Towards the prediction of fisheries on seasonal to multi-annual time scales

Global climate change is disrupting the species composition and food web structure of marine ecosystems. These impacts are driven by variability in ocean conditions that affect the productivity and distribution of marine organisms. Ecosystem based fisheries management can help reduce repercussions and promote resilience in the face of changing ocean conditions. 

Rationale: Coupled global Earth System Models (ESMs) have demonstrated sNill in predicting physical and biological variables important to fish and fisheries on seasonal to decadal time scales of significance to management (ParN et al. 2019, Tommasi et al. 2017, Yeager et al. 2018). This potential has only recently been recognized and documented and is thus underutilized for management activities. Our objective is to more deeply examine the variability, uncertainty, and source of predictability of these environmental drivers, and the sensitivity of fisheries to them. These analyses will provide a broader understanding of the physical and biogeochemical (BGC) processes necessary for robust fisheries projections. 

Proposed Work: The overarching task is to incorporate an offline fish model with the results of ESMs run in initialized prediction mode to generate and assess model predictions. The work to be completed includes simulating large ensembles of seasonal to decadal scale predictions of the physical and BGC variables of relevance to fish, which will then be used to force a mechanistic fish model, FEISTY (PetriN et al. 2019). FEISTY was built to be coupled with ESMs and is able to simulate the regional variability in fish catches that spans four orders of magnitude. We will force FEISTY offline with output from NCARГs CESM Decadal Prediction Large Ensemble (CESMYDPLE) and GFDLГs CM2.1 prediction experiments (Yeager et al. 2018, ParN et al. 2019). We will utilize the decadal and seasonalYtoYannual integrations that have already been completed with these two models, respectively. Analyses will I) estimate the amount of fisheries production variance explained by different physical and BGC drivers, II) estimate the spatiotemporal variability in the predictability of the most important physical and BGC drivers, III) examine the spatiotemporal variability in the predictability of fisheries, and IV) maNe comparisons between CESM and ESM2M COBALT. The ensemble approach allows for quantification of the uncertainties that arise from different climate, ocean, and BCG models. Additionally, the results of these offline ESM FEISTY simulations will provide a baseline for the predictions made with an online version to be developed under a separate proposal (PI M. Long, Incorporating fish in Earth system predictions). 

Broader Impacts and Competition Relevance: The proposed worN will address Priority Area C: Improve the modeling of climate ocean predictability pathways and its representation in prediction/projection systems. There is a strong management need for projections of future fisheries yields with accurate estimates of uncertainty. This worN provides an opportunity to accelerate research into large scale fisheries responses to climate change and the associated uncertainties. Our research will improve the modeling and understanding of climate related physical pathways that drive biogeochemical variability in U.S. marine ecosystems and their impacts on living marine resources. Ultimately, this linkage between a fish model and ESMs is an essential step in developing a comprehensive frameworK for integrated ecosystem assessment. 

Climate Risk Area: Marine Ecosystems