AI-driven flood modeling and 3D visualization tools are transforming how communities prepare for and respond to climate risks. In this NVIDIA GTC 2024 session, Guy Schumann and Guillaume Gallion from RSS-Hydro explore how next-generation geospatial applications and high-fidelity visualizations, including NVIDIA Omniverse, can enhance disaster resilience by providing dynamic tools for decision-making, public education, and responder training.
They dive into how AI-powered platforms like FloodSENS use satellite and sensor data combined with machine learning models running on NVIDIA GPUs to create highly accurate flood maps, even in challenging conditions like cloud cover.
The key technologies and approaches include:
FloodSENS platform
FloodSENS is a machine-learning-powered flood mapping tool using U-Net models trained on optical satellite data. It delivers high-accuracy flood detection across different biomes, even under cloud cover, by integrating auxiliary datasets like digital elevation models (DEM) and water flow grids. The model runs on NVIDIA T4 Tensor Core and NVIDIA L4 Tensor Core GPUs for inference and has shown consistent results in Pakistan, Spain, and Sudan.
3D visualization with NVIDIA Omniverse
The presentation highlights how RSS-Hydro leverages NVIDIA Omniverse for high-end 3D visualizations. These immersive models help decision-makers better understand flood patterns, visualize the impact of extreme weather events, and prepare more effectively for disasters.
AI-powered flood modeling
The team showcases advanced hydraulic models capable of simulating fluvial (riverine) and pluvial (surface water) flooding. These models integrate land features such as topography, infrastructure, and urban layouts. Running on high-performance computing systems like MeluXina, equipped with NVIDIA A100 Tensor Core GPUs, the models enable ultra-high-resolution simulations at city and regional scales.
What you’ll learn:
- Data-driven flood assessment: Explore how AI models can create detailed flood maps using satellite and sensor data, improving situational awareness during disasters.
- Visualization for decision support: Understand the benefits of 3D visualization tools in risk communication, public education, and responder training.
- AI’s role in disaster management: Learn how machine learning models and high-performance GPUs can improve flood prediction, risk assessment, and real-time response strategies.
- Real-world implementations: Gain insights into successful deployments of RSS-Hydro’s technology in collaboration with partners like Microsoft Azure, UN WFP, and the ESA HeManEO program.
Watch the session Using Next-Generation Geospatial Applications and 3D Visualizations to Improve Climate Risk Resilience Strategies Explore more videos on NVIDIA On-Demand, and gain valuable skills and insights from industry experts by joining the NVIDIA Developer Program.
This content was partially crafted with the assistance of generative AI and LLMs. It underwent careful review and was edited by the NVIDIA Technical Blog team to ensure precision, accuracy, and quality.
Strengthening Climate Resilience with AI-Powered Flood Modeling and 3D Visualizations
