A tool for rapid assessment of erosion risk to support decision-making and policy development at the Ngenge watershed in Uganda

This study tests a rapid, user-friendly method for assessing changes in erosion risk, which yields information to aid policy development and decision-making for sustainable natural resources management. There is currently a lack of timely, up-to-date and current information to support policy development on sustainable natural resources management in Uganda. The study was carried out in the Ngenge watershed, a typical catchment in the Ugandan Highlands, characterised by deforestation in favour of subsistence agriculture without adequate soil and water conservation measures. The watershed is experiencing soil erosion, sedimentation and flooding problems which are threatening agricultural productivity and food security. Sustainable management of environmental resources is needed to ensure a livelihood for the rural population which is dependent on the land. Historical erosion risk was evaluated in three steps using multi-temporal satellite data. First, current erosion risk was assessed by combining slope and vegetation cover during periods of high intensity rainfall. The data used for the assessment was obtained from public (free) satellite images. Erosion risk was then linked to land use and finally to the change in vegetation cover over the years 1980–2000. The analysis of erosion risk using rainfall, slope and NDVI (Normalised Difference Vegetative Index) as a proxy for vegetation cover gives an indication of the current erosion risk in the area. The results of historical vegetation cover change analysis indicate an overall increase in areas under erosion risk in the study area from 1980 to 2000. This method of erosion risk mapping provides a quick and straightforward means for identifying priority areas for interventions for soil and water resource management. Considering that resources are limited, the interventions to be appropriate have to be focused mainly on areas affected by degradation.