Delineation of Nalwekomba Inland Valley Wetland in Eastern Uganda and Prediction of Future Landcover

Abstract

Inland valley wetlands (IVWs) in Africa are potential hotspots for agriculture and artisanal fishing. They present good opportunities for the rural communities in mitigating against impacts of climate change due to their abundant water supply, diverse vegetation attributes and relatively fertile soils. Unfortunately, few small inland-valley wetlands in Sub-Saharan-Africa have been delineated, mapped and land cover (LC) information documented despite countries like Uganda having a strong wetland policy and inventory. The aim of this study was to delineate Nalwekomba IVW and catchment in Eastern Uganda, predict its future land use/cover changes (LULCC) by the year 2040, and quantify extent of LC losses and gains in the period. The Wetland Identification Model, an Arc Hydro toolset for predicting wetlands with remotely sensed data was utilized together with topographic indices as inputs to a machine learning algorithm (Random Forest) in the delineation of Nalwekomba wetland and future LULC prediction achieved using Land Change Modeler. Results reveal that Nalwekomba wetland covers 72.2 km2 with a catchment of 216.24 km2 . In the year 2020, the catchment of Nalwekomba wetland had grassland as dominant cover (30.4%), followed by tree cover (29.1%), cropland (23.6%), shrubland (10.2%), built-up & settlement (4.3%), and open water (2.5%). Future LULC predictions reveal grassland will remain dominant LC but will be reduced by 4.6% to 28.9%, tree cover may reduce by 22.5%, and shrubs by 3.88%. Land cover-area increases are expected for built-up & settlement (71.9%); open water (48.4%), and cropland (17.2 %) by 2040. These results implicate human impacts on the wetland’s land cover as the wetland’s direct-use activities increase. In-depth ecosystem investigations in relation to human impacts are suggested to provide information for effective ecosystem management