Land-cover changes in sub-Saharan Africa (1982-1991): Application of a change index based on remotely sensed surface temperature and vegetation indices at a continental scale

There is an increasing need for data on land-cover changes at broad spatial scales. In this study, a remote sensing-based technique for land-cover-change analysis was applied to the African continent for the last decade. Ten years of daily, continental-scale satellite remote sensing data (from the National Oceanic and Atmospheric Administration's advanced very-high resolution radiometer) were analyzed. Deviations in the seasonal trajectory of the land surface-characterized by its brightness temperature and a vegetation index-were interpreted in terms of land-cover change on a yearly basis, from 1982 to 1991. Land-cover-change magnitudes were measured and mapped for every year. The major categories of seasonal patterns of changes were identified. Multiyear patterns of change also were categorized, and all areas affected by a continuous decrease and gain in vegetation cover over the decade were mapped. Results were related to published high-resolution remote sensing-based studies. This study's results demonstrate that interannual land-cover changes in Africa mostly involve erratic variations in land-cover conditions due to interannual climatic variability and temporary modifications in seasonality. Continuous, unidirectional change processes (decrease or gain in vegetation cover) affected less than 4% of sub-Saharan Africa during the study period. Although such linear changes are just a fraction of the total land-cover changes detected, their effect on ecosystems and sustainability of livelihood might be significant. In the semiarid regions, temporary modification of the biophysical attributes of the surface driven by variations in rainfall is the dominant process. Future work will focus on finer-scale interpretation and validation of patterns of changes. (C) Elsevier Science Inc., 1997.