Using remotely sensed data to map variability in health and wealth indicators in Accra, Ghana

Accra, Ghana is a developing world city with pronounced disparities in health and wealth. This research focuses on mapping variations in health and wealth disparities within Accra using very high resolution remotely sensed imagery. Using the 2000 Ghanaian census at the enumeration area (EA) level and a multispectral, Quickbird image with a spatial resolution of 2.4 m, we examine our ability to map small area, spatial variations in health and wealth indicators. Regression trees are used to map variations in built up area and vegetation within the city. Results indicate that there is a strong correlation between indicators of wealth and health including cooking fuel type, population density, and percentage of women with secondary education level, with remotely sensed estimates of vegetation and built up area at the both the EA and the neighborhood level.