Abstract:
Long-term oceanographic and meteorological data were used to develop models that describe the observed variation in the recruitment of Pacific mackerel (Scomber japonicus) in the California
Current region. The models that were found to best describe recruitment included both density-dependent factors and environmental factors. Models incorporating only density-dependent factors
accounted for a maximum of 24 percent of the observed variation in recruitment. Multiple regressions models including both density-dependent and environmental-dependent factors were fitted to the data available for two time periods; 1931-1968 and 1946-1968. A model including three environmental factors and a density function was selected as the best fit to the longer period. In this model increased recruitment was associated with increased sea surface
temperature, reduced sea level, and reduced atmospheric pressure during the spawning season. This model accounted for 59 percent of the variation in recruitment from 1931-1968. The model fitted
to the 1946-1968 period included a density function and two environmental factors, on which data was not available prior to 1946. Increased recruitment with this model was associated with increased coastal upwelling and decreased offshore convergence during the spawning season. This model accounted for 76 percent of the variation in recruitment from 1946-1968. Yield per recruit simulations gave greatly different results than dynamic pool simulations. Maximum yield with a yield per recruit model occurs with an age at recruitment of 1 or less and instantaneous fishing mortalities (F) in excess of 1.0. Dynamic pool simulations that incorporate a Ricker spawner-recruit function predict that extinction of the stock will occur with the above fishing strategy. Dynamic pool simulations with both density-dependent and environmental-dependent recruitment functions were assessed to determine maximum yield with different ages at recruitment and different quota options.
