Document Type

Conference Paper

Publication Date

2025

Publication Title

Workshop on "Tackling Climate Change with Machine Learning" at ICLR 2025

Pages

1-9

Conference Name

Workshop on "Tackling Climate Change with Machine Learning" at ICLR 2025, 24-28 April 2025, Singapore

Abstract

Coastal areas like Virginia Beach, USA, are increasingly vulnerable to flooding. To mitigate the impact of flooding, it is crucial for the City of Virginia Beach to have reliable 72-hour-ahead (3 days) forecasts of water levels at key gauge locations. To support this effort, several sensors have been installed throughout the city to monitor water levels and other environmental parameters such as wind speed, precipitation, and atmospheric pressure. Leveraging sensor data from one of these locations, we developed an uncertainty-aware deep learning model to forecast water levels. We employed deep quantile regression (DQR) to quantify variability in the predictions and examined the performance of three different model architectures. In addition to exclusively including historical data, we investigated the improvement wind forecasts provide to the accuracy of 72-hour-ahead water level predictions. The results show a twelvefold improvement in the flood forecast for a real flooding event.

Rights

© 2025 Climate Change AI. All rights reserved.

Included with the kind written permission of the copyright holder.

Original Publication Citation

Hasan, M., Schram, M., McSpadden, D., Katragadda, S., Udomvisawakul, A., Richter, H., & Liu, F. (2025, April 24-28, 2025). Uncertainty-aware deep learning framework for forecasting coastal water level in Virginia Beach. Workshop on "Tackling Climate Change with Machine Learning" at ICLR 2025, Singapore. https://www.climatechange.ai/papers/iclr2025/49

ORCID

0000-0003-1812-0572 (Richter), 0000-0001-6615-0739 (Liu)

Download

Share

COinS