Predictive models for freshwater fish community composition

grantor: University of TorontoEmpirical models were developed to predict fish species occurrence, richness and community composition for 286 temperate lakes located in south-central Ontario, Canada based on whole-lake measures of habitat. Detailed analysis and comparison of traditional (i.e., logistic regression, discriminant analysis) and alternative (i.e., classification trees, and artificial neural networks) modeling statistical techniques show that predictive success differs among species and approaches. Neural networks (NNs) are focused upon in subsequent chapters given their potential utility in solving non-linear problems related to pattern recognition and prediction. Details of model construction, optimization and validation are illustrated, and new methods for quantifying the explanatory value of NNs are developed. Next, the utility of NNs to aid in understanding and predicting species abundance using nearshore measures of lake habitat are illustrated. Together, my thesis provides ecological and methodological insights into modeling of fish communities, both of which have considerable value for the study, management and conservation of aquatic ecosystems.M.Sc