Spatial Pattern and Environmental Driving Factors of Treeline Elevations in Yulong Snow Mountain, China

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The study of the factors tree vegetation spatial distribution is a fairly urgent geographical task. However, conducting relevant research requires thorough planning and correct selection of possible variables. The manuscript analyzes spatial pattern and factors of treeline elevations, which is an urgent task. However, the methodological part of the study cannot be considered correct.

1) An objective analysis of treeline factors requires a more significant spatial sampling of objects with different climatic conditions. The experimental data used in the study is not sufficient to formulate conclusions about the treeline factors.

2) Table 2. The used set of independent variables cannot be considered correct. Some soil characteristics (for example, soil organic carbon storage) are incorrectly used as predictors of treeline. This is due to the fact that these soil characteristics largely depend on the vegetation cover. That is, it is incorrect to use soil characteristics as predictors of treeline in the context of the study.

3) A serious note concerns the source data. Tree height estimates derived from LiDAR measurements may be inaccurate and need to be verified. This is especially important given the precision that was required in the study. Tree heights obtained from LiDAR require verification based on actual data.

Additional notes:

Line 199. The Pearson correlation coefficient formula is considered well known. Therefore there is no need to show it.

The manuscript uses the Pearson correlation coefficient (line 280). At the same time, in the text to Figure 4 the Spearman correlation coefficient is indicated (line 298).

Comments on the Quality of English Language

The manuscript contains frequent repetitions of the same words. For example, lines 140-143 (“sampling”).

Author Response

Thank you for your thorough review and valuable feedback on my manuscript titled “Spatial pattern and environmental driving factors of treeline elevations in Yulong Snow Mountain, China” I appreciate your detailed comments and suggestions, which have been very helpful in improving the quality of the study. Below, I provide my responses to each of your points:

 

1. Comment 1: An objective analysis of treeline factors requires a more significant spatial sampling of objects with different climatic conditions. The experimental data used in the study is not sufficient to formulate conclusions about the treeline factors.

   Response: Thank you for pointing this out. I acknowledge the limitation of the spatial sampling in the current study. Our study investigates the distribution of treelines and the key driving factors at a local scale. In April 2024, the article "Uppermost global tree elevations are primarily limited by low temperature or insufficient moisture" was published in Global Change Biology. This study is similar in content to the aforementioned research and reaches the same conclusion: temperature and precipitation are the dominant factors. This similarity indicates the credibility of the data and methods used in this study. The methodology section has been reorganized and detailed in the attached document.

 

2. Comment 2: Table 2. The used set of independent variables cannot be considered correct. Some soil characteristics (for example, soil organic carbon storage) are incorrectly used as predictors of treeline. This is due to the fact that these soil characteristics largely depend on the vegetation cover. That is, it is incorrect to use soil characteristics as predictors of treeline in the context of the study.

   Response: I appreciate your insight regarding the use of soil characteristics as predictors. I have revised the selection of independent variables to exclude soil characteristics that are influenced by vegetation cover. The revised Table 2 now includes variables that are more appropriate for predicting treeline elevations. Additionally, I have re-run the analysis with these new variables and updated the results section to reflect these changes.

 

3. Comment 3: A serious note concerns the source data. Tree height estimates derived from LiDAR measurements may be inaccurate and need to be verified. This is especially important given the precision that was required in the study. Tree heights obtained from LiDAR require verification based on actual data.

   Response: Thank you for highlighting the potential inaccuracies in tree height estimates derived from LiDAR. The treeline on Yulong Snow Mountain is mostly distributed along steep cliffs, and we currently lack the conditions to conduct field verification in these areas. However, the LiDAR data has already been used and validated in other studies. For instance, Mathew et al.'s research, "Improvement in the delineation of alpine treeline in Uttarakhand using spaceborne light detection and ranging data," validated this data using field data and found that the treeline derived from LiDAR closely matches field data (The treeline derived from LiDAR was found to have root mean square error of ∼60 m with respect to the ground verified treeline location. The NDVI treeline was overestimated in comparison to the LiDAR treeline by an average surface distance of 290, 232, 257, and 237 m in the south, north, west, and east aspects, respectively.

 

4. Comment 4: Line 199. The Pearson correlation coefficient formula is considered well known. Therefore there is no need to show it.

   Response: I have removed the Pearson correlation coefficient formula from line 199, acknowledging that it is well known and does not need to be explicitly shown in the manuscript.

 

5. Comment 5: The manuscript uses the Pearson correlation coefficient (line 280). At the same time, in the text to Figure 4 the Spearman correlation coefficient is indicated (line 298).

 

   Response: Thank you for pointing out this inconsistency. I have reviewed the manuscript and re-examined my data. After verifying whether the data conforms to a normal distribution, we found that some data are not suitable for Pearson correlation. Therefore, we decided to switch to Spearman correlation. The specific results are detailed in the attached document.

 

I have attached the revised manuscript for your review. I believe these changes address your concerns and enhance the overall quality of the study. Thank you again for your valuable feedback.

 

Please let me know if there are any additional suggestions or if further clarification is needed.

 

Sincerely,

Chuan Lin

2024.7.10

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper looks at modelling treeline elevations based on a number of factors.

 

The paper writing is generally coherent, but there are a number of awkwardly constructed sentences and ideas throughout the paper that could do with revision and more scientific language could be used.  Also there are a number of spacing issues, lack of capital letters and poor use of punctuation.

 

The introduction isn't particularly logical in it's structure, it seems more like some independent ideas stitched together.  The introduction needs to be reworked so that context in which the work sits is more clearly written.

 

How were the 12 explanatory environmental variables used for building models chosen? What is their justification?

 

Neither Table 1 or 2 are cited in the document.  The origin and some background information on the datasets mentioned in Table 2 would be helpful in the document, a simple citation isn't sufficient.  Such as how the data was collected?

 

Why was a SVM model used? The justification on lines 347-349.  I can't help think that a probabilistic approach would be more informative. eg. Figure 3 could show error bars so the effectiveness of the model could be more clearly determined.

 

The variables in Figures 4 and 5 don't coincide. eg. S_Slope?

 

The paper requires quite a bit of work before it can be accepted.  Through the paper the flow of the message, the description of the work is poor and makes it difficult to understand.  Decriptions of what figures shows need to be improved.

 

There are quite a few results littered throughout the text, please find a way to present these so that the reader can see them at a glance. eg. l.257-271

 

l.40 - "... position of realze their fundamental niches..."

 

l.74 - "first time to identify" - really? first time Machine Learning used?

 

l.86 - what does "~" mean in this context, why not a "-"?

 

l.152 - a confusion matrix is mentioned, but I don't think I ever see this in the paper?

 

l.161 - "model.in" ?

 

l.164 - Reference not properly cited.

 

l.172 - Cite KD-tree model, you can't assume the reader will be familiar with this.

 

l.199 - Not sure you need to explain the pearson correlation coefficient formula?

 

l.212 - Reference not properly cited.

 

l.245-247 - Be consistent with thousand separator, eg. 5,794 but not 5794?

 

l.250 - How were the errors on the fluctuations calculated?

 

l.251 - Figure 2, what does isoyherm mean in the image?

 

l.276 - Figure 3b, the Thermal treeline seems to disappear on the left and right sides of the plot

 

l.284 - 0.03 is a weak positive correlation?  Almost non-existant.

 

l.296 - "...between environmental variables and environmental variables..." is a strange way to phrase it.  "indicating significant relationship (p>0.05)" seems like an incorrect definition.  There are very few significant results in this matrix, and seem to be lower correlations, it seems like there is a problem here.

 

l.377 - Reference not properly cited.

 

l.388 - "Extremely weak (r=0.001)" ?  Ultimately non-existant?

 

l.574 - The author list on this reference seems like it requires revision.

 

Comments on the Quality of English Language

The paper writing is generally coherent, but there are a number of awkwardly constructed sentences and ideas throughout the paper that could do with revision and more scientific language could be used.  Also there are a number of spacing issues, lack of capital letters and poor use of punctuation.

