Soil salinity is a pervasive environmental stressor triggering land degradation in irrigated areas of arid regions, notably in the Aral Sea Basin of Central Asia. This study presents a high-resolution diagnostic approach to map soil apparent electrical conductivity (ECa) by integrating proximal Electromagnetic Induction (EMI) surveys with multi-sensor satellite imagery within the Google Earth Engine (GEE) platform. Using an EM38-MK2 sensor, we collected 21,396 in-situ ECa measurements across 526.4 ha of agricultural fields in the Khorezm region of Uzbekistan. These measurements were combined with multi-temporal Sentinel-2 (NDVI, NDWI, EVI), Sentinel-1 (VV, VH), and ALOS-2 PALSAR-2 (HH, HV) composites from the 2024 irrigation season. A Random Forest (RF) regression model was trained to predict ECa, with robust validation performance (R² = 0.925, RMSE = 4.27 dS/m). According to the analysis NDVI during critical crop growth stages and Sentinel-1 VH backscatter were found out as the most influential predictors, underlining the synergy between crop water stress and soil salinity. The classification of the resulting 20m resolution salinity map revealed that 69.3% of the studied area is affected by high to extreme salinity while 18.4% (96.4 ha) of the lands fall into the severe salinity category. Our findings demonstrate that the integration of multi-sensor remote sensing with dense EMI surveys provides a scalable, cost-effective framework for precise assessment of soil salinity, providing essential insights for sustainable land management in irrigated croplands.