Predicting the vertical distribution of fish populations: analysis of cisco, Coregonus artedii, and yellow perch, Perca flavescens
We develop a model based on fish behavior in temperature and oxygen gradients that yields quantitative predictions of the vertical distribution of a fish population throughout the water column. The model was tested against observed vertical distributions of cisco, Coregonus artedii, and yellow perch, Perca flavescens, in 1981 and 1982 from five Wisconsin lakes. In some cases, the model seemed adequate for cisco, but in most cases, they occupied a temperature lower than their final preferendum. Occupation of lower temperature is consistent with a response to less than ad libitum food rations expected in these oligotrophy to mesotrophic lakes. In Lake Mendota, which is eutrophic with an anaerobic hypolimnion, cisco occupied temperatures higher than predicted by the model. For perch distributions, avoidance of high light intensities appears important. We did not observe effects of interspecific segregation between cisco and perch in their vertical distributions beyond that expected from differences in their preferred temperatures. Deviations of actual distributions from predictions of our relatively simplistic two-factor model can be used to help identify and evaluate other important physical and biotic factors influencing vertical distributions.