Urban environments face serious climate challenges due to rapid urban expansion, exacerbating the surface urban heat island (SUHI) phenomenon. This study proposes a hybrid framework based on Multi-sensor Fusion technology within the cloud computing environment of the Google Earth Engine platform to overcome the limitations of single data sources. Focusing on Baghdad from 2018 to 2023, Multi-spectral, radar, and atmospheric satellite datasets (Landsat 8/9, Sentinel-1, Sentinel-2, Sentinel-5P) were integrated to accurately extract morphological and environmental variables at the standardized residential neighborhood scale. A spatiotemporal model was constructed based on the random forest algorithm to evaluate the relative impact of urban drivers. The model showed high predictive efficiency, achieving a coefficient of determination (R² = 0.842) and a root mean square error (RMSE = 1.007°C). The results revealed that urban morphology (NDBI contribution of 29.43%) and air quality degradation (NO₂ contribution of 25.38%) are the main drivers of surface warming. This study provides empirical evidence of the synergistic polluted-heat-dome effect, offering a geospatial tool for urban planners to adopt sustainable, climate-resilient mitigation strategies in arid metropolitan areas.