Use of Landsat Data for Moose-Habitat Analyses in Alaska

Abstract
Vegetative type maps covering .apprx. 13,000,000 ha of east-central Alaska were made using Landsat multispectral scanner data to analyze moose (A. alces) habitat. Modified clustering techniques were used: the clustering method has significant advantages over other methods for processing Landsat data. The most important advantage is that the maximum number of readily separable information classes contained in the data is identified and extracted early in the analysis. Other resource analysis applications are possible from this same basic array of classes.