Journal of Agriculture and Ecology Research International

Human activities are altering natural land cover and influencing climate-related processes, making land-cover monitoring and forecasting important for sustainable resource management. This study assessed historical and projected land use and land cover (LULC) changes in the Upper Krishna River Basin of Maharashtra, India. LULC maps for 1990, 2000, 2010, and 2020 were prepared from Landsat and Sentinel imagery using a random forest classifier in Google Earth Engine. A digital elevation model, slope data, and a road-distance map were integrated with the classified maps in the QGIS-MOLUSCE cellular automata–artificial neural network framework. The model was trained with the 1990 and 2010 LULC maps, validated by comparing simulated and observed conditions for 2020, and then used to project LULC for 2030. The simulation achieved an overall accuracy of 74.50% and a validation kappa value of 0.72. Between 1990 and 2020, woodland and agricultural land declined, whereas built-up and barren areas expanded; the area under water bodies remained comparatively stable. The 2030 projection indicates further decreases in water bodies, woodland, and agricultural land, together with continued increases in built-up and barren areas. These projected changes suggest continuing pressure on vegetated land and surface-water resources. The findings provide a spatial basis for land-use planning, watershed management, and conservation-oriented decision-making in the Upper Krishna River Basin.