A deep convolutional neural network model for rapid prediction of fluvial flood inundation

Abstract Most of the two-dimensional (2D) hydraulic/hydrodynamic models are still computationally too demanding for real-time applications. In this paper, an innovative modelling approach based on a deep convolutional neural network (CNN) method is presented for rapid prediction of fluvial flood inundation. The CNN model is trained using outputs from a 2D hydraulic model (i.e. LISFLOOD-FP) to predict water depths. The pre-trained model is then applied to simulate the January 2005 and December 2015 floods in Carlisle, UK. The CNN predictions are compared favourably with the outputs produced by LISFLOOD-FP. The performance of the CNN model is further confirmed by benchmarking against a support vector regression (SVR) method. The results show that the CNN model outperforms SVR by a large margin. The CNN model is highly accurate in capturing flooded cells as indicated by several quantitative assessment matrices. The estimated error for reproducing maximum flood depth is 0–0.2 m for the 2005 event and 0–0.5 m for the 2015 event at over 99% of the cells covering the computational domain. The proposed CNN method offers great potential for real-time flood modelling/forecasting considering its simplicity, superior performance and computational efficiency.

[1]  Zening Wu,et al.  Depth prediction of urban flood under different rainfall return periods based on deep learning and data warehouse. , 2020, The Science of the total environment.

[2]  Gwo-Fong Lin,et al.  Development of a real-time regional-inundation forecasting model for the inundation warning system , 2013 .

[3]  J. M. Van Der Knijff,et al.  Please Scroll down for Article International Journal of Geographical Information Science Lisflood: a Gis-based Distributed Model for River Basin Scale Water Balance and Flood Simulation(2008)'lisflood: a Gis-based Distributed Model for River Basin Scale Water Balance and Flood Simulation',internatio , 2022 .

[4]  S. Calmant,et al.  Large‐scale hydrologic and hydrodynamic modeling of the Amazon River basin , 2013 .

[5]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[6]  Paul D. Bates,et al.  Improving the stability of a simple formulation of the shallow water equations for 2‐D flood modeling , 2012 .

[7]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[8]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[9]  Jeff Heaton,et al.  Early stabilizing feature importance for TensorFlow deep neural networks , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[10]  J. Nash,et al.  River flow forecasting through conceptual models part I — A discussion of principles☆ , 1970 .

[11]  Yoshua Bengio,et al.  Algorithms for Hyper-Parameter Optimization , 2011, NIPS.

[12]  Jasper Snoek,et al.  Practical Bayesian Optimization of Machine Learning Algorithms , 2012, NIPS.

[13]  Li-Chiu Chang,et al.  Clustering-based hybrid inundation model for forecasting flood inundation depths , 2010 .

[14]  Li-Chiu Chang,et al.  Building ANN-Based Regional Multi-Step-Ahead Flood Inundation Forecast Models , 2018, Water.

[15]  P. Bates,et al.  A simple inertial formulation of the shallow water equations for efficient two-dimensional flood inundation modelling. , 2010 .

[16]  P. Bates,et al.  A subgrid channel model for simulating river hydraulics and floodplain inundation over large and data sparse areas , 2012 .

[17]  Giriraj Amarnath,et al.  Modelling the flood-risk extent using LISFLOOD-FP in a complex watershed: case study of Mundeni Aru River Basin, Sri Lanka , 2015 .

[18]  Andri Fachrur Rozie,et al.  Text Classification for Sentiment Prediction of Social Media Dataset using Multichannel Convolution Neural Network , 2018, 2018 International Conference on Computer, Control, Informatics and its Applications (IC3INA).

[19]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[20]  Lothar Fuchs,et al.  An ensemble neural network model for real-time prediction of urban floods , 2019, Journal of Hydrology.

[21]  Yang Liu,et al.  A flood inundation modelling using v-support vector machine regression model , 2015, Eng. Appl. Artif. Intell..

