Visualizing catch-age analysis: a case study

Catch-age analysis has become a standard tool for modern stock assessment. Widely available graphical techniques can be customized to provide an intuitive framework for this complex procedure. We illustrate a graphical approach using data from a stock of Pacific ocean perch (Sebastes alutus). These data encompass typical problems in stock assessment, with missing data for some years and changes in age determination and survey methodologies midway through the time series. Preliminary visual inspection of the raw data reveals important features that must be accommodated in the model development. We tailor a state-space model to address these requirements and provide appendices that completely define the model and its statistical application. We suggest graphical methods for evaluating model output, including residuals and standard errors of population biomass. Such portrayals of uncertainty help define realistic limits to the information inherent in fishery data.