Effects of varying natural mortality and selectivity on the assessment of eastern Baltic cod (Gadus morhua Linnaeus, 1758) stock

In 2014, the International Council for the Exploration of the Sea (ICES) was unexpectedly unable to provide an analytical assessment of eastern Baltic cod stock; factors such as data issues, assessment methodology, and the ecological situation of cod were indicated as the reasons for this failure. Some evidence suggests that the natural mortality (M) of cod could increase substantially in forthcoming years and that the selectivity could change. In this paper, age-structured and stock-production assessment models were applied to simulate the dynamics of cod stock; in the models, both constant and increasing natural mortalities were permitted. In the age-structured model, the effects of selectivity related to the cod size on the cod assessment were also analysed. In addition, stock with characteristics similar to Baltic cod stock and increasing natural mortality was generated and assessed with the age-structured model using both constant and increasing M. It was shown that models with increasing natural mortality of cod in recent years perform much better than models with constant natural mortality in terms of the distribution of residuals and retrospective patterns. The models with size-dependent selectivity did not perform better than other standard assessments. The assessment of generated stock (where natural mortality was increasing) with constant natural mortality in the assessment model showed a poor distribution of residuals and strong retrospective patterns, similar to the ICES assessment with constant M. The conducted simulations strongly suggest that the main reason for the poor recent cod assessment is the increase in natural mortality, which is not considered in the assessment methodology.