Abstarct: The alarming increase in unauthorized groundwater withdrawals in recent years has become one of the major challenges of water resources management in Iran. Identifying illegal agricultural wells, particularly under conditions of limited access to field data, is of critical importance. This dissertation aims to identify areas with a high probability of illegal wells in Bastam, Shahroud County, using advanced GIS analyses, the fusion of Landsat 8 and Sentinel-2 satellite imagery, and electricity consumption data from smart meters. Following preprocessing and pixel-baxsed fusion of images in ENVI software, a land cover map was generated using five different classification methods. Among them, the Artificial Neural Network (ANN) method, with a Kappa coefficient of 0.93, was selected as the most accurate. In the spatial analysis stage, data from 209 licensed agricultural wells and the main waterways, along with the classified agricultural land map, were incorporated into the GIS environment. Subsequently, Euclidean distance analysis was applied to assess the distance of various points from licensed wells and waterways, Kernel Density Estimation (KDE) was employed to evaluate the spatial density of wells, and finally, a hybrid approach integrating the results of these two methods was developed. Correlation analysis and scatter plots demonstrated that the hybrid method outperformed other approaches in terms of accuracy and stability. To further enhance model accuracy, electricity consumption data from 197 agricultural wells equipped with smart meters were extracted and integrated as a complementary information laxyer in the final analysis. The results indicate that this integrative approach, relying on open-access and readily available data, can provide an accurate, cost-effective, and scalable tool for identifying areas with a high probability of illegal wells under data-scarce conditions. This method offers valuable support to water resources decision-makers in achieving sustainable groundwater management.