Author Response

Thank you for your thorough review and valuable feedback on my manuscript titled “Spatial pattern and environmental driving factors of treeline elevations in Yulong Snow Mountain, China” I appreciate your detailed comments and suggestions, which have been very helpful in improving the quality of the study. Below, I provide my responses to each of your points:

 

1. Comment: The paper writing is generally coherent, but there are a number of awkwardly constructed sentences and ideas throughout the paper that could do with revision and more scientific language could be used. Also, there are a number of spacing issues, lack of capital letters, and poor use of punctuation.

   Response: Thank you for highlighting these issues. I have revised the manuscript to improve sentence construction and coherence, ensuring that scientific language is used consistently throughout. Additionally, I have corrected all spacing issues, capitalizations, and punctuation errors.

2. Comment: The introduction isn't particularly logical in its structure, it seems more like some independent ideas stitched together. The introduction needs to be reworked so that the context in which the work sits is more clearly written.

   Response: I have restructured the introduction to present a more logical flow of ideas. The revised introduction now clearly outlines the context of the study, the research gap, and the objectives of the work.

3. Comment: How were the 12 explanatory environmental variables used for building models chosen? What is their justification?

   Response: I have added a detailed explanation in the Methods section, describing the criteria and rationale for selecting the environmental variables. The reasons for selecting these variables have been described in detail in the article.

4. Comment: Neither Table 1 nor Table 2 are cited in the document. The origin and some background information on the datasets mentioned in Table 2 would be helpful in the document, a simple citation isn't sufficient. Such as how the data was collected?

   Response: I have ensured that Tables 1 and 2 are properly cited in the text. Additionally, I have included detailed background information on the datasets in Table 2, including how the data was collected and their sources.

 

5. Comment: Why was an SVM model used? The justification on lines 347-349. I can't help but think that a probabilistic approach would be more informative. eg. Figure 3 could show error bars so the effectiveness of the model could be more clearly determined.

   Response: I have expanded the justification for using the SVM model in the Methods section, detailing its advantages for this specific study. This probability refers to the generalization that the recognition model considers to be a true treeline, this operation is implemented using the 'decision_function(X)' function in the scikit-learn library in Python. Additionally, I have included error bars in Figure 3 to better illustrate the model’s effectiveness.

 

6. Comment: The variables in Figures 4 and 5 don't coincide. eg. S_Slope?

   Response: I have corrected this error, please refer to the attachment for details. The reason for the inconsistency of variables in Figures 4 and 5 is also explained in the text

 

7. Comment: The paper requires quite a bit of work before it can be accepted. Through the paper, the flow of the message, the description of the work is poor and makes it difficult to understand. Descriptions of what figures show need to be improved.

   Response: I have thoroughly revised the manuscript to improve the flow of the narrative and the clarity of the descriptions. The explanations of the figures have been enhanced to ensure they are easy to understand and accurately convey the intended message.

 

8. Comment: There are quite a few results littered throughout the text, please find a way to present these so that the reader can see them at a glance. eg. l.257-271

   Response: I have consolidated the results into summarized sections and included additional tables and figures to present the data more clearly. This allows readers to grasp the key findings at a glance.

9. Comment: Specific line-by-line comments (l.40, l.74, l.86, l.152, l.161, l.164, l.172, l.199, l.212, l.245-247, l.250, l.251, l.276, l.284, l.296, l.377, l.388, l.574)

   Response: I have addressed each specific comment as follows:

   - l.40: Corrected to “...position to realize their fundamental niches...”

   - l.74: Deleted “first time machine learning” in this specific context.

   - l.86: Changed “~” to “-” for consistency.

   - l.152: Included the confusion matrix in the Results section.

   - l.161: Removed 'model. in'

   - l.164: Properly cited the reference.

   - l.172: Added a citation and brief explanation for the KD-tree model.

   - l.199: Removed the Pearson correlation coefficient formula.

   - l.212: Properly cited the reference.

   - l.245-247: Standardized the use of thousand separators.

   - l.250: Explained the calculation of errors on the fluctuations.

   - l.251: Clarified the meaning of “isotherm” in Figure 2.

   - l.276: Adjusted Figure 3b to ensure the Thermal treeline is visible.

   - l.284: Revised to reflect that 0.03 is a very weak positive correlation.

