Top-down modeling and bottom-up dynamics: Linking a fisheries-based ecosystem model with climate hypotheses in the Northern California Current

In this paper we present results from dynamic simulations of the Northern California Current ecosystem, based on historical estimates of fishing mortality, relative fishing effort, and climate forcing. Climate can affect ecosystem productivity and dynamics both from the bottom-up (through short- and long-term variability in primary and secondary production) as well as from the top-down (through variability in the abundance and spatial distribution of key predators). We have explored how the simplistic application of climate forcing through both bottom-up and top-down mechanisms improves the fit of the model dynamics to observed population trends and reported catches for exploited components of the ecosystem. We find that using climate as either a bottom-up or a top-down forcing mechanism results in substantial improvements in model performance, such that much of the variability observed in single species models and dynamics can be replicated in a multi-species approach. Using multiple climate variables (both bottom-up and top-down) simultaneously did not provide significant improvement over a model with only one forcing. In general, results suggest that there do not appear to be strong trophic interactions among many of the longer-lived, slower-growing rockfish, roundfish and flatfish in this ecosystem, although strong interactions were observed in shrimp, salmon and small flatfish populations where high turnover and predation rates have been coupled with substantial changes in many predator populations over the last 40 years.