Land Use Transition Under a Tense Human–Land Relationship: A GWR Analysis of Conflicts Between Construction Land and Cropland

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The research topic selected for this study appears rather conventional. The phenomenon of construction land expansion and its encroachment on cultivated land in developing China has become an established fact widely recognized in academic circles. Over the past two to three decades, this issue has been extensively investigated and validated by numerous scholars through diverse perspectives and methodologies. Many existing studies have provided more profound examinations than the present paper regarding the impacts of construction land expansion on food security and regional ecological security. Compared to these comprehensive studies, the current research merely analyzes spatiotemporal variations through the conversion between two land types, which significantly limits its practical relevance and fails to directly address core issues.

 

Regarding the analysis of factors influencing land type conversion, the paper lacks sufficient justification in selecting influencing factors. It merely examines the individual effects of single factors on construction land expansion without considering the interactive mechanisms between different factors. This makes the factor analysis appear rather superficial and lacking in theoretical depth. As for the regional differences in the relationship between construction land and cultivated land mentioned in the discussion, the paper could have provided greater scientific value by focusing on how these differences should inform region-specific land policy formulation. Exploring policy directions and their rationales in depth would have been more valuable than simply presenting these regional differences.

 

For Figure 1 (Overview map of the study area), layout optimization is recommended, including adjusting the placement of the north arrow and scale bar. If multiple images are included, they should be labeled as (a), (b), etc., with corresponding explanations. The font size should be increased to improve readability. In the Driver factor section, the software and its version number used should be specified. Key model evaluation metrics for OLS and GWR, including R², adjusted R², and AICc values, should be provided.

 

It is suggested to separate the Conclusion and Discussion section into two distinct parts: 4.1 Discussion and 4.2 Conclusion. This structural adjustment would improve the paper's organization and readability. The current version combines these elements, which may reduce the clarity of the research findings and their implications. By separating them, the discussion can focus on interpreting results and comparing them with existing literature, while the conclusion can concisely summarize key findings and their significance.

Author Response

Comments 1: The research topic selected for this study appears rather conventional. The phenomenon of construction land expansion and its encroachment on cultivated land in developing China has become an established fact widely recognized in academic circles. Over the past two to three decades, this issue has been extensively investigated and validated by numerous scholars through diverse perspectives and methodologies. Many existing studies have provided more profound examinations than the present paper regarding the impacts of construction land expansion on food security and regional ecological security. Compared to these comprehensive studies, the current research merely analyzes spatiotemporal variations through the conversion between two land types, which significantly limits its practical relevance and fails to directly address core issues.

Response 1: We appreciate the reviewer’s concerns regarding the maturity of the research topic. We have revised the Introduction and Discussion sections to improve the manuscript’s originality and analytical clarity. In the Introduction, we now more clearly explain how the study contributes to the literature by introducing two complementary indicators, DEP and CON, which capture both the pressure from construction land expansion and the resulting loss of cropland. The revised Discussion provides a more in-depth interpretation of the spatial patterns, linking them with differentiated policy recommendations across regions. These improvements aim to strengthen the study’s practical relevance and clarify its value within the broader context of land-use research. 

Comments 2: Regarding the analysis of factors influencing land type conversion, the paper lacks sufficient justification in selecting influencing factors. It merely examines the individual effects of single factors on construction land expansion without considering the interactive mechanisms between different factors. This makes the factor analysis appear rather superficial and lacking in theoretical depth. As for the regional differences in the relationship between construction land and cultivated land mentioned in the discussion, the paper could have provided greater scientific value by focusing on how these differences should inform region-specific land policy formulation. Exploring policy directions and their rationales in depth would have been more valuable than simply presenting these regional differences.

Response 2:We appreciate the reviewer’s constructive feedback. First, we would like to clarify a potential misunderstanding: the geographically weighted regression (GWR) model used in this study is multivariate, not a series of separate univariate regressions. All explanatory variables were included in the model simultaneously, and the maps of spatially varying coefficients reflect the localized effect of each variable while controlling for the others. We have clarified this point in the methodology section to avoid confusion.

Second, to strengthen the theoretical depth of the analysis, we have added an explanation (lines 383–402) outlining the rationale behind our variable selection. These variables were chosen based on established land-use change literature and their relevance to both socioeconomic and ecological mechanisms influencing cropland loss and construction land expansion in China.

Third, in response to the suggestion to deepen the policy discussion, we have strengthened the interpretation of spatially differentiated results in the Discussion section (4.3. Suggestions for Differentiated Policies).  This revision provides a more substantive link between empirical findings and actionable policy insights.