   - l.296: Rephrased for clarity and corrected the definition regarding significant relationships.

   - l.377: Properly cited the reference.

   - l.388: Revised to indicate that r=0.001 is effectively non-existent.

   - l.574: Reviewed and revised the author list as required.

 

I have attached the revised manuscript for your review. I believe these changes address your concerns and significantly improve the overall quality of the study. Thank you again for your invaluable feedback.

 

Please let me know if there are any additional suggestions or if further clarification is needed.

 

Sincerely,

Chuan Lin

2024.7.10

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The study is about the classification fort reeline detection and evaluation with environmental driving factors. I have a few suggestions to improve the manuscript.

Comments

Part of the comments are based on the location of the page (P) and line (L):

 

1.      In the introduction section, the authors did not mention the literature about classification or mapping treelines using machine learning algorithms. There are studies about treeline such as:

·         Nguyen, T. A., Kellenberger, B., & Tuia, D. (2022). Mapping forest in the Swiss Alps treeline ecotone with explainable deep learning. Remote Sensing of Environment, 281, 113217.

·         Maher, C. T., Dial, R. J., Pastick, N. J., Hewitt, R. E., Jorgenson, M. T., & Sullivan, P. F. (2021). The climate envelope of Alaska's northern treelines: implications for controlling factors and future treeline advance. Ecography, 44(11), 1710-1722.

·         Wang, Z., Ginzler, C., Eben, B., Rehush, N., & Waser, L. T. (2022). Assessing changes in mountain treeline ecotones over 30 years using CNNs and historical aerial images. Remote Sensing, 14(9), 2135.

Authors should include literature on this topic by emphasizing the novelty of their study. Especially, highlight the differences between their paper with Maher et al. (2021). Additionally, the selected methods also need to be supported by references.

2.      Tables were not referred to in the text. The features in Table 1 and Table 2 were produced from which dataset?

3.      Authors should clearly highlight why SVM is used for classification and XGBoost for feature ranking. They can also use the XGBoost algorithm for the classification or they can use an explainable artificial intelligence algorithm for determining feature importance. The main reasons for method choices should be explained.

4.      Why did the authors use 80% training-20% test data in the XGBoost algorithm? They used 70% training - 30% test data in SVM. Why didn't you use the same amount of data?

5.      The authors stated in P.4 L.1142-143 that they established 20 sampling points. However, in P.6 L. 229-230, they also stated that 4561 points were used as training dataset. How many samples did the authors use in the classification and feature importance analysis? Please describe the data used more clearly.

6.      Showing classification results on a map could be better for interpreting results.

7.      In P.10 L.330-335, Since the results obtained are generated from these parameters, the paragraph should be given at the beginning of the section.

Author Response

Thank you for your thorough review and valuable feedback on my manuscript titled “Spatial pattern and environmental driving factors of treeline elevations in Yulong Snow Mountain, China” I appreciate your detailed comments and suggestions, which have been very helpful in improving the quality of the study. Below, I provide my responses to each of your points:

 

1. Comment: In the introduction section, the authors did not mention the literature about classification or mapping treelines using machine learning algorithms. There are studies about treeline such as:

   - Nguyen, T. A., Kellenberger, B., & Tuia, D. (2022). Mapping forest in the Swiss Alps treeline ecotone with explainable deep learning. Remote Sensing of Environment, 281, 113217.

   - Maher, C. T., Dial, R. J., Pastick, N. J., Hewitt, R. E., Jorgenson, M. T., & Sullivan, P. F. (2021). The climate envelope of Alaska's northern treelines: implications for controlling factors and future treeline advance. Ecography, 44(11), 1710-1722.

   - Wang, Z., Ginzler, C., Eben, B., Rehush, N., & Waser, L. T. (2022). Assessing changes in mountain treeline ecotones over 30 years using CNNs and historical aerial images. Remote Sensing, 14(9), 2135.

   Response: Thank you for pointing that out. I have updated the introduction to include these important references. The revised introduction now discusses the application and potential of machine learning in treeline research, and the final paragraph emphasizes the novelty of our study.

