Data‐driven flood emulation: Speeding up urban flood predictions by deep convolutional neural networks

• Published in: Journal of flood risk management Vol. 14; no. 1
• Main Authors: Guo, Zifeng, Leitão, João P., Simões, Nuno E., Moosavi, Vahid
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01.03.2021; John Wiley & Sons, Inc; Wiley
• Subjects: Accuracy; Artificial neural networks; Catchment area; Catchments; Computer applications; Computing time; convolutional neural network; Data; data‐driven emulation; Environmental risk; fast water depth prediction; Flood analysis; Flood forecasting; flood modelling; Flood predictions; Flooding; Floods; Hydrologic data; Hyetographs; Model accuracy; Neural networks; Predictions; Risk analysis; Risk assessment; Simulation; Spatial discrimination; Spatial resolution; Urban areas; Water depth
• ISSN: 1753-318X; 1753-318X
• DOI: 10.1111/jfr3.12684

Computational complexity has been the bottleneck for applying physically based simulations in large urban areas with high spatial resolution for efficient and systematic flooding analyses and risk assessment. To overcome the issue of long computational time and accelerate the prediction process, this paper proposes that the prediction of maximum water depth can be considered an image‐to‐image translation problem in which water depth rasters are generated using the information learned from data instead of by conducting simulations. The proposed data‐driven urban pluvial flood approach is based on a deep convolutional neural network trained using flood simulation data obtained from three catchments and 18 hyetographs. Multiple tests to assess the accuracy and validity of the proposed approach were conducted with both design and real hyetographs. The results show that flood prediction based on neural networks use only 0.5% of the time compared with that of physically based models, with promising accuracy and generalizability. The proposed neural network can also potentially be applied to different but relevant problems, including flood analysis for flood‐safe urban layout planning.