Landslide Susceptibility Mapping and Driving Mechanisms in a Vulnerable Region Based on Multiple Machine Learning Models

Round 1

Reviewer 1 Report

The paper is effectively structured, presenting the contents in a clear and well-organised manner. It begins by providing a brief overview of the topic covered, before proceeding to a more in-depth discussion of the analysis of landslide susceptibility using machine learning algorithms. 

It would be useful to detail the number of Landsat and Sentinel-2 imagery, in addition to their resolution used for the analysis, in such a way as to understand the amount of data consulted. 

In conclusion, the paper is also a well-written one with appropriate syntax and style and provides reliable and trustworthy information, thus to make it a valuable methodological reference in the literature.

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This manuscript researched on the machine learning of landslide susceptibility with four different models. The introduction is wonderful, give some well summarize of the research. Compare with the introduction, thd methodolgy is common and the conclusions is plain.

some advices for the authors,

1) As the authors metioned, the integrated models can improve the accuracy, and the author choose four traditional models for assessment. It will be better if the authors can use the integrated model.

2)line 124, the expression "inoculation condition". I'm not sure if it is correct, I didn't read this words before.

3) The authors did not talk about the assessment unit, which is important for susceptibility assessment.

4) Some data detail is lack, like the date of sentinel-2a image, mean precipitation.

5) the landslide events are between 1949-2020, which is too long, reader cannot get the current landslide status.

6) Line 391, the natural breaks method does not divide the data in average.

7) Chapter 4.2, the authors give many papers and their assessment factors, but the summary is short and rather plain.

8) line 569, the machine learning widely used for landslide susceptibility is not several decades, it is just these five years.

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Please find the attached file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

 

The paper is indeed interesting, but it needs further scientific editing.

1. Some important numerical findings should be provided in the Abstract.

2. The logical flow in the Introduction is not well-developed, and it is not clear for the readers what are the key knowledge gaps and objectives of the researchers to fill them.

3. Please be clear about the purpose of the article and the research innovation aspects at the end of the introduction, and use the suggested references to enrich your study:

10.1007/s12524-020-01155-y

10.1109/JSTARS.2022.3189528

10.1007/s12524-019-00966-y

4. Please add new table and present some climatic properties of the study area.

5. The conclusion is justified, but it could be extended, highlighting the advantages of the proposed method and specifying exactly what is new?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

I confirmed that the paper is properly revised.

Reviewer 4 Report

 The authors have addressed all my comments for this paper.