Machine Learning for Generalizable Prediction of Flood Susceptibility

TL;DR

This paper develops machine learning models trained on remote sensing and historical data to predict flood susceptibility across multiple river basins, aiming to improve generalizability and accessibility of flood prediction.

Contribution

It introduces a supervised machine learning framework that leverages geographically-distributed data for generalizable flood susceptibility prediction across basins.

Findings

Models show promising prediction performance with precision-recall metrics.

Remote sensing data enhances flood susceptibility modeling.

Compared to naive baselines, models improve prediction accuracy.

Abstract

Flooding is a destructive and dangerous hazard and climate change appears to be increasing the frequency of catastrophic flooding events around the world. Physics-based flood models are costly to calibrate and are rarely generalizable across different river basins, as model outputs are sensitive to site-specific parameters and human-regulated infrastructure. In contrast, statistical models implicitly account for such factors through the data on which they are trained. Such models trained primarily from remotely-sensed Earth observation data could reduce the need for extensive in-situ measurements. In this work, we develop generalizable, multi-basin models of river flooding susceptibility using geographically-distributed data from the USGS stream gauge network. Machine learning models are trained in a supervised framework to predict two measures of flood susceptibility from a mix of river basin attributes, impervious surface cover information derived from satellite imagery, and historical records of rainfall and stream height. We report prediction performance of multiple models using precision-recall curves, and compare with performance of naive baselines. This work on multi-basin flood prediction represents a step in the direction of making flood prediction accessible to all at-risk communities.