Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I would like to thank the editor for inviting me to review this article and my specific comments are listed below:

1. The keywords should reflect the research focus of the article, and it is suggested that “climate change” and “land suitability” be added to enhance the accuracy of the search.

***The originality of the article must be strengthened. iThenticate report shows that the duplication rate reaches 20%. Authors should ensure the originality of their articles and avoid any form of plagiarism to ensure compliance with standards of academic integrity.

1. Introduction (lines 32-51). The article mentions the special characteristics of semi-arid regions, but does not clearly explain why North Minas Gerais was chosen as the study area. It is recommended that the authors add to the introduction a discussion of the importance and representativeness of the study area.

2. Methodology (lines 131-136). The Methods section should detail the basis for the selection of the Random Forest algorithm and how the algorithm parameters were adjusted for specific types of data.

3. Training and validation (lines 207-224). The training and validation section needs to detail the approach to model validation, including the strategy for cross-validation and the selection of evaluation metrics.

4. Prediction scenarios (lines 225-231). How future climate data will be combined with soil and terrain data when applying the Random Forest algorithm to projections needs to be described in more detail.

5. Results (lines 232-294). The climate projections section should provide more quantitative analysis of how the projection models affect land use and land cover types.

6. Discussion (lines 314-469). The authors discuss the performance of the model but do not provide a comparison with current techniques. It is recommended that the authors compare the performance of the current model with the advantages and disadvantages of existing techniques.

7. Conclusion (lines 435-457). A discussion of the limitations of the study and how they can be addressed in future research should be provided in the conclusion.

 

Author Response

Please see attachment. 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

According to this study, the RF algorithm performed relatively well and achieved an accuracy of 60%. The 60% accuracy is too high. My question is how to decrease this percent of accuracy to acceptable level such as 20%.

The changes in land occupation depends on how the area is used. For instance, the key factor that drive the pastures conditions is the grazing pressure. The authors do not give any information on how the different components of the land are used. Another example is the Eucalyptus: this tree is known to use a lot of water. The density of trees will determine the amount of water that these trees are going to use. The same thing for cropland, what are the kind of crops which will determine how much water these crops are utilizing. For savanna, what plants are there. Another aspect that should be taken into consideration is the changes in vegetation species with climate change. This point is critical and was not addressed.

 

Comments on the Quality of English Language

The english has to be improved

Author Response

Please see attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript is really well-written and the topic is crucial in the context of  climate change. The authors should implement the discussion highlighting also the limitations of the approach, for instance the resolution of satellite data. The authors could also compare their results with similar approach in other countries to identify possible match and mismatch that can be of help for the readers.

Author Response

Please see attachment. 

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

The paper investigates the impact of climate change on land use using an adequate method of climate change simulation. The introduction introduces very well to the problem and the aim of the research is clear. Meanwhile, the reason for doing this simulation of land use change is not clear (lines 85-87). What are those policy decisions? 

The description of the study area mentions that the study area has mostly tropical climate and the rest is semiarid but from the map in Figure 1 it can be seen that precipitation is 700-1300mm. In my opinion, it is a humid subtropical climate, typically in proximity to regions with a tropical savanna climate. In this case, the title about the semiarid region should change. 

The land use pastures should be clarified because, for me, grassland and savannas are also pasture land. What is the difference? Please provide a short description. 

The results show that the most significant land use change will be in eucalyptus, which is expected to decrease by up to 79%. Meanwhile, this species is drought resistant and maybe this prediction is wrong and the people will keep them. The native forest might decrease if the people will try to replace and increase their pastures. 

The last lines of the conclusions mention that this kind of research serves as a foundation for developing action plans to optimize agro-environmental monitoring but does not describe what are the policies that could be based on this research. Please mention some specific measures that should be taken by land planners or policymakers based on the findings of this research. 

Finally, I am not sure that the land use "forest" will decrease due to climate change, but it will probably occur due to human action to increase their pasture land and compensate for the loss of their cropland. Thus, what is the adequate policy to protect the native forests that will come under pressure?

Author Response

Please see attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed some of my concerns, but the similarity rate of this article must be reduced. A similarity rate of 20% is a very dangerous number, indicating that there are many concerns about the content of the article, at least indicating that the originality of this research is not high. The authors did not pay attention to this issue in this revision. The authors did not pay attention to this issue in this revision. They also claimed that their plagiarism check in another system showed a low repetition rate.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf