Using remotely sensed imagery to detect urban change -: Viewing Detroit from space

The data integration capabilities of Geographic Information Systems (GIS) are providing new opportunities for urban analysis. This article describes the work of a multidisciplinary ream who used spatial data from NASA's Landsat earth observation satellite to analyze the net vegetation change between 1975 and 1992 in Detroit, Michigan. Furthermore, by integrating the satellite-derived change data with census data from 1970, 1980, and 1990, this research shows how the extensive demographic changes that occurred in Detroit over the past quarter century have resulted in physical landscape changes detectable from space. Strong correlations were found between the patterns of social, economic, and demographic data and the pattern of vegetation change seen in the satellite imagery, both for certain points in time and changes over time. These correlations suggest that the imagery reflects processes of urban growth, inner-city decline, population shifts, and change in urban form.