Determining Brown Trout Populations from Habitat Measurements. Carrie Kioski School of Science and Natural Resources Lake Superior State University Trout habitat modeling continues to be an important tool in the management of brown trout. Modeling takes into account the features of the stream habitat and uses these values to predict the abundance of brown trout. The objectives of this study were to discover which habitat variables impact trout populations and to create a model to predict these numbers. Percent cover, substrate, and depth were the variables measured in this study. These values were gathered by dividing the stream up into several transect areas across the stream. The method of electroshocking was employed in collecting the data of size, weight, and numbers ofthe brown trout. The data was arranged in a correlation worksheet and those variables that had high correlation values were placed into a forward stepwise regression. Large woody debris, with an r-value of 0.535, and mean depth, with an r-value of O.657, were found to be the variables that significantly impacted the numbers oftrout; especially those in the 6-inch size class. The model that contained these variable proved to be a good predictor of trout population with an r2-value of O.579 (p