Predicting Eastern Mediterranean Flash Floods Using Support Vector Machines with Precipitable Water Vapor, Pressure, and Lightning Data
Round 1
Reviewer 1 Report
General concept comments:
Introduction should be rewritten and organized more logically to better show the importance and significance of this research.
The methodology of this study is reasonable and feasible.
Conclusions should be stated more concisely and clearly.
Abstract
This section requires major revision
1. It suggested to add a short problem statement because the first sentences are too general to address the direction of this study.
2. Major findings and contributions should be included in this section.
3. kindly remove the citation in abstract section
4. kindly restructure this section. Start with a short problem statement, methodology and major findings.
Introduction:
1.The introduction is not well organized. The authors should review the advance, challenges and unsolved problems in this field and state the importance and novelty of this study. And the comparison with other studies can be placed in the results and discussion section.
2. In my opinion, this section is not critical when referring to the title of the manuscript.
3. This section is more on reporting style and (does look like a report).
3. Toward the end of this section, it is recommended to include previous related research work and has to be critically discussed.
4. pls explain with some consistent keywords (in abstract, intro,and conclusion) to show gap in knowledge.
Materials and research method
1. abbreviation is to be defined at first mentioned` in text
2. This section should be more organised, structured and concise.
Result and Discussion
1. it is suggested to add standard deviation or standard error where applicable
2. RMSE ad R square should be discussed on compared with many other related studies. The discussion should not only, be within one or two previous works.
4. Even though the aim of the study is to present a better version of Ziskin and Reuveni, in my opinion, the discussion should be done in wider coverage to support the betterment of findings from this study.
5. Ziskin and Reuveni can be the benchmark, but to support /prove the work, more related previous studies are required
5. Some of the discussions are too long and more on reporting.
6. Data Presentation of this manuscript, must be improved.
Conclusion
Can be improved to reflect the objectives of the study.
Author Response
We would like to thank the reviewer for the time and effort which spent reviewing our paper. All comments, suggestions, and questions were carefully considered, and all necessary corrections were made in the revised manuscript.
Please see the attached letter.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
It is seen that a great deal of effort has been made for the study. However, there are deficiencies regarding the novelty of the study.
As stated in the study, this study only consists of the addition of lightning data to the previous study presented by Ziskin and Reuveni (2022). Also, in another study by one of the Authors (Reuveni et al. 2023) the statement "In this study, we extend the previous work by integrating nearby lightning data as a new feature in our studied dataset. The inclusion of this feature is motivated by the observation that heavy rainfall, which can lead to flood events, is often accompanied before by an increase in lightning activity. The experimental results show that the adding a 24-hour vector of nearby lightning activity improves the precision score significantly" were given. This means that the importance of the addition of lightning data was also given in the previous study by the author. This situation weakens the originality of the study.
Therefore, I suggest that this study is not suitable for publication in Remote Sensing.
Author Response
We would like to thank the reviewer for the time and effort which spent reviewing our paper. All comments, suggestions, and questions were carefully considered, and all necessary corrections were made in the revised manuscript.
Please see the attached response letter.
Author Response File: Author Response.pdf
Reviewer 3 Report
This manuscript introduces nearby lightning information as a new dataset feature into the flood prediction method proposed by other researchers. The results of score metrics indicate that this method achieves an improved performance than previous work in flood prediction. This is a meaningful and valuable work. The article is very interesting, and overall the structure and writing are very good. There are some issues that the author needs to pay attention to and provide additional explanations:
1. What is the research period of this manuscript?
2. How are the WWLLN and ICDN positioning systems used in combination?
3. When lightning events and which Lightning Detection Network are used in Figure 1?
4. What is the criteria to distinguish flood and non-flood events?
5. What is the DoY information?
Author Response
We would like to thank the reviewer for the time and effort which spent reviewing our paper. All comments, suggestions, and questions were carefully considered, and all necessary corrections were made in the revised manuscript.
Please see the attached response letter.
Author Response File: Author Response.pdf
Reviewer 4 Report
This article evaluates a methodology to predict flash floods using support vector machines in the East Mediterranean. The article is interesting and worth publication in Water Journal. Prior the acceptance I propose some comments for the authors that should be included in the text.
In the introduction add some comments about flash flood forecasting skills in the recent literature. For instance, the advantages on weather numerical prediction models and the associate increase of high local resolution makes feasible the timely prediction of flash flood events combining also hydrological models (https://doi.org/10.1002/met.2079, https://doi.org/10.3390/atmos12070902).
Additional highlight the importance for rational flood risk management as also depict from 2007/60 European directive.
The machine learning used only for susceptibility mapping or even for extreme precipitation forecast?
Clearly state the research gap and the novel points of the current approach.
Line 298. Give some examples from similar researches using the same scores.
Conclusion chapter is not very informative. Please give some more information about the results and the main concussion derived.
Author Response
We would like to thank the reviewer for the time and effort which spent reviewing our paper. All comments, suggestions, and questions were carefully considered, and all necessary corrections were made in the revised manuscript.
Please see the attached response letter.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Dear author,
Kindly rephrase the sentence to avoid 'our'.
It is suggested to use this study and passive sentences.
Author Response
Answer to Reviewer #1
We would like to thank the reviewer for the time and effort which spent reviewing our paper. All comments, suggestions, and questions were carefully considered, and all necessary corrections were made in the revised manuscript.
General comments:
Dear author,
Kindly rephrase the sentence to avoid 'our'.
It is suggested to use this study and passive sentences.
The manuscript has been revised accordingly to the reviewer comments and suggestions.
Author Response File: Author Response.pdf
Reviewer 2 Report
Reuveni et al. (2023) was published as an abstract, but it is seen that the details were shared with the scientific community. It is also seen that the most important point of this study, "The experimental results show that adding a 24-hour vector of nearby lightning activity improves the precision score significantly" has already been given. This situation reduces the originality of the article. However, it is also seen that the details will also be published at the meeting.
Author Response
Answer to Reviewer #2
We would like to thank the reviewer for the time and effort which spent reviewing our paper. All comments, suggestions, and questions were carefully considered, and all necessary corrections were made in the revised manuscript.
General comments:
Reuveni et al. (2023) was published as an abstract, but it is seen that the details were shared with the scientific community. It is also seen that the most important point of this study, "The experimental results show that adding a 24-hour vector of nearby lightning activity improves the precision score significantly" has already been given. This situation reduces the originality of the article. However, it is also seen that the details will also be published at the meeting.
The reviewer seems to misunderstand (or disregard) the normal process of scientific progress. It is routinely practiced and almost globally accepted that new scientific results are first presented at international conferences, quite often in parallel with submission for publication at a refereed journal. By its nature, an abstract submitted to a conference is just a succinct description of a much larger research, whose comprehensive scope can only be described in a full-length manuscript.
Thus, we see no contradiction between including the abstract in the conference website and the 2-minute oral presentation that our paper was awarded at the EGU General Assembly in Vienna last week (April 28th) – and its acceptance and publication as a full paper in Remote Sensing.
There is really no comparison between the exposure of the content of our research at the conference – in front of a few tents of people in the audience at most - and the vast readership of the journal. In summary, we reject the reviewer's approach and fail to see the logic in his/her claim that earlier exposure of results at a conference "reduces the originality" of our results. If anything, it only serves to highlight them and draw those that are interested in full details to read the paper.
Author Response File: Author Response.pdf
