Rapid Zooplankton Assessment: Evaluating a Tool for Ecosystem-Based Fisheries Management in the Large Marine Ecosystems of Alaska

Abstract

Ecosystem-based fisheries management (EBFM) remains an aspirational goal for management throughout the world. One of the primary limitations of EBFM is the incorporation of basic lower trophic level information, particularly for zooplankton, despite the importance of zooplankton to fish. The generation of zooplankton abundance estimates requires significant time and expertise to generate. The rapid zooplankton assessment (RZA) is introduced as a tool whereby nontaxonomic experts may produce rapid zooplankton counts shipboard that can be applied to management in near real time. Zooplankton are rapidly counted shipboard and placed into three broad groups of zooplankton relevant to higher trophic levels: large copepods (> 2 mm), small copepods (< 2 mm), and euphausiids. A Bayesian, hierarchical linear regression modeling approach was used to validate the relationship between RZA abundances and laboratory-processed abundances to ensure the rapid method is a reliable indicator. Additional factors likely to impact the accuracy of the RZA abundance predictions were added to the initial regression model: RZA sorter, survey, season, and large marine ecosystem (Bering Sea, Chukchi/Beaufort Sea, and Gulf of Alaska). We tested models that included the random effect of sorter nested within survey, which improved fits for both large copepods (Bayes R² = 0.80) and euphausiids (Bayes R² = 0.84). These factors also improved the fit for small copepods when the fixed effect of season was also included (Bayes R² = 0.65). Additional RZA data were used to predict laboratory-processed abundances for each zooplankton category and the results were consistent with model training data: large copepods (Bayes R² = 0.80), small copepods (Bayes R² = 0.64), and euphausiids (Bayes R² = 0.88). The Bayesian models were therefore able to predict laboratory-processed abundances with an associated error when accounting for these fixed and random effects. To demonstrate the utility of zooplankton data in management, zooplankton time series from the Bering Sea shelf were shown to vary in relation to warm and cold conditions. This variability impacted commercially important fish, notably Walleye Pollock (Gadus chalcogrammus), and these time series were used by managers using a risk table approach. The RZA method provides a rapid zooplankton population estimation in near real time that can be applied to the management process quickly, thus helping to fill a gap in EBFM.