Past and future land use/land cover changes from multi-temporal Landsat imagery in Mpologoma catchment, eastern Uganda

Abstract

Land use/cover (LULC) change is a major concern in Africa’s river basins and policy makers, environmentalists and other stakeholders tackling biodiversity and sustainable development issues in these watersheds require accurate information on past, present and future LULC projections to develop management strategies for the concerned watersheds. This study assessed the historical, current and future LULC changes in Mpologoma catchment. Remote sensing and supervised classification were used to analyze 33-year multitemporal LULC changes in Mpologoma catchment while future patterns for the next two decades were predicted using the Cellular Automata-Markov modelling technique in TerrSet’s Land Change Modeler. Initially, in 1986, the catchment was dominated by grassland (32.08%). However, most grassland (92.77%) was gradually converted to subsistence farming (75%) and built-up (15.7%). Grassland, woodland and wetland annually declined at a rate of 5.52%, 2.47% and 0.63% respectively while farmland and built-up expanded at 9.32% and 6.22% respectively and by 2019 subsistence farming was the dominant class (53.16%). Prediction results showed that by 2039, woodland, grassland, wetland and open water will decrease while there will be major increases in built-up and commercial farming from 11.61% to 27.91% and 0.18% to 0.34% respectively. Subsistence farming will continue to be the dominant land use by 2039 attributed to gains from woodland (4.7%), grassland (3.7%) and wetland (4.9%). These LULC changes indicate an intensifying land use pressure in Mpologoma catchment and provide useful information for land use planners, environmentalists and policymakers in this catchment to consider when planning for sustainable management of the watershed.

Keywords

• Land change modeler
• Multi-layer perceptron
• Remote sensing
• TerrSet
• Watershed monitoring