Hierarchical Bayesian estimation of recruitment parameters and reference points for Pacific rockfishes (Sebastes spp.) under alternative assumptions about the stock-recruit function

Hierarchical Bayesian meta-analysis can be a useful method for improving estimation of key parameters for harvested fish populations. In hierarchical models, data from multiple populations are used simultaneously to obtain estimates of parameters for individual populations and characterize the variability among populations. Many populations of Pacific rockfishes (Sebastes spp.) have declined off the US West Coast since the 1980s, and there is also concern for their conservation in Canada. We develop a hierarchical Bayesian meta-analysis to improve estimates of stock-recruit parameters, characterize management-related parameters (e.g., optimal harvest rate), and address uncertainties in the structural form of the stock-recruit function for Pacific rockfishes. We estimate steepness and optimal harvest rates for 14 populations of Pacific rockfishes under alternative assumptions about the underlying stock-recruit function (Beverton-Holt and Ricker). We provide a posterior predictive distribution of steepness for rockfishes that can be used as a prior in future assessments for similar populations. We also evaluate whether F-40% is an appropriate proxy for F-MSY for Pacific rockfishes and show that uncertainty in the natural mortality rate can have a significant effect on management advice derived from meta-analyses of stock-recruit data.