3M cod assessment for different assumptions over M

Abstract

M is one of the most influential quantities in fisheries stock assessment and in providing management advice. Cod 3M Bayesian XSA model assumes a lognormal prior over M being constant for all ages and years. Using the same data as in the 2013 assessment, four different scenarios over M were tested: the one assumed in the assessment (M constant over ages and time), one with age variable over three classes and constant for time, one with time variable over three periods and constant over ages, and one variable over three age classes and three periods of time. Although the results indicate the forth scenario as the best one, inconsistencies in them make very difficult to choose one of the scenarios based on the information available and further studies are needed. Nevertheless, it would be advisable that this uncertainty on M will be tested in the MSE process. The scenario 4 could be a good starting point to analyse the uncertainty on M in the 3M cod MSE.