Elevation data is generally associated with topographic maps and expressed by contours and spot elevations. However, elevation data is also essential to the proper classification of land use and land cover by remote sensing techniques. Absolute elevation governs various types of vegetative growth as does the degree and direction of slope. However, the effect of terrain aspect (slope and direction) on reflectance is of even greater significance. The angular relationship of a surface to the sun can significantly effect the radiometric response. For flat areas this effect is contract but she slopes are involved they must be considered if automated classification is to be applied. To overcome the terrain aspect effect, rating of multi-spectral responses is used in an attempt to eliminate the terrain aspect. However, this is only partially effective and when the terrain aspect controls group cover it is of little or no value.

The proper solution to the aspect problem is to obtain and utilize suitable digital elevation data in conjunction with remote sensor response. The relationship of the aspect to the response can be developed by theory of empirical test and when properly applied can neutralize the aspect response. Then the classification of land use and land cover can proceed without the unwanted radiometric anomalies of terrain aspect.

In relatively well mapped areas such as the United States and Europe, digital data can be developed from topographic maps or from the stereo aerial photographic movie. For poorer mapped areas (which involved most of the world's land areas), a satellite designed to obtain stereo data offers the best hope for a digital elevation database. Such a satellite, known as Mapsat, has been defined by the U.S. Geological Survey. Utilizing modern solid state technology, there is no reason why such stereo data cannot be acquired simultaneously with the multispectral response, thus simplifying the overall problem of land use and land cover classification.