An alternative classification method far northern Wisconsin lakes

Wisconsin has nearly 15 000 lakes with great variation in limnology, morpho metry, and origin. Classification of lakes into groups is a continuing goal . This study examines two alternative approaches to lake classification, on e common and the other somewhat novel. Both approaches used lake morphometr y and limnological variables and were compared for ability to form groups a nd assign lakes to groups with a high probability of correct classification . The first approach used nonhierarchical cluster analysis to form lake gro ups and discriminant analysis to put lakes into these groups. The second ap proach formed lake groups by iterative dichotomous splitting of the samplin g space into smaller and smaller subspaces. Each binary split was done usin g nonhierarchical cluster analyses on a subset of the original variables. T his iterative splitting resulted in a hierarchical classification tree with reduced dimensionality in comparison with the original data set. At each b ranch, multiple logistic regression was used to place lakes into nodes of t he tree. Validation of both approaches was performed with a resubstitution analysis of the model building data set as well as a separate validation da ta set. The decision tree method yielded significantly lower rates of miscl assification and was more easily interpreted than the discriminant analysis approach.