Can less be more? New ConEvolHer article on stock assessment uncertainty
ICES Journal of Marine Science published an article by ConEvolHer members Fabian Zimmermann and Katja Enberg on the impact of assessment and survey frequency for stock estimates and uncertainty. The study shows that considering the prevailing overall high assessment uncertainty, reducing the assessment or survey frequency may not necessarily decrease the quality of stock estimates and could therefore be considered as a possibility to lower the costs of stock assessment.
The article explores how changes in the assessment or survey frequency influences the estimates of stock size and their uncertainty, using the assessments of blue whiting (Micromesistius poutassou) and Norwegian spring-spawning herring (Clupea harengus) as case studies. To do so, assessment runs with configurations corresponding to current ICES stock assessments of both stocks are contrasted with alternative scenarios in which the survey frequency has been reduced from annual to biannual, or the assessment is carried out every second year. The following comparison of resulting stock size estimates and their confidence intervals as well as predicted catches shows that generally assessment runs with lower survey frequency do not deviate substantially from the current approach. Problematic performances occur mostly for scenarios where a survey data point is missing in the assessment year (i.e. an alternating occurrence of assessment and surveys is assumed) or for specific years that are subject to underlying issues in the assessment model or input data.
The authors conclude that within suitable assessment and management frameworks a lower frequency of surveys and assessments could be an interesting possibility to reduce assessment costs, especially for stocks in a good state and with stable population dynamics. The study also demonstrates the considerable uncertainty in current stock estimates, despite using two stocks with input data of high quality and quantity. Such uncertainty and inaccuracy is a widespread problem in stock assessment and management advice that has not been diminished significantly by increasingly sophisticated assessment methods and the associated large investment of financial and scientific resources. The present article underlines this dilemma and points towards potential solutions which reduce costly assessment efforts and instead. Reduced survey and/or assessment frequency should be incorporated into management strategy evaluations and resulting management plans.