Lastly, we acknowledge that GWR does not account for interactions between explanatory variables. While our primary objective was to explore spatial heterogeneity in main effects, we agree that future studies could benefit from models capable of capturing interaction effects and complex causal relationships, such as multiscale GWR or spatial structural equation modeling (SEM).

Comments 3: For Figure 1 (Overview map of the study area), layout optimization is recommended, including adjusting the placement of the north arrow and scale bar. If multiple images are included, they should be labeled as (a), (b), etc., with corresponding explanations. The font size should be increased to improve readability. In the Driver factor section, the software and its version number used should be specified. Key model evaluation metrics for OLS and GWR, including R², adjusted R², and AICc values, should be provided.

Response 3:Thank you for these detailed and constructive suggestions. In response, we have redrawn Figure 1 to optimize the layout. The positions of the north arrow and scale bar have been adjusted, and the font size has been increased to enhance readability. Where multiple maps are presented, subfigure labels (e.g., (a), (b)) and corresponding explanations have been added for clarity.

In addition, we have specified the software and version used in the analysis in line 206, and included key model evaluation metrics for both OLS and GWR models (R², adjusted R², and AICc values) in lines 430–438 of the revised manuscript. These additions aim to improve the transparency and replicability of the analysis.

Comments 4: It is suggested to separate the Conclusion and Discussion section into two distinct parts：4.1 Discussion and 4.2 Conclusion. This structural adjustment would improve the paper's organization and readability. The current version combines these elements, which may reduce the clarity of the research findings and their implications. By separating them, the discussion can focus on interpreting results and comparing them with existing literature, while the conclusion can concisely summarize key findings and their significance.

Response 4: Thank you for this valuable suggestion. In the revised manuscript, we have restructured the paper to clearly separate the Discussion and Conclusion into two independent sections—Section 4 (Discussion) and Section 5 (Conclusion)—as recommended. This restructuring improves the manuscript’s clarity by enabling the discussion to focus on interpreting the findings and spatial mechanisms, while the conclusion concisely summarizes key results and implications.

Additionally, to improve the manuscript’s completeness and transparency, we have added a new Section 6 (Limitations and Future Directions). This section discusses data-related uncertainties, methodological boundaries, and proposes future research directions including the integration of national spatial planning documents. We hope this further strengthens the manuscript’s overall contribution.

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors, I consider the topic to be highly relevant and timely. Overall, the manuscript is logically structured. However, I have several major and specific comments regarding its current version.

Major Comments

The manuscript's most significant weakness is its insufficient theoretical grounding and the near-complete absence of a robust discussion section. The introduction lacks a thorough theoretical background on key concepts such as land use change, soil sealing, and the context of food security and environmental issues. The authors need to more clearly define the research gaps they aim to address and more precisely position their work in relation to the existing literature.

This is related to the absence of a real discussion section in the end of paper that would contextualize the findings within the current scientific literature. The text is thus largely limited to a description of the results and cartograms, without offering a deeper interpretation or explanation of the causal mechanisms behind the observed phenomena. This gives the impression that the authors attempted to cover an overly broad topic, resulting in analyses that are somewhat superficial, whereas the subject deserves a more in-depth treatment.

Furthermore, the paper lacks a critical assessment of the limitations arising from the data and methods used, which is crucial for a scientific paper. Similarly, recommendations for future research and the specific policy recommendations promised in the abstract are only vaguely and generally mentioned in the conclusion.

Comments on individual parts of the paper

1. Abstract: The abstract lacks essential information about the data and methods used. The claim regarding the proposal of "differentiated policies" is not sufficiently developed or substantiated in the main body of the text.
2. Introduction: I recommend significantly expanding this section with a theoretical review. The note on China as the "most populous developing country" should be verified and, if necessary, updated.
3. Methods:
• The abbreviation ACO is not defined in the text (line 127 -131), which impedes a full understanding of the methodology.
• The selection of socioeconomic indicators, while supported by citations, lacks a clear justification for why these specific indicators were chosen and what their presumed influence on the studied phenomena is.
4. Results:
• The presentation of the results is largely descriptive, confined to stating the trends shown in the maps. A deeper analysis and explanation of the causes of the observed phenomena are missing (e.g., why urbanization has opposing effects on DEP and CON).
• Some findings are require detailed explanation. For instance, the reported decrease in built-up land in some provinces (with a minimum of -17.28%) is a highly unusual trend that must be carefully explained.
• The authors should consider whether their results might be influenced by the initial share of cropland in each province. It is logical that regions with a high proportion of cropland will also exhibit a higher rate of its conversion to built-up areas.
5. Discussion and Implications:
• As mