Assessing Spatiotemporal Land Use and Land Cover Changes Using Geoinformatics Insights from the Godavari-Purna River Sub-Basin, India

Abstract

The land use land cover (LULC) change detection is an effective approach for understanding spatial and temporal landscape dynamics driven by both natural processes and human activities, especially in environmentally sensitive river basins such as the Godavari-Purna River sub-basin, Maharashtra. In this study, LULC changes from 2015 to 2025 were assessed using multi-temporal Sentinel-2 imagery and geoinformatics techniques. Maximum likelihood classification, used for supervised classification in ArcGIS 10.8, identified seven land use classes. Classification accuracy was validated using field observations and high-resolution Google Earth Pro software data. The results indicate noticeable land transformation during the study period. A decadal decline in agricultural land occurred from 89.98% to 87.86%, while built-up areas increased from 2.25% to 3.70%, reflecting rapid urban growth. Minor variations were observed in tree cover, water bodies, flooded vegetation, rangeland, and bare ground, largely influenced by land conversion and seasonal hydrological changes. The classification achieved high reliability, with overall accuracies of 95% (2015) and 94% (2025) and Kappa coefficients of 0.86 and 0.81, respectively. Overall, the study demonstrates the strong capability of remote sensing and GIS for monitoring LULC dynamics and provides valuable inputs for sustainable watershed planning and land resource management in the Purna River Basin.