dc.description.abstract | Discarding of small fish has been and still is a large problem in many fisheries. It is a problem because most of the discarded fish die and are a direct loss to the biomass of the stock. When discards occur and it is not accounted for in the assessment, the total mortality from the stock is underestimated. Thus, it will cause bias in the VPA estimates and in analysis based on these estimates. In the present work, discards in the commercial trawl fishery for Northeast Arctic cod, during the years 1946 - 1998, were estimated from two points of view. The first was to assume that the USSR did not have any discards and adjust the other countries' catch at age distributions according to the USSR landings. The second was to use selectivity properties and abundance estimates to estimate age distributions in the catches and adjust the catches according to these. The differences between the estimated and reported catch numbers were then regarded as discards. New VPA numbers at age were estimated and the biological reference point Fmed was calculated. The results from the first approach were considered not to be any reliable. The results from the second approach were as expected, high in the 1950s and 1960s with a decreasing trend towards the 1980s. The results indicate also that the USSR had discards or that they had errors in the reported catches. Discards were shown to have a large influence on the VPA estimates, especially stock numbers at age three, which resulted in a small increase in the value of Fmed. Due to the large errors in the VPA stock numbers of Northeast Arctic cod, as indicated by the estimates of discards, it is suggested that the Arctic Fisheries Working Group should revise the catch numbers at age that are used in the VPA. | en_US |