LST and VTCI estimation from UAV-mounted thermal imaging in an agriculture-dominated CZO in the Ganga basin, India

Unmanned Aerial Vehicles (UAVs) provide high-resolution monitoring of critical zone parameters such as land surface temperature (LST), soil moisture, and vegetation indices. These tools overcome the spatial limitations of satellite-based monitoring, offering precise data for water and crop management in regions like the Ganga basin CZO, India. This study assessed crop water stress and soil moisture using indices like the vegetation temperature condition index (VTCI) and crop water stress index (CWSI), derived from LST and Normalized Difference Vegetation Index (NDVI). UAV data validation against ground-based measurements yielded correlations of R² = 0.65–0.78 and RMSE between 4–6 K. The texture-temperature vegetation dryness index (TTVDI) enhanced soil water content (SWC) estimation, ranging from 4% to 47.4%, by accounting for soil texture variability. Seasonal UAV flights revealed significant spatial variability in water stress, with upstream regions experiencing increased dryness. These insights support data-driven irrigation planning and water resource management. Accurate LST retrieval methods using radiative transfer models, NDVI thresholds, and atmospheric corrections improved the precision of crop stress assessments. Integrating UAV data with ground measurements strengthens validation frameworks, promoting sustainable water use strategies. The protocol developed in this work for high resolution assessment of crop water stress should be scalable to the upcoming satellite missions for thermal imaging such as TRISHNA and therefore has the potential for similar assessment over large areas.