 

2. Comment: Tables were not referred to in the text. The features in Table 1 and Table 2 were produced from which dataset?

   Response: I have ensured that all tables are properly cited in the text. Additionally, I have clarified the origin of the features listed in Table 1 and Table 2, including detailed information about the datasets from which these features were derived.

 

3. Comment: Authors should clearly highlight why SVM is used for classification and XGBoost for feature ranking. They can also use the XGBoost algorithm for the classification or they can use an explainable artificial intelligence algorithm for determining feature importance. The main reasons for method choices should be explained.

   Response: I have added a detailed explanation in the methodology section to justify the use of SVM for classification and XGBoost for feature ranking. The reasons for choosing these methods are based on their respective advantages in handling the specific characteristics of our dataset and research objectives. We used SVM because our data can be broadly divided into upper and lower boundaries, and we believe SVM is particularly suitable for this study as it can find a "margin" to effectively distinguish between these two classes, helping us identify the data we need. The XGBoost model was chosen for its excellent data fitting capabilities and high result reliability.

 

4. Comment: Why did the authors use 80% training-20% test data in the XGBoost algorithm? They used 70% training - 30% test data in SVM. Why didn't you use the same amount of data?

   Response: I have revised the manuscript to explain the rationale behind the different training-test splits used for SVM and XGBoost. The 80% training and 20% testing split for XGBoost was chosen to optimize the performance of the model's feature importance analysis. On the other hand, the 70% training and 30% testing split for SVM was used to balance the training and validation stages of classification, ensuring more accurate identification of the treeline.

 

5. Comment: The authors stated in P.4 L.142-143 that they established 20 sampling points. However, in P.6 L.229-230, they also stated that 4561 points were used as training dataset. How many samples did the authors use in the classification and feature importance analysis? Please describe the data used more clearly.

   Response: I apologize for the confusion caused. I have revised the manuscript to clearly describe the data used in classification and feature importance analysis. Specifically, the reference to "20 sampling points" was incorrect; 20 actually refers to the number of visually interpreted sampling areas. Within these 20 sampling areas, we visually interpreted and obtained 3,680 treeline points.

 

6. Comment: Showing classification results on a map could be better for interpreting results.

   Response: I have added new subplots b and c to Figure 2 in the first part of the "Results" section to visually represent the classification results. These maps provide a clearer interpretation of the spatial distribution of treelines and enhance the overall comprehensibility of the findings.

 

7. Comment: In P.10 L.330-335, since the results obtained are generated from these parameters, the paragraph should be given at the beginning of the section.

   Response: I have moved the paragraph in P.10 L.330-335 to the beginning of the relevant section, ensuring that the results are introduced in a logical and coherent manner. This adjustment improves the flow of the manuscript and aligns the presentation of results with the methodology.

 

I have attached the revised manuscript for your review. I believe these changes address your concerns and significantly improve the overall quality of the study. Thank you again for your invaluable feedback.

 

Please let me know if there are any additional suggestions or if further clarification is needed.

 

Sincerely,

Chuan Lin

2024.7.10

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear Authors,

I have carefully reviewed your manuscript, titled “Spatial pattern and environmental driving factors of treeline elevations in Yulong Snow Mountain, China.” I believe that some improvements could enhance the scientific value and impact of your study. I have provided my specific evaluations and recommendations below.

1. The primary question addressed by the research is as follows: The objective of this study is to identify the spatial patterns and environmental driving factors of treeline elevations in Yulong Snow Mountain, China. The objective is to identify deviations between actual treeline elevations and thermal treeline predictions and to determine the key environmental variables influencing these deviations.

2. The utilisation of machine learning techniques (SVM and XGBoost) in treeline research represents an innovative approach. The focus on Yulong Snow Mountain, a biodiversity hotspot, adds regional specificity and relevance to the study. This study addresses a gap in understanding local environmental factors that cause deviations in treeline elevations, which is crucial for predicting ecosystem responses to climate change. Furthermore, it addresses the need for high-resolution, machine-learning-based treeline identification. 

3. Methodological Advancement: The application of advanced machine learning algorithms to identify and analyse treeline patterns represents a significant methodological advancement.

