Artificial Intelligence Improves Flood Forecasting

According to a new study, artificial intelligence weather forecasting models could improve the accuracy of flood forecasting. The new AI system is shown to be as accurate as, or more accurate than, current leading methods, providing earlier lead times and warning times of large flooding events.

Human-caused climate change has increased the frequency of flooding in some regions, so the need for better flood forecasting is great. Current forecasting methods are limited by their reliance on stream gauges, which are not distributed evenly across the globe. Similarly, areas without river gauges, like in developing countries, are even harder to forecast. 

A new AI model uses 5,680 existing river gauges to predict daily streamflow in ungauged watersheds over a 7 day forecast period. It was able to provide flood predictions 5 days in advance that were as reliable or more reliable than the current Global Flood Awareness System’s (GloFAS) same day predictions. The model was also able to predict extreme weather events with a return window of 5 years equal to or better than the GloFAS predictions for events with a 1 year return window.

Results suggest that the AI model can provide flood warnings for both small and extreme events in basins with a longer lead time than previous methods and could improve access to reliable flood forecasting for developing regions.