Physics-Aware Downsampling with Deep Learning for Scalable Flood Modeling
This work addresses the problem of scalable flood forecasting for disaster management, though it is incremental as it builds on existing downsampling methods with deep learning enhancements.
The paper tackles the computational scalability issue in flood modeling by training a deep neural network for physics-aware downsampling of terrain maps, achieving a significant reduction in computational cost while maintaining accurate flood predictions.
Background: Floods are the most common natural disaster in the world, affecting the lives of hundreds of millions. Flood forecasting is therefore a vitally important endeavor, typically achieved using physical water flow simulations, which rely on accurate terrain elevation maps. However, such simulations, based on solving partial differential equations, are computationally prohibitive on a large scale. This scalability issue is commonly alleviated using a coarse grid representation of the elevation map, though this representation may distort crucial terrain details, leading to significant inaccuracies in the simulation. Contributions: We train a deep neural network to perform physics-informed downsampling of the terrain map: we optimize the coarse grid representation of the terrain maps, so that the flood prediction will match the fine grid solution. For the learning process to succeed, we configure a dataset specifically for this task. We demonstrate that with this method, it is possible to achieve a significant reduction in computational cost, while maintaining an accurate solution. A reference implementation accompanies the paper as well as documentation and code for dataset reproduction.