The availability of satellite data and the advantages offered by automatic machine-processing of such data have opened up new and exciting possibilities for developing ground cover maps. Two Landsat analyses techniques (an unsupervised clustering algorithm called Landsat Signature Development Program, and an interactive method based on the Multispectral Image Analyzer) are used to compare computer generated character maps to known earth surface features. Data samples are shown and applications are discussed. Reference is made to the value of the digital computer in natural and man-made features mapping and monitoring, and suggestions are given for further research.

Date of this Version