The dynamics of Baltic fish stocks on the basis of a multispecies stock-production model

A multispecies stock-production model was developed and used to estimate the dynamics of Baltic cod (Gadus morhua callarias), herring (Clupea harengus), and sprat (Sprattus sprattus) stocks from 1982 to 1992. Inputs to the model were estimates of the fishing effort, recruitment indices, mean weight of fish in the stock, von Bertalanffy's growth parameters, and residual natural mortality. The model's parameters that were not known in advance (e.g., food preference parameters, catchabilities, calibration coefficient for recruitment indices) were estimated by minimizing the deviations of observed catches and food composition of cod from the Values arrived at from the model. The standard errors of the fitted parameters of the model were evaluated using a bootstrap procedure. Sensitivity analyses showed that growth parameters had the largest influence on the model outcome. The model produced estimates of the population dynamics of Baltic fish stocks consistent with the estimates from age-structured models. In the period analyzed, the biomass of cod decreased by about 80%, the biomass of herring was stable, and sprat biomass fluctuated, increasing finally to a record high level. The estimated decrease of cod biomass is reflected in a decreasing predation mortality of herring and sprat.