections from ~2030 onward, more so under RCP 8.5 than RCP 4.5. The 2 MT cap had less impact on Pacific cod, while for arrowtooth flounder, it stabilized catches, especially under RCP 8.5. The analysis indicated increased risk of decline in pollock and cod across all scenarios after 2050.

The results show that EBFM policies, while not explicitly designed for climate change adaptation, provided a critical window of opportunity to prepare for and adapt to change in the Eastern Bering Sea fisheries. The 2 MT cap stabilized pollock catches and biomass, highlighting the potential of EBFM to enhance resilience in the near term. However, the study also revealed the limits of adaptation, with significant declines projected after 2050 even under EBFM, particularly under high-emission scenarios. The differing responses of species to the 2 MT cap, with pollock benefitting more than cod, highlight the species-specific nature of climate change impacts and the need for tailored management strategies. The identification of a thermal tipping point provides crucial information for proactive management. The study underscores the need for climate-resilient management approaches that account for increased variability and potential for sudden collapses, particularly in the face of extreme warming.

This study demonstrates the near-term benefits of EBFM in mitigating climate change impacts on fisheries, but also highlights its limitations in the face of severe climate change. The 2 MT cap showed efficacy for pollock, but not for Pacific cod. A critical temperature threshold was identified, informing management strategies. Future research should explore adaptive management strategies and consider the potential for hyperstability under EBFM.

The study focused on existing EBFM policies and did not evaluate alternative management approaches, such as adaptive or climate-informed policies. The model used three species, while the 2 MT cap encompasses many additional species; this simplification could influence the results. While the ATTACH model demonstrates good performance in simulating TAC and catch, uncertainty remains. The exclusion of observation and estimation errors in the MSE could have underestimated overall uncertainty.