Vegetation Change Prediction with Geo-Information Techniques in the Three Gorges Area of China

A computerized parametric methodology was applied to monitor, map, and estimate vegetation change in combination with 3S (RS-remote sensing, GIS-geographic information systems, and GPS-global positioning system) technology and change detection techniques at a 1:50000 mapping scale in the Letianxi Watershed of western Hubei Province, China. Satellite images (Landsat TM 1997 and Landsat ETM 2002) and thematic maps were used to provide comprehensive views of surface conditions such as vegetation cover and land use change. With ER Mapper and ERDAS software, the normalized difference vegetation index (NDVI) was computed and then classified into six vegetation density classes. ARC/INFO and ArcView software were used along with field observation data by GPS for analysis. Results obtained using spatial analysis methods showed that NDVI was a valuable first cut indicator for vegetation and land use systems. A regression analysis revealed that NDVI explained 94.5% of the variations for vegetation cover in the largest vegetation area, indicating that the relationship between vegetation and NDVI was not a simple linear process. Vegetation cover increased in four of areas. This meant 60.9% of land area had very slight to slight vegetation change, while 39.1% had moderate to severe vegetation change. Thus, the study area, in general, was exposed to a high risk of vegetation cover change.