Landslide susceptibility analysis predicts where landslides are likely to occur, which is very important for human. There have been many methodologies to assess landslide susceptibility. However, a standardized methodology, procedure and software for susceptibility assessment is still required. This study developed suite of tools for statistically-based landslide susceptibility modelling. However, the current version looks like a specification and the scientific contributions need to be stressed. In addition, how does this software provide a standardized susceptibility analysis or how can people use it at a standard way? The differences between this study and Bornaetxea et al. (2018) should be well presented. The authors stated that LAND-SUITE provides a tool that can assist the user to reduce some common source of errors coming from the data preparatory phase, and to perform more easily, more flexible and more informed statistically based landslide susceptibility applications. But I cannot get such information. How can this tool help to reduce the errors in the data preparation and why is this tool more informed?

Major concerns:

Abstract:

1. Line 10: The authors stated that some physically based models are available. What is the purpose of this statements?
2. What is the limitations of LAND-SE? What is the difference between LAND-SUITE and LAND-SE?
3. Some results and conclusions in the specific applications should be added in the abstract.

Introduction:

4. Line 59: shows why the authors further developed the LAND-SUITE model? If so, more details should be given rather than only cite the table in the existing literature.
5. Line 64: the advantage of LAND-SUITE over LAND-SE should be stressed.

Data requirements and specifications

6. This section needs to be reorganized. Currently it is not logical and difficult to get the central information.

Software description

7. A simple schematics can be provided to show the link of three modules.
8. A table can be added to show the functions of each module.

LAND-SUITE application

9. The authors need to present how to use this tool at a standard way in the applications rather than just show some results.

Final remarks

10. Some prospects and limitations can be provided in this section.

Code availability

11. I cannot access the software via the doi: 10.5281/zenodo.5650810.

Minor concerns:

Line 25-30: add some references

Line 41: add some references

Line 41: “are available”?

Line 49: propose -> proposed

Line 55, give the version of R

This paper is written in detail and provides a practical tool including data preprocessing, variable analysis and landslide susceptibility zonation for people who engaged in landslide susceptibility assessment. However, this paper is more like the software instruction manual which mainly introduced the extension of LAND-SE. In general, there are few studies on model innovation, whether physical model or empirical statistical model development in this paper.

This software suite LAND-SUITE mainly based on statistically-based landslide susceptibility assessment models. As the prediction from statistically-based models is not only influenced by input variables, training sample distributions, but also the number and representative of samples, how to carry out landslide susceptibility assessment under the condition of lack of samples is also one of the problems that need to be solved. I would like suggest the authors further discuss the physically based models in this paper, or consider physical model as one of the directions of software extension in future research. 

Line 35-36, “As a matter of fact, a standardized methodology, procedure and software for susceptibility assessment is still missing.” How to define the standardized methodology for landslide susceptibility mapping?

Line 41, “In the literature, are available….”. Grammar mistake.

In Fig. 8, it is clear that landslide susceptibility map from LDA and LRM is quite similar, while the spatial pattern of landslide susceptibility map from QDA and NNM is similar in most places. The NNM result mainly contribute to the final CFM result. Please further explain the possible reasons.