Wetland change detection and inundation north of lake George, western Uganda using landsat data

Abstract

A remote sensing approach has been adopted to detect seasonal change and inundation in the wetlands north of Lake George, western Uganda that is being impacted by anthropogenic factors. Five Landsat (TM and ETM+) satellite imagery of August 1987, January 1995, September 1999, March 2001 and December 2001 were used. During change detection in the wetland region, three broad classes (water bodies, exposed areas and vegetated zones) were generated in a Geographic Information System (GIS) using the Normalised Difference Water Index (NDWI), Normalised Difference Vegetation Index (NDVI), and unsupervised classification supported with ground data. Inundation mapping was carried out using density slicing on image sets with similar precipitation inputs. Change detection shows a slight decrease in vegetated and exposed wetlands areas from August 1987 through March 2001. Inundation mapping presents an increase in waterlogged expanses from January 1995 to December 2001. There is a likelihood of present anthropogenic factors playing a significant role in denuding the wetland land cover. A similar remote sensing approach could be used for monitoring temporal and spatial aspects of other wetlands in the region.

Keywords: Wetland, Change detection, Inundation, Lake George, Remote Sensing