Detection of cod (Gadus morhua) subpopulations by chemical and statistical analysis of pollutants

The hypothesis was postulated that various subpopulations may be represented in samples of fish living in an open fjord system. A statistical model based on mercury and octachlorostyrene concentrations in cod from two monitoring programs in a Norwegian fjord, was applied to test the hypothesis. The model was (mercury concentration) = b · (weight), where the coefficient b reflects the mercury concentration in the environment and to is the weight exponent. The effects of weight, time, and sex were considered. The distributions of b values, or normalized b values, were investigated by the normal plot technique. Discontinuities in the graphs were interpreted as indicative of the presence of two subpopulations. Weight-independent concentration ratios were also used in the classification of the fish on the octachlorostyrene data. The two subgroups were the same as those revealed by the mercury data. The statistical analysis of mercury and octachlorostyrene data supported the hypothesis that different subpopulations may be represented in samples of fish found in the present and similar fiord systems. The size of a fish sample must permit the testing of a general hypothesis of nonuniform distribution of concentrations of pollutants in the fish.