Preprints
https://doi.org/10.5194/hess-2021-614
https://doi.org/10.5194/hess-2021-614
 
13 Dec 2021
13 Dec 2021
Status: this discussion paper is a preprint. It has been under review for the journal Hydrology and Earth System Sciences (HESS). The manuscript was not accepted for further review after discussion.

Deep Learning Methods for Flood Mapping: A Review of Existing Applications and Future Research Directions

Roberto Bentivoglio1, Elvin Isufi2, Sebastian Nicolaas Jonkman3, and Riccardo Taormina1 Roberto Bentivoglio et al.
  • 1Department of Water Management, Faculty of Civil Engineering and Geosciences, Delft University of Technology
  • 2Department of Intelligent Systems, Faculty of Electrical Engineering, Mathematics and Computer Science, Delft University of Technology
  • 3Department of Hydraulic Engineering, Faculty of Civil Engineering and Geosciences, Delft University of Technology

Abstract. Deep Learning techniques have been increasingly used in flood risk management to overcome the limitations of accurate, yet slow, numerical models, and to improve the results of traditional methods for flood mapping. In this paper, we review 45 recent publications to outline the state-of-the-art of the field, identify knowledge gaps, and propose future research directions. The review focuses on the type of deep learning models used for various flood mapping applications, the flood types considered, the spatial scale of the studied events, and the data used for model development. The results show that models based on convolutional layers are usually more accurate as they leverage inductive biases to better process the spatial characteristics of the flooding events. Traditional models based on fully-connected layers, instead, provide accurate results when coupled with other statistical models. Deep learning models showed increased accuracy when compared to traditional approaches and increased speed when compared to numerical methods. While there exist several applications in flood susceptibility, inundation, and hazard mapping, more work is needed to understand how deep learning can assist real-time flood warning during an emergency, and how it can be employed to estimate flood risk. A major challenge lies in developing deep learning models that can generalize to unseen case studies and sites. Furthermore, all reviewed models and their outputs, are deterministic, with limited considerations for uncertainties in outcomes and probabilistic predictions. The authors argue that these identified gaps can be addressed by exploiting recent fundamental advancements in deep learning or by taking inspiration from developments in other applied areas. Models based on graph neural networks and neural operators can work with arbitrarily structured data and thus should be capable of generalizing across different case studies and could account for complex interactions with the natural and built environment. Neural operators can also speed up numerical models while preserving the underlying physical equations and could thus be used for reliable real-time warning. Similarly, probabilistic models can be built by resorting to Deep Gaussian Processes.

Roberto Bentivoglio et al.

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on hess-2021-614', Anonymous Referee #1, 28 Dec 2021
    • AC1: 'Reply on RC1', Roberto Bentivoglio, 26 Jan 2022
  • RC2: 'Comment on hess-2021-614', Anonymous Referee #2, 28 Dec 2021
    • AC2: 'Reply on RC2', Roberto Bentivoglio, 26 Jan 2022
  • RC3: 'Comment on hess-2021-614', Anonymous Referee #3, 02 Jan 2022
    • AC3: 'Reply on RC3', Roberto Bentivoglio, 26 Jan 2022

Status: closed

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on hess-2021-614', Anonymous Referee #1, 28 Dec 2021
    • AC1: 'Reply on RC1', Roberto Bentivoglio, 26 Jan 2022
  • RC2: 'Comment on hess-2021-614', Anonymous Referee #2, 28 Dec 2021
    • AC2: 'Reply on RC2', Roberto Bentivoglio, 26 Jan 2022
  • RC3: 'Comment on hess-2021-614', Anonymous Referee #3, 02 Jan 2022
    • AC3: 'Reply on RC3', Roberto Bentivoglio, 26 Jan 2022

Roberto Bentivoglio et al.

Roberto Bentivoglio et al.

Viewed

Total article views: 1,508 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
933 556 19 1,508 17 13
  • HTML: 933
  • PDF: 556
  • XML: 19
  • Total: 1,508
  • BibTeX: 17
  • EndNote: 13
Views and downloads (calculated since 13 Dec 2021)
Cumulative views and downloads (calculated since 13 Dec 2021)

Viewed (geographical distribution)

Total article views: 1,441 (including HTML, PDF, and XML) Thereof 1,441 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 08 Aug 2022
Download
Short summary
Deep Learning methods have been increasingly used in flood mapping as an alternative to traditional modeling techniques. While promising results have been obtained, our review shows significant challenges in building Deep Learning models that can generalize across multiple scenarios, account for complex interactions, and provide probabilistic predictions. We argue that these shortcomings could be addressed by transferring recent fundamental advancements in Deep Learning.