Jonathan Reum1
1NWFSC
May 08, 2018 9:00 (PST): FSH 105
Characterizing sources of uncertainty in future projections of the Eastern Bering Sea food web using a multi species-size spectrum model
In this talk I’ll give an overview of my ongoing efforts to (1) calibrate and validate a multi-species size spectrum model of the Eastern Bering Sea and (2) generate future projections of the EBS food web using 11 different downscaled global climate model (GCM) projections. The size spectrum model can represent different hypothesized pathways through which climate may influence system dynamics. Specifically, temperature can influence predator feeding rates and natural morality and the productivity of low trophic level groups can also be modified in accordance with down-scaled estimates for the EBS using each GCM. While there are several potential sources of structural and parameter uncertainty that may influence the projection envelope, I focus on how projection uncertainty is apportioned according to GCM, climate impact hypothesis, and fishing mortality scenario over time. In a preliminary simulation experiment, near-term projection uncertainty (2020 – 2050) of biomass for some species (e.g., forage fish, snow crab, pollock) was dominated by uncertainty related to fishing mortality scenario. However, uncertainty stemming from GCMs and the specific climate hypothesis was more important for long-term (2075-2100) projection uncertainty. The reverse pattern was observed for other species (e.g., Pacific cod, Pacific halibut). I’ll discuss how this information can be useful for prioritizing future research and and developing ensemble projections. This modeling work is part of the Alaska Climate Integrated Modeling Project (ACLIM).