Simulation and Prediction of Land Use/Cover Changes Based on CLUE-S and CA-Markov Models: A Case Study of a Typical Pastoral Area in Mongolia

Round 1

Reviewer 1 Report

Taking Gurvanbulag, a typical pastoral area in central Mongolia, as an example, this study simulated LUCC in 2019 using CLUE-S and CA-Markov models and provided two scenarios ( CT, EP ) to predict the spatiotemporal evolution of LUCC in 2030 and 2040. There are some suggestions for this article below:

1.     What does “Error! Reference source not found” mean in the article? Is there an error in using the document management software to cite documents?

2.     Line 133. The meaning of each figure (a), (b), and (c) can be elaborated in the title of Figure 1.

3.     Line 237. The reasons for the selection of land use pattern impact factors could be elaborated on in more detail.

4.     In Table 2, how the values of different scenarios and different land use types are obtained. Moreover, the setting rules of EP scenarios should be introduced in detail. Similarly, in line 278, there should be a detailed description of how the EP scenario is set up and why it is set up this way.

5.     The "+-" symbols in Figure 5, Figure 8, Figure 11, and Figure 13 may be caused by the distortion of the picture, and can be changed to a north compass in a new map.

6.     Why the CLUE-S model predicts much smaller results for Other land types than the CA_Markov model would indicate that the CLUE-S model is more suitable for typical pastoral areas than the CA_Markov model.

7.     Why is it possible to infer that the CLUE-S model meets the development needs of local policies better than the CA_Markov model in terms of suppressing land degradation when the area of Other land predicted by the CLUE-S model is smaller than the area of Other land predicted by the CA_Markov model? Do you mean that the government is better able to curb land degradation when planning based on the prediction results of the CLUE-S model?

Author Response

请参阅附件。Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Taking Mongolia as the research object, CLUE-S model and CA Markov model are used to simulate land use/cover change under different scenarios. Through the comparison of the two models, the experiment was carried out correctly. Appropriate visualization is used in the manuscript to show the differences between different model simulation results and scenarios. However, before publishing in the journal, the author should revise the manuscript, because there are many problems in the manuscript. My suggestions are as follows:

1.      Line130, Line146, Line151, Line259: Please quote references correctly and check the full text citations.

2.      Line134: What is the data space resolution? Please give a brief description.

3.      Line161, Line289, Line292, Line294, Line306, Line309, Line322, Line324, Line345, Line346, Line349, Line364, Line378, Line390, Line403, Line415, Line438, Line458, Line555: Please indicate the correct drawing number in the text!

4.      Line318: Refer to “Death to Kappa: birth of quantity disagreement and allocation disagreement for accuracy assessment”, I suggest that the author increase the overall accuracy and Fom coefficient for accuracy verification.

5.      Line369: “-” does it represent the land type that has not been converted? Please explain it.

6.      Line382-Line384, Line408-Line410: Why the reduction of land use types predicted by the CLUE-S model is less than that predicted by the CA Markov model?

7.      Fig5, Fig8, Fig11, Fig13: “±” is the compass?

8.      Be careful not to "oversell" the significance of the CLUE-S model. It is marginally better than the CA-Markov model, but in some cases, it is superior to the CLUE-S model.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper compares simulation results of future LUCC in pastoral areas by two models, and provides knowledge on sustainable use and development of land resources. The results of the research will make a great contribution both in the academic aspect of LUCC simulation method and in the practical aspect of supporting the land development policy in pastoral areas. In order to improve the quality of the paper, it is desirable that the following minor revisions are made. 

1) In the introduction, the authors state as follows.

“However, the above-mentioned study areas are all simulations and predictions of LUCC in cities with rapid economic development or high population density, whereas pastoral areas with relatively moderate economic development and low population density are rarely reported in the comparative studies of the two models.”

However, in the discussion and conclusion, the applicability characteristics in pastoral areas were not discussed in comparison with economically developed areas. It is desirable that consideration of the methodological applicability from the viewpoint of regional characteristics will be additionally described. Is the finding that the CA_Markov model was significantly better than the CLUES model limited to pastoral regions?

 2) LUCC is represented as an acronym for land use/land cover change and land use/cover change in the abstract and introduction, respectively. In order to avoid confusion with LULC, either notation should be unified.

 3) References in the text of all figures and tables are errors.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf