Recorded bimodal length frequency distributions of Arctic charr, Salvelinus alpinus (L.), and brown trout, Salmo trutta L.: an effect of both population structure and sampling bias

Abstract

The length distributions of gillnet catches obtained by fishing on unexploited or slightly exploited Arctic charr, Salvelinus alpinus (L.), and brown trout, Salmo trutta L., populations are frequently bimodal. To test the hypothesis that these length distributions may not only be due to sampling bias, but also an effect of population structure, fleets of gillnets were used in sampling from allopatric populations of Arctic charr and brown trout, where the population numbers had been estimated by mark–recapture experiments. In both populations, the length distributions of the populations and the corresponding gillnet catches appeared as bimodal. The second mode of the length frequency distributions consisted of individuals from many age classes accumulated within a restricted length range, which is in turn a consequence of stagnation or low annual growth in old fish. Our results show that the gillnet catches by the use of fleets of gillnets described the overall bimodal length frequency distributions of the populations, although the fraction of larger fish may be significantly overestimated due to a strong increase in gillnet catchability by increasing fish size.