| Popis: |
Information about the occurrence and proportions of endangered wild Atlantic salmon stocks in Baltic Sea catches is invaluable to the fishery managers responsible for sustaining these populations. The usefulness of stock mixture analysis (SMA) from characters of sampled fish was evaluated by applying the method in an estimation of actual and simulated catch compositions. Samples for seven variable genetic loci, expressed by allozymes, and smolt age from 17 stocks from Finland and Sweden which are potential contributors to the fishery were used with similar samples collected in 1993 from catches in a Finnish trapnet fishery along the east coast of the Bothnian Sea. Identified by their character similarities, three stock management units and their estimated catch contributions were: resident Neva stock (29%), wild baseline stocks (28%), and migratory hatchery stocks (43%). The contributions of the units varied during the fishing season. With allozyme data alone, the management units were not so distinctive. Nevertheless, three major wild stocks (Tornionjoki, Kalixalven, and Simojoki), which accounted for 68% of the total wild smolt production of the area, were similar in their allozymes and distinctive from the other stocks. The increase in reliability of wild stock composition estimates when smolt age was added to the allozyme loci was evaluated by simulation. Smolt age improved the precision of composition estimates for both the combined wild baseline stocks and their aforementioned subset comprising three major stocks; it controlled bias better for the combined wild baseline stocks than their subset. |
