A Spatiotemporal Deep Learning Approach for Urban Pluvial Flood Forecasting with Multi-Source Data

Round 1

Reviewer 1 Report

This study presents a deep-learning-based forecast model for spatial and temporal prediction of pluvial flooding. Experimental results show the good performance of the proposed method. In my opinion, the main innovation of this article is development of a prediction model for pluvial flooding based on deep learning that can predict the spatial and temporal evolution of the flooding situation, especially the introduction and experiment.

1)  The abstract needs to be re-described. The current abstract does not clearly inform the organization of the issues addressed by the proposed approach? What are the steps to achieve the method proposed in this article?

2) In INTRODUCTION part, the introduction of background is too simple. Many of the latest work has not been introduced, which is not enough to fully explain the significance of this study. In particular, the application of image processing technology in other fields should also be introduced. Therefore, the authors are suggested to add some literatures. e. g.,

[1] Super-Resolution Mapping Based on Spatial-Spectral Correlation for Spectral Imagery [J]. IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(3): 2256-2268.

[2] Target-Constrained Interference-Minimized Band Selection for Hyperspectral Target Detection, IEEE Transactions on Geoscience and Remote Sensing, 2021, 59(7): 6044-6064.

3) The current experimental data are relatively simple. Can experimental data from different regions and seasons be added to verify the universality of the proposed method?

4) Is there room for improvement in the proposed method in this paper? It is suggested that the authors add the analysis of the shortcomings of the proposed method in the conclusion part, and what can be improved in the future?

5) In addition, there are some grammatical errors in the article, which need further careful proofreading.

Author Response

Dear Reviewer,

thank you for reviewing the manuscript, rating it, and commenting. Below you will find my responses to the individual points. In the attached document you will find our responses to each of the points.

Many greetings,

Benjamin Burrichter

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present an interesting paper on a relevant topic, particularly with storm event behavior trends expected in a climate change context.

The paper is well structured and presented.

It would be very useful if the authors could provide an analysis of the improvements of the proposed methodology. That is, what improvements does this methodology provide in terms of prediction accuracy and calculation time with respect to other previously available ones?

Author Response

Dear Reviewer,

thank you for reviewing the manuscript, rating it, and commenting. Below you will find my responses to the individual points. In the attached document you will find our responses to each of the points.

Many greetings,

Benjamin Burrichter

Author Response File: Author Response.docx

Reviewer 3 Report

This article presents a spatiotemporal deep learning approach for urban pluvial flood forecasting with multi-source data. It is an interesting, up to date and well-written article match the journal scope. However, improvements are suggested to the authors.

Line 42. In the recent literature there are some recent approaches indicating that flood forecasting could be achieved using hydrological modeling and numerical weather prediction (https://doi.org/10.1002/met.2079).

Line 108. Clearly highlight the novel points of the current approach.

Line 289. Add the position of the study area in the European Union. A location map is necessary.

Line 306. How the data collected? Which was their spatial resolution, cover period, homogeneous data?

Line 487. Add some comments about the meaning of the errors metrics (RMSE/CSI) and their advantage/disadvantage according to the literature (https://doi.org/10.1016/j.atmosres.2022.106017).

Line 623. It is more preferable to have the figures in the results section and not in the discussion.

Finally, add and justify some targets for future research.  

 

 

 

 

 

 

Author Response

Dear Reviewer,

thank you for reviewing the manuscript, rating it, and commenting. Below you will find my responses to the individual points. In the attached document you will find our responses to each of the points.

Many greetings,

Benjamin Burrichter

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thanks for the authors' reply. I don't have any other questions.

Reviewer 3 Report

All comments adressed an now the article is ready for publication.