Aeolian sand migration induced land degradation and desertification hotspots identification in the semi-arid rain shadow regions of Anantapur, India

This study presents a multi-temporal geospatial modeling approach to identify and map land degradation and desertification hotspots in the aeolian-dominated semi-arid regions of Bommanahal, Beluguppa, and Kanekal Mandals of Anantapur District, Andhra Pradesh, India. Landsat 4-5 TM (1990), Landsat 7 ETM+ (2000, 2010), and Landsat 8 OLI/TIRS (2020) datasets were processed through a standardized workflow comprising radiometric calibration, atmospheric correction, LST retrieval, and spectral index computation. Three diagnostic indices: Normalized Difference Vegetation Index (NDVI), Topsoil Grain Size Index (TGSI), and Normalized Difference Salinity Index (NDSI), were integrated with Land Surface Temperature (LST) to quantify vegetation stress, soil texture variability, and salinity conditions. Correlation and regression analyses were employed to evaluate the relationships between LST and index-derived DN values, after which stratified sample extraction and buffer-based zonal statistics were used to delineate surface degradation intensity. A composite hotspot map was generated using mask extraction and cell statistics to merge the most degraded pixel clusters, identifying approximately 192 km² as severe degradation zones. Model performance was validated using ROC-AUC analysis, yielding an accuracy of 0.851. The study demonstrates a reproducible GIS workflow for semi-arid degradation assessment and provides a robust spatial framework for targeted land restoration, sustainable resource planning, and long-term environmental management in vulnerable dryland ecosystems.