4. Specific Methodological Improvements:

Feature Selection Explanation: It is recommended that a more detailed explanation of the feature selection process and the rationale behind the choice of specific environmental variables be provided. For instance, the assertion that "We select environmental variables from four aspects: climate impact, soil matrix, disturbance activity, and terrain" requires further explanation.

The spatial and temporal resolutions of the data used must be clarified. It is necessary to clarify the implications of the spatial and temporal resolutions of the data used and how these might affect the results. For example, the spatial resolution of the data may impose limitations. Consequently, the values of environmental variables for treelines within the same grid remain consistent. However, this raises the question of the impact of this consistency on the results.

Model Validation: It is recommended that more robust validation techniques be employed for the machine learning models, such as external validation with independent datasets or comparison with other established models.

It is important to consider the role of controls and replicates in the study. It is imperative to implement controls to mitigate the influence of confounding variables. For instance, the assertion that "To ensure the accuracy of selected treeline points, we aimed to exclude human-made treelines as much as possible" could be substantiated by providing further details on the methods employed to control for human activities.

5. Consistency of Conclusions and Main Questions Addressed:

It is important to ensure that the conclusions drawn are consistent with the evidence and arguments presented. The conclusions reached are generally consistent with the evidence and arguments presented. Nevertheless, the relationship between specific findings and broader ecological implications could be more clearly delineated. For example, the assertion that "temperature and precipitation are the primary factors influencing the spatial distribution of D-treeline at local scales" could be expanded upon to discuss the wider ecological contexts in which these factors operate.

The study addresses the following main questions: The study addressed all of the main questions posed using SVM and XGBoost models, which were validated by comprehensive data analysis.

6. Appropriateness of References:

The paper would be enhanced by the inclusion of more recent studies on the application of machine learning in ecological research.

7. Additional Comments on Tables, Figures, and Data Quality:

The tables and figures are well-structured and effectively convey the study's findings. Nevertheless, certain enhancements could be implemented.

Clarity and Detail: It is recommended that all figures and tables be clearly labelled and that sufficient detail be provided for their interpretation. For instance, the statement "Figure 3 shows the average treeline obtained along the latitude and longitude directions..." requires further clarification.

A comparative analysis should be conducted. Include comparative figures that demonstrate the study's results in the context of findings from similar studies, thereby enhancing the relevance and context of the data.

The quality of the data must be ensured. It is of the utmost importance to ensure the quality of the data by providing more information on the data preprocessing steps, the handling of missing values, and the robustness of the data sources. For example, the statement that "high-resolution (HR) satellite images from the Google Earth platform were used" could be enhanced by providing further details on the impact of the resolution.

Author Response

Thank you for your detailed review and constructive feedback on our manuscript titled “Spatial pattern and environmental driving factors of treeline elevations in Yulong Snow Mountain, China.” We appreciate your time and effort in evaluating our work and providing insightful recommendations. Below, we provide our responses to each of your comments:

 

1. Primary Question Addressed by the Research:

   Response: Thank you for summarizing the primary question of our research accurately. We will ensure that the objective of identifying deviations between actual treeline elevations and thermal treeline predictions, along with the key environmental variables influencing these deviations, is clearly stated in the revised manuscript.

 

2. Utilization of Machine Learning Techniques:

   Response: We appreciate your recognition of the innovative approach of using SVM and XGBoost in our study. We will emphasize the regional specificity and relevance of our research in the revised introduction to highlight its significance in addressing local environmental factors affecting treeline elevations.

 

3. Methodological Advancement:

   Response: Thank you for acknowledging the methodological advancement in our study. We will further elaborate on the use of advanced machine learning algorithms to identify and analyze treeline patterns in the revised manuscript.

 

4. Specific Methodological Improvements:

   - Feature Selection Explanation:

    Response: We agree that a more detailed explanation of the feature selection process is needed. We will provide a comprehensive rationale for choosing specific environmental variables and further clarify our selection criteria in the Methods section.

   - Spatial and Temporal Resolutions:

    Response: We will clarify the spatial and temporal resolutions of the data used in our study and discuss their implications on the results. We will address the potential limitations imposed by the spatial resolution and how this consistency within the same grid may impact our findings.

