Parameters That Aren’t: Time-Varying Effects in Stock Assessment

Recent research in stock assessment modeling has been directed towards diagnosing and compensating for “parameters that aren’t”, i.e., biological, anthropogenic, and environmental effects causing the value of modeled parameters to change over time.  Changes in parameters over time could accelerate as global climate and regime shifts modify trophic relationships, while fisheries management at fine spatial scales or for ecosystem components may have different sensitivity to time-varying effects than single-species models.  These and other concerns have prompted research into time-varying parameters for gear selectivity, catchability, predation mortality, growth, population productivity, and many others.  Recent research has also sought to diagnose time-varying effects using retrospective analysis, has investigated methods for projecting likely changes in parameters under future conditions, and has explored the sensitivity of management strategies to likely time-varying effects. This symposium will focus on the modification of existing models and the development of new models to account for time-varying effects.  Because assessment model types are parameterized differently, different models are likely to permit different types of modifications for, or may already be robust to, particular time-varying effects.  Thus, the discussion will also involve the selection of assessment model or control rule types that are appropriate given different time-varying effects.  In particular, presentations will include the application of time-varying methods to a particular case study, or will involve a theoretical discussion regarding the circumstances when methods are appropriate.  Such methods will grow in importance as environmental effects continue to induce changes in fisheries and ecosystem.