Kiva Oken; Northwest Fishery Science Center, NOAA

https://attendee.gotowebinar.com/register/2585985820324822359

Scientific advances in ocean modeling and institutional investments such as NOAA’s Climate, Ecosystem, and Fisheries Initiative are leading to increased interest and feasibility in including environmental information in stock assessments. One of the most common linkages from environmental information to the stock assessment’s population model is via recruitment. However, research on when and how this additional source of data improves management advice from stock assessments is limited, providing little guidance on where to focus and how to implement efforts to include environmental data in stock assessments. In this study, we simulated age-structured populations using stock synthesis, one of the most widely-used tools for stock assessment today. We then sampled data from the simulated populations and fit stock assessment models to the simulated data. Estimation models included environmental recruitment indices of varying quality, included as a “survey” of recruitment deviations, as is common in stock synthesis (versus including the environmental data as part of the population’s recruitment model, more common in state-space approaches such as WHAM). Across multiple life histories, fishing histories, and data collection histories, we found that under this common implementation, recruitment indices improved estimates of spawning output during the forecast period due to improved estimates of recruitment during the final model years. However, models with high-quality indices also demonstrated increased bias in estimates of unfished recruitment and natural mortality. Work is ongoing to identify the source of this bias and strategies to minimize it.