| dc.description.abstract | The massive effect of soil erosion is one of the major sources of land degradation that affects human lives in many ways which occur mainly due to deforestation, poor agricultural practices, overgrazing, wildfire and urbanization. In view of this, a study was conducted in Rwampara district in Uganda to predict and assess soil erosion as it is influenced by long-term LULC changes.
Rwampara, being a hilly district, has significant soil erosion issues. To examine the effects of land cover on water erosion, the land cover in the district was studied for 1999, 2010, 2020 and then predicted the expected land cover in the next forty years using geospatial techniques such as remote sensing and Geographic Information System (GIS). The average annual soil loss was calculated using GIS-based Universal Soil Loss Equation (USLE) model. To compute the temporal and spatial calculation of soil erosion in the model, rainfall erosivity factor, topographic factor, land cover and management factor as well as erodibility factor were used.
In Rwampara district, a classification scheme which defines the land cover classes was considered and new developments (agriculture, and built-up areas) are emerging every day. Seven major land cover classes were identified for mapping the case study district including; built-up areas, bushland, grassland, tropical rain forest, small scale farming, wetland and woodland. Quantification of spatial and temporal dynamics of land use/cover changes was accomplished by using three satellite images (1999, 2010, 2020), and classified them via supervised classification algorithm and finally applied post-classification change detection technique in GIS. The Cellular Automata–Markov model in Land Change Modeler (Terrset) was also applied to simulate and predict future land cover maps.
The increase was observed in agricultural areas, built-up area, small scale farming and forest cover from 1999 throughout 2020. On the other hand bushland, grassland, wetland and woodland followed a declining trend. The driving force behind this change was population growth and the government and NGO’s interventions in strengthening forest natural resource management policies. The Cellular Automata–Markov model was applied to simulate and predict future LULC maps. The Markov-CA was used to predict the land-use change in 2020 and project changes in 2060 by extrapolating current trends. The projected land cover for 2060 revealed more built-up areas(from 8.6% to 13.6%), a potential expansion in the croplands(43% to 64.4%), complete vanishing of woodlands and drastic reduction of the rest of the land covers, which leads to rising to notable increases in erosion risk.
Therefore, using reclassification in GIS, erosion was categorized as low erosion, moderate erosion, high erosion and very high erosion to calculate the effect/ severity soil potential loss on of the land-use/cover. Generally in 2020, very high erosion coverage was observed in the central steep parts of the district and depending on the total area and percentage coverage of each land cover class in the district, very high erosion was more observed in woodland, tropical rain forest and wetland with 10.82%, 8.09% & and 4.74% respectively. Bushland, small scale farming, grassland and built-up areas were less affected by soil erosion though some patches of built-up areas had high soil erosion losses.
Therefore land use planning ought to be undertaken in light of the mounting pressure for settlement and agriculture occurring in the study area for appropriate decisions on soil conservation | en_US |
