3D Flood Inundation Forecasting & Mapping Using Satellite + DEM Data with Deep Learning

This study develops a comprehensive 3D flood inundation forecasting and mapping framework by integrating satellite imagery, Digital Elevation Model (DEM) data, and advanced deep learning techniques to enhance flood prediction accuracy and disaster management efficiency. The proposed approach effectively models complex spatial and terrain-driven flood dynamics, enabling precise estimation of flood extent and water depth. Statistical evaluation confirms that the dataset is complete, consistent, and well-structured, supporting robust model training and validation. By leveraging architectures such as CNN, U-Net, and CNN-LSTM, the framework improves computational efficiency and predictive performance compared to traditional hydrological and hydraulic models, while also supporting scalable and real-time flood monitoring applications. Future work will focus on incorporating real-time rainfall, river discharge, and IoT-based sensor data to strengthen temporal forecasting capabilities. The integration of high-resolution LiDAR-derived DEMs and advanced deep learning models such as Transformer-based architectures may further enhance 3D flood depth estimation and extreme event prediction. Additionally, deployment on cloud-based geospatial platforms and validation across diverse river basins will improve generalizability, operational scalability, and decision-support effectiveness for early warning systems.

1.    P.D.P.O. Peramuna et al., “Enhancing 2D hydrodynamic flood model predictions in data-scarce regions through integration of multiple terrain datasets” Journal of Hydrology, Elsevier, Volume 648,2025.Page No.132343. https://doi.org/10.1016/j.jhydrol.2024.132343 2.    Hassan Sabeh et al., “Evaluating terrain-based HAND-SRC flood mapping model in low-relief rural plains using high resolution topography and crowdsourced data” Journal of Hydrology, Elsevier, Volume 652,2025.Page No.132649. https://doi.org/10.1016/j.jhydrol.2024.132649 3.    Qingsong Xu et al., “Urban flood modeling and forecasting with deep neural operator and transfer learning” Journal of Hydrology, Elsevier, Volume 661, Part B, 2025.Page No. 133705. https://doi.org/10.1016/j.jhydrol.2025.133705 4.    Chiranjit Singha et al., “Advancing flood risk assessment: Multitemporal SAR-based flood inventory generation using transfer learning and hybrid fuzzy-AHP-machine learning for flood susceptibility mapping in the Mahananda River Basin” Journal of Environmental Management, Elsevier, Volume 380,2025.Page No. 124972. https://doi.org/10.1016/j.jenvman.2025.124972 5.    Yuta Kurihara et al., “Flood direct damage assessment due to Typhoon Ulysses by satellite images” International Journal of Disaster Risk Reduction, Elsevier, Volume 118,2025.Page No. 105067. https://doi.org/10.1016/j.ijdrr.2024.105067 6.    Anupal Baruah et al., “FIMserv v.1.0: A tool for streamlining Flood Inundation Mapping (FIM) using the United States operational hydrological forecasting framework” Environmental Modelling & Software, Elsevier, Volume 192,2025.Page No.106581. https://doi.org/10.1016/j.envsoft.2025.106581 7.    Akshay S et al., “Detection of Forest Area Density using Satellite Images” Procedia Computer Science, Elsevier,2025. 8.    Hamid Gholami et al., “Mapping flood risk using a workflow including deep learning and MCDM– Application to southern Iran” Urban Climate, Elsevier, Volume 59,2025.Page No.102272. https://doi.org/10.1016/j.uclim.2024.102272 9.    Jie Zhao et al., “Urban Flood Mapping Using Satellite Synthetic Aperture Radar Data” IEEE,2025. 10.  Yaling Lin et al., “Characterizing the capability of public DEMs for mapping global floodplain bathymetry” Journal of Hydrology, Elsevier, Volume 658, 2025.Page No.133205. https://doi.org/10.1016/j.jhydrol.2025.133205