   - Model Validation:

Response: We acknowledge the need for robust validation techniques. We will incorporate additional validation methods, such as external validation with independent datasets and comparisons with other established models, to strengthen our model validation process.

   - Controls and Replicates:

Response: I have revised and added the description of this section, please refer to Method 2.2.1 for details. We will provide further details on the methods employed to control for human activities and other confounding variables. This will include a clearer description of how we aimed to exclude human-made treelines and other potential biases in our data collection process.

 

5. Consistency of Conclusions and Main Questions Addressed:

   Response: We will ensure that our conclusions are consistent with the evidence and arguments presented in the manuscript. We will expand on the relationship between specific findings and broader ecological implications, particularly the roles of temperature and precipitation in influencing the spatial distribution of D-treeline at local scales. Detailed content can be found in the attachment.

 

6. Appropriateness of References:

   Response: We have revised the introduction section, please refer to the attachment, and added the latest research on the application of machine learning in ecological research.

 

7. Additional Comments on Tables, Figures, and Data Quality:

   - Clarity and Detail:

    Response: We will ensure that all figures and tables are clearly labeled and provide sufficient detail for interpretation. We will clarify the description of Figure 3 and other figures to enhance their comprehensibility.

   - Comparative Analysis:

    Response: We will include comparative figures that demonstrate our results in the context of findings from similar studies. This will enhance the relevance and context of our data.

   - Data Quality:

Response: We have added a more data processing process in the methods section, please refer to the attachment for details

 

We appreciate your valuable feedback and believe that these revisions will significantly improve the quality and impact of our manuscript. We have attached the revised manuscript for your review.

Thank you again for your insightful comments.

 

Sincerely,

Chuan Lin

2024.7.10

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript has been significantly modified taking into account the noted comments. In particular, significant changes have been made to the “Data and Methods” section. The set of independent variables used in the study was adjusted (Table 2). The manuscript may be recommended for publication subject to some corrections.

1) Figure 4. In the upper part of the figure, you need to correct “CLIMAT” to “CLIMATE”.

2) Figure 5. Right side. On the graphs (Figure 5b and Figure 5d), the full name of the horizontal axis need be given. For example, replace “hdf” with “Anthropogenic disturbance”.

Comments on the Quality of English Language

It is recommended to check the manuscript for minor inaccuracies.

Author Response

Thank you very much for your thorough review and valuable comments on our manuscript. We appreciate your feedback and suggestions for improvement. Below are our responses to your comments:

1. Figure 4 Correction: We have corrected "CLIMAT" to "CLIMATE" in the upper part of Figure 4.
2. Figure 5 Correction: We have provided the full name for the horizontal axis in Figures 5b and 5d. Specifically, "hdf" has been replaced with "Anthropogenic disturbance".
3. Regarding the quality of the English language, we have reviewed the manuscript again and corrected minor inaccuracies.

Thank you once again for your constructive feedback. We believe that the revisions have improved the quality of the manuscript. We look forward to your final evaluation and hope for a favorable recommendation for publication.

Best regards,

Chuan Lin

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The authors seem to have addressed my concerns and I believe the paper is generally in condition for publication.

 

Comments on the Quality of English Language

I would recommend that the authors check the English, as there are still some awkwardly constructed sentence in the text.

Author Response

Thank you for your positive feedback and for acknowledging the revisions we made to address your concerns. We are pleased to hear that you believe our paper is generally in condition for publication.

Regarding the quality of the English language, we appreciate your suggestion. We will carefully review the manuscript once more to correct any awkwardly constructed sentences and ensure the language is clear and precise.

Thank you again for your valuable comments and suggestions. We look forward to the final evaluation.

Best regards,

Chuan Lin

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you to the authors for their responses and corrections.

Author Response

Thank you for your positive feedback and for taking the time to review our manuscript. We appreciate your acknowledgement of our responses and corrections.

If there are any further suggestions or comments, please let us know. Otherwise, we look forward to the final evaluation and hope for a favorable recommendation for publication.

Best regards,

Chuan Lin

Author Response File: Author Response.docx