Cloud-scalable Global Climate-driven Flood Modeling using the HEC-RAS 2D Engine.

Floods constitute the most costly and prevalent climate-sensitive acute natural hazard, posing increasing risks to global communities and critical infrastructure. We introduce a novel, high-resolution global flood modeling framework engineered for cloud-scalable execution, leveraging the US Army Corps of Engineers’ HEC-RAS 2D hydraulic engine. This system performs physics-based simulation of fluvial, pluvial, and coastal flood hazards under both baseline and future CMIP6 climate projections. The framework integrates advanced climate forcings, including high-resolution gridded precipitation, dynamically downscaled streamflow, and sea-level projections derived from a combination of regional and global datasets. Fluvial boundary conditions are synthesized via a hybrid approach combining regional frequency analysis with machine learning–based hydrograph generation. Pluvial and coastal components explicitly incorporate extreme rainfall statistics and cyclonic surge dynamics, respectively. Simulations are conducted across multiple Shared Socioeconomic Pathways (SSPs) and time horizons, resulting in flood inundation layers for different return periods. Flood depth layers are derived by projecting surface water elevation onto a newly developed high-resolution digital terrain model (DTM). This downscaling process rigorously maintains the hydrodynamic fidelity of the HEC-RAS 2D model, thereby enabling granular, asset-level flood exposure and risk assessments. By seamlessly integrating physically-based hydrodynamics with a globally scalable computational architecture, this framework significantly advances quantitative flood risk assessment, supporting rigorous, climate-informed decision-making for applications spanning insurance, engineering design, and long-term resilience planning.