<p>Satellite based monitoring within a Geographic Information System (GIS) framework provides an effective approach for identifying actionable target areas in semi-arid regions. Sorghum is a climate-resilient crop widely cultivated in these regions, which are often characterized by erratic rainfall and limited irrigation availability. However, monitoring sorghum growth and condition has remained challenging for informed decision-making. Recent advances in satellite sensors and the improved accessibility of large volumes of satellite data through the Google Earth Engine (GEE) platform have created new opportunities for crop monitoring. Sentinel-2A MSI data (Rabi 2022–23) were used to evaluate regression models for estimating LAI from NDVI. A total of 160 field-based LAI observations were collected from Solapur and Ahmednagar districts, Maharashtra, representing diverse agro-ecological conditions. A secondary objective was to isolate sorghum pixels across the study area in order to assess spatial variability in crop distribution. The results revealed that second-order polynomial regression models outperformed linear, logarithmic, exponential, and power models, with higher coefficients of determination (R² &gt; 0.40) and lower root mean square error (RMSE: 0.83–0.89) when evaluated at the district level. When the datasets from both districts were combined, the model performance was moderate (R² &gt; 0.25; RMSE = 1.20), reflecting the influence of spatial variability. Crop area classification showed good accuracy, with Kappa coefficients exceeding 0.70 and estimates within ± 5% of official government statistics. Overall, the findings highlight the potential of integrating NDVI derived from satellite data with ground-based LAI measurements using polynomial regression models for accurate and near real-time crop monitoring. This approach can support climate-smart agriculture, precision farming, and policy-level planning in semi-arid regions.</p>