Abstract Land use land cover change (LULCC) is a complex phenomenon influenced by human activities and environmental factors that significantly affect ecosystems and sustainable development. To effectively understand this dynamic process and integrate it into planning and natural resource management, current and projected LULC information is crucial. This study aimed to assess the LULCC from 2003 to 2024 and project for the next 20 years in the Jemma sub-basin of Ethiopia. We employed random forest (RF) machine-learning algorithms for classification, and a CA-ANN model to project future LULC changes using the MOLUSCE plugin in QGIS. Elevation, distance from rivers, and distance from roads were used as inputs in the projection process. The major land use types identified in the study area were cropland, grazing land, water bodies, forest land, bare land, and built-up areas. Classification accuracy ranged from 83 % to 90 %, and the kappa coefficient ranged from 0.78 to 0.88. Between 2003 and 2024, cropland, built-up areas, and water bodies increased by 6.5, 2.3, and 0.2 %, respectively, while forest land increased by 1.7 %. Conversely, grazing and bare land decreased by 9.9 % and 0.7 %, respectively, between 2003 and 2024. In 2034, cropland, water bodies, forest land, and built-up areas are projected to increase by 1.99 %, 0.01 %, 0.56 %, and 0.18 %, respectively; however, these land use types show a slight decreasing trend from 2034 to 2044, except for built-up areas. Urban areas are expected to increase by 0.48 % between 2034 and 2044, primarily at the expense of grazing and bare lands. The findings suggest that significant land use changes are shaping the ecological and socioeconomic dynamics of the sub-basin. Therefore, promoting the wise utilization of natural resources and effective land management practices is essential to ensure ecological and environmental sustainability in the sub-basin.
Cherinet et al. (2026) studied this question.