Proposed AdaFuse, an uncertainty-aware framework that combines Random Forest, XGBoost, and CNN-LSTM for remote sensing-based water quality prediction. The framework uses Sentinel-2 satellite imagery, spatiotemporal features, and adaptive fusion to predict water quality parameters such as chlorophyll-a, turbidity, and dissolved oxygen. The model achieved R² scores of 0.92, 0.90, and 0.88 respectively.