Prioritizing data collection for assessment purposes

Fishery stock assessment scientists do their best to identify uncertainties and fully describe them using the best modeling approaches available. Models are only as good as the input data, and “gather more data” always seems to be a research recommendation at the end of the assessment process.  But, what kind of data?  We cannot have more of all types of data when we are working on a budget, so how do we make an intelligent recommendation of which type of data to prioritize? It costs about $1,200 per sea day for observer coverage in the southeast.  Each otolith costs approximately $20-$32 to age.  Each recreational angler intercept survey runs $35-$40 depending on the season, and it costs $70 per successful phone interview. The research fishery vessel is the most expensive and variable at $6,000-$22,000 for a day of sampling.  In our study, we explore the impacts of improving data collection of various data types on the stock assessment of a simulated population of fish.  We couple those simulations with a cost-benefit analysis to determine the contributions of each data type.  Through this heuristic experiment, we hope to focus the discussion of which data are most beneficial to stock assessments.