[22]  Q. Liang,et al.  A full-scale fluvial flood modelling framework based on a high-performance integrated hydrodynamic modelling system (HiPIMS) , 2019, Advances in Water Resources.

[23]  Moncef Gabbouj,et al.  Real-Time Patient-Specific ECG Classification by 1-D Convolutional Neural Networks , 2016, IEEE Transactions on Biomedical Engineering.

[24]  Jihn-Sung Lai,et al.  A Support Vector Machine Forecasting Model for Typhoon Flood Inundation Mapping and Early Flood Warning Systems , 2018, Water.

[25]  Jerónimo Puertas,et al.  A rapid flood inundation model for hazard mapping based on least squares support vector machine regression , 2019, Journal of Flood Risk Management.

[26]  R. Moore,et al.  A surface water flooding impact library for flood risk assessment , 2016 .

[27]  Andrew Smith,et al.  Optimisation of the two-dimensional hydraulic model LISFOOD-FP for CPU architecture , 2018, Environ. Model. Softw..

[28]  S. Kanae,et al.  A physically based description of floodplain inundation dynamics in a global river routing model , 2011 .

[29]  Moncef Gabbouj,et al.  Convolutional Neural Networks for patient-specific ECG classification , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[30]  Li-Chiu Chang,et al.  Online multistep-ahead inundation depth forecasts by recurrent NARX networks , 2012 .

[31]  Peter C. Young,et al.  Visualization approaches for communicating real‐time flood forecasting level and inundation information , 2010 .

[32]  Chris Eliasmith,et al.  Hyperopt: a Python library for model selection and hyperparameter optimization , 2015 .

[33]  Paresh Chandra Deka,et al.  Support vector machine applications in the field of hydrology: A review , 2014, Appl. Soft Comput..

[34]  Li-Chiu Chang,et al.  Regional flood inundation nowcast using hybrid SOM and dynamic neural networks , 2014 .

[35]  Boualem Boashash,et al.  1-D CNNs for structural damage detection: Verification on a structural health monitoring benchmark data , 2018, Neurocomputing.

[36]  J. Neal,et al.  Reducing Inconsistencies in Point Observations of Maximum Flood Inundation Level , 2013 .

[37]  Sandhya Patidar,et al.  A machine learning approach for forecasting and visualising flood inundation information , 2020, Proceedings of the Institution of Civil Engineers - Water Management.

[38]  R. Moore,et al.  Use of high‐resolution NWP rainfall and river flow forecasts for advance warning of the Carlisle flood, north‐west England , 2009 .

[39]  Shien Tsung Chen,et al.  Physical Hybrid Neural Network Model to Forecast Typhoon Floods , 2018 .

[40]  P. Bates,et al.  A simple raster-based model for flood inundation simulation , 2000 .

[41]  Brett F. Sanders,et al.  PRIMo: Parallel raster inundation model , 2019, Advances in Water Resources.

[42]  Onur Avci,et al.  1D Convolutional Neural Networks and Applications: A Survey , 2019, Mechanical Systems and Signal Processing.

[43]  Li-Chiu Chang,et al.  Building an Intelligent Hydroinformatics Integration Platform for Regional Flood Inundation Warning Systems , 2018, Water.

[44]  Keith Beven,et al.  Probabilistic flood risk mapping including spatial dependence , 2013 .

[45]  Markus Disse,et al.  Framework for Offline Flood Inundation Forecasts for Two-Dimensional Hydrodynamic Models , 2018, Geosciences.

[46]  Michael Tschannen,et al.  Convolutional recurrent neural networks for electrocardiogram classification , 2017, 2017 Computing in Cardiology (CinC).

[47]  J. Neal,et al.  Modelling of flood hazard extent in data sparse areas: a case study of the Oti River basin, West Africa , 2017 .

[48]  Gerard B. M. Heuvelink,et al.  About regression-kriging: From equations to case studies , 2007, Comput. Geosci..