Multi-temporal imagery from the Landsat7 satellite (ETM+) was classified using unsupervised and supervised techniques to develop a land use / land cover map of the Madiama Commune in Mali. The desired outcome of the imagery classification is to identify the basic land uses in the region, and to make inferences on overall land suitability and productivity as a resource for the commune. Since different land uses/land covers are best represented at different dates, radiometric information from different images is expected to improve classification in a difficult environment. Imagery from three dates (April 2000, August 2000, and October 2000) provide coverage of primary phenological periods in the region. Combining multiple dates improved the accuracy of the classification compared to single-date classifications. Additional field data is required to provide an adequate basis for classification and for evaluation of alternative techniques.