flood-modeling

LISFLOOD-FP implementation of model-agnostic self-calibrating loop. Inspired by AutoResearch

Map-first, climate-informed flood hazard assessment for data-scarce basins in R: rainfall extreme value analysis (GEV), rainfall-runoff simulation, terrain-based flow routing and water-depth mapping in one reproducible pipeline, with a built-in stationary-vs-nonstationary test for changing rainfall extremes.

Agentic AI tool for automated QC of SRH-2D hydraulic models. Parses mesh, boundary conditions, and materials: generates structured reports and answers natural language questions about your model.

Interactive flood-risk simulation for urban Chennai — terrain-based drainage analysis with a real-time "what-if" rainfall scenario engine. DEM pipeline → simplified hydraulic proxy (HEC-RAS upgrade planned) → live web dashboard.

AI-powered dam-break flood prediction & digital twin for hydrodynamic simulation, satellite-validated flood mapping, and real-time evacuation planning. Built with React, FastAPI, PostGIS & Google Earth Engine for Smart India Hackathon 2026 (SIH26161) uncertainty-aware disaster response for Indian rivers.

Geospatial framework for flood-risk assessment of culverts using multi-source data and ML models.

GPU 2D shallow-water flood model of Washington DC running live in the browser — built from open public data (USGS, OSM, FEMA). GPL-3.0.

Wrapper UI over HEC-RAS's engine that simplifies common workflows via a guided flow while keeping full native parameter access to every field, guaranteed by a sourced feature catalog.

RIFT-UConn: CUDA-accelerated 2D flood modeling with ATS-HDF5 rainfall, dynamic Green-Ampt infiltration, and urban drainage.

Guadalupe River flood-analysis utilities using USGS/NOAA data, HEC-RAS controller checks, and peak-flow reporting.

A Python-based geospatial analysis and optimization pipeline for designing climate-adapted, flood-resilient infrastructure at Roskilde Festival.

Terrain-routed flood-wave arrival model + human exposure for the 2026 Nepal Bhote Koshi–Trishuli flood. Validated at 3 gauges over 176 km.

High-resolution coastal flood modeling using morphological reconstruction algorithms

Official PyTorch implementation of MFPIT (Multimodal Flood Physics-Regularized Transformer) for water occurrence mapping in the Indus Delta. Fuses multimodal Earth observation (Sentinel-1/2, MODIS) and climatological forcing variables under a soft hydrological mass-balance constraint.

GPU-accelerated flood modelling research prototype seeking non-commercial collaboration

Modular Python pipeline for HAND-based flood inundation and damage estimation — from a (lat, lon) point and rainfall to a per-cell flood depth and damage map.

ML-driven 2D mesh generation for HEC-RAS - learns expert meshing practice from terrain and validates every mesh by simulation

AI-Augmented Water Resources Decision Support Platform integrating rainfall monitoring, runoff modeling, reservoir optimization, and flood-risk analysis.

Can a specific vehicle ford a flooded road? A three-level model ladder (depth threshold, AR&R hazard criterion, and coupled MPM water plus rigid-vehicle simulation on GH200/A100) with every verdict gated and reproducible. NSF SCIPE REU 2026, GeoElements Lab, UT Austin.

Proyección de anegamientos en regiones de Chile ante precipitación extrema: modelo semi-hidrológico HAND calibrado con huellas históricas (MODIS/Sentinel-1), sobre DEM Copernicus y lluvia GFS/IFS