TY - JOUR

T1 - Data driven real-time prediction of urban floods with spatial and temporal distribution

AU - Berkhahn, Simon

AU - Neuweiler, Insa

N1 - Funding Information: We thank Felix Schmid 2 2 for discussions. Furthermore, we thank Johannes Steinhauer 3 3 for his help in creating the figures for this work. The authors would like to thank the anonymous reviewers for their valuable comments and suggestions, which significantly improved the quality and clarity of this manuscript. The research is being conducted with financial support of the state of Lower Saxony and the BMU funded research project FURBAS (Entwicklung und Implementierung einer effizienten und nutzerfreundlichen Modellkette zur Frühwarnung von urbanen Sturzfluten in Hannover) [BMU, 67DAS224A].

PY - 2024/1/1

Y1 - 2024/1/1

N2 - The increase in extreme rainfall events due to climate change, combined with urbanisation, leads to increased risks to urban infrastructure and human life. Physically based urban flood models capable of producing water depth maps with sufficient spatial and temporal resolution are generally too slow for decision makers to react in time during an extreme event. We present a surrogate model with high temporal and spatial resolution for real-time prediction of water levels during a pluvial urban flood. We used machine learning techniques to achieve short computation times. The recursive approach used in this work combines convolutional and fully coupled multilayer architectures. The database for the machine learning was pre-simulated results from a physically based urban flood model. The forcing input of the prediction is precipitation and the output is water level maps with a temporal resolution of 5 min and a spatial resolution of 6 x 6 meters. The prediction performance can be considered promising for testing the model in real operational applications.

AB - The increase in extreme rainfall events due to climate change, combined with urbanisation, leads to increased risks to urban infrastructure and human life. Physically based urban flood models capable of producing water depth maps with sufficient spatial and temporal resolution are generally too slow for decision makers to react in time during an extreme event. We present a surrogate model with high temporal and spatial resolution for real-time prediction of water levels during a pluvial urban flood. We used machine learning techniques to achieve short computation times. The recursive approach used in this work combines convolutional and fully coupled multilayer architectures. The database for the machine learning was pre-simulated results from a physically based urban flood model. The forcing input of the prediction is precipitation and the output is water level maps with a temporal resolution of 5 min and a spatial resolution of 6 x 6 meters. The prediction performance can be considered promising for testing the model in real operational applications.

KW - Artificial neural network

KW - Convolutional neural network

KW - Real-time forecast

KW - Recursive prediction

KW - Temporal distribution

KW - Urban flooding

UR - http://www.scopus.com/inward/record.url?scp=85180538188&partnerID=8YFLogxK

U2 - 10.1016/j.hydroa.2023.100167

DO - 10.1016/j.hydroa.2023.100167

M3 - Article

AN - SCOPUS:85180538188

VL - 22

JO - Journal of Hydrology X

JF - Journal of Hydrology X

M1 - 100167

ER -