Sampling and a mortality model of a Norwegian fjord cod ( Gadus morhua L.) population

Abstract

An assessment model was developed of a coastal cod population integrating data on catch-at-age from commercial catches and survey data and tag returns from commercial and survey trawl fisheries. Cod were tagged during the period 1990–1995 and recaptured in the commercial and research survey. Quarterly trawl surveys were conducted during 1990–1996, and cod were also sampled from the commercial catches during 1994–1996. The model was fitted to the different sets of data using a weighted least square minimisation. The model appeared to fit the data sets reasonably well, and the sensitivity analysis indicated that the major tensions tended to be between the survey and the commercial catch-at-age data. The biomass of 2+ cod was estimated to be 2–3 t km −1 and the age structure of the stock was dominated by the strong 1987 year-class. The estimated total mortality rate ( Z=0.4–0.5 per year) of 2+ cod was low compared to most other cod stocks, mostly because of a low fishing mortality rate ( F≤0.3 per year). This is a much lower fishing mortality than observed in coastal cod stocks in southern Norway. This low fishing mortality rate implies that VPA-type models would have poor convergence and hence give low precision. Thus, the integrated model using tagging data may be used for stocks exposed to low fishing mortality and where there is a scarcity of commercial catch-at-age time series.