Files and links (2)
Published (Version of record)CC BY-NC V4.0, Open Access
Published (Version of record)CC BY-NC V4.0, Open
Journal article
FloodTransformer: Efficient real-time high-resolution flood forecasting
Environmental modelling & software : with environment data news, Vol.197, pp.1-17
02/2026
Metrics
1 Record Views
Abstract
Flood forecasting is crucial for disaster planning and risk management, yet conventional hydrodynamic-based 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 high-resolution 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: variable-size 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.
•An AI-hydrodynamic hybrid approach for real-time flood forecasting.•Calibrated 3Di model produces diverse simulations for robust AI training•FloodTransformer achieves efficient, large-scale, and high-resolution prediction.•A tokenised encoder–decoder enables multi-step predictions in a single run.•Physics-informed multi-task optimisation enhances model accuracy and consistency.
Details
- Title
- FloodTransformer: Efficient real-time high-resolution flood forecasting
- Creators
- Zhanzhong Gu - University of Technology SydneyJiachen Kang - University of Technology SydneyWenzheng Jin - University of Technology SydneyFeifei Tong - The University of AdelaideY. Jay Guo - University of Technology SydneyWenjing Jia - University of Technology Sydney
- Publication Details
- Environmental modelling & software : with environment data news, Vol.197, pp.1-17
- Publisher
- Elsevier Ltd
- Number of pages
- 17
- Grant note
- This work was supported by the New South Wales (NSW) State Emergency Service (SES) via the NSW Digital Restart Fund.
- Identifiers
- 991013338688502368
- Copyright
- © 2025 The Authors.
- Academic Unit
- Faculty of Science and Engineering
- Language
- English
- Resource Type
- Journal article