FloodTransformer: Efficient real-time high-resolution flood forecasting

Files

hdl_148775.pdf (3.89 MB)
  (Published Version)

Date

2026

Authors

Gu, Z.
Kang, J.
Jin, W.
Tong, F.
Guo, J.Y.
Jia, W.

Editors

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Journal article

Citation

Environmental Modelling & Software, 2026; 197:106832-1-106832-17

Statement of Responsibility

Zhanzhong Gu, Jiachen Kang, Wenzheng Jin, Feifei Tong, Jay Guo, Wenjing Jia

Conference Name

Abstract

Flood forecasting is crucial for disaster planning and risk management, yet conventional hydrodynamicbased approaches are often slow in response and computationally intensive. We present a hybrid framework leveraging traditional hydrodynamic modelling with a novel AI model to enable accurate, real-time, and highresolution flood prediction. To address the computational challenges of large-scale, dense flood prediction, we develop an efficient flood prediction model, FloodTransformer, which possesses three key novelties: variablesize cell embedding, tokenised time-sequence encoding, and physics-informed multi-task optimisation. These components effectively capture complex spatiotemporal dependencies, allowing accurate sequential predictions in a single run. Comprehensive evaluations on both simulated and historical flood events demonstrate FloodTransformer’s excellent accuracy and efficiency: NSE 0.9445, KGE 0.9759 for water-depth prediction, and IoU 0.8180, F1 0.8997 for inundation classification, outperforming all comparative models. With 3s inference enabling multiple horizons in one pass, FloodTransformer offers a robust and practical solution for operational flood risk management.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

© 2025 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/bync/ 4.0/).

License

Grant ID

Call number

Persistent link to this record