Deep Learning Models for Flood Predictions in South Florida

TL;DR

This paper develops deep learning surrogate models to rapidly and accurately predict river water levels in South Florida, significantly outperforming traditional physics-based models in speed and accuracy during extreme weather events.

Contribution

The paper introduces and evaluates several deep learning models as surrogates for complex hydrological simulations, achieving over 500x speedup and improved accuracy during extreme conditions.

Findings

DL models outperform HEC-RAS in accuracy during storms

Speedup exceeds 500x compared to physics-based models

Using recent and forecasted variables improves prediction accuracy

Abstract

Simulating and predicting the water level/stage in river systems is essential for flood warnings, hydraulic operations, and flood mitigations. Physics-based detailed hydrological and hydraulic computational tools, such as HEC-RAS, MIKE, and SWMM, can be used to simulate a complete watershed and compute the water stage at any point in the river system. However, these physics-based models are computationally intensive, especially for large watersheds and for longer simulations, since they use detailed grid representations of terrain elevation maps of the entire watershed and solve complex partial differential equations (PDEs) for each grid cell. To overcome this problem, we train several deep learning (DL) models for use as surrogate models to rapidly predict the water stage. A portion of the Miami River in South Florida was chosen as a case study for this paper. Extensive experiments show that the performance of various DL models (MLP, RNN, CNN, LSTM, and RCNN) is significantly better than that of the physics-based model, HEC-RAS, even during extreme precipitation conditions (i.e., tropical storms), and with speedups exceeding 500x. To predict the water stages more accurately, our DL models use both measured variables of the river system from the recent past and covariates for which predictions are typically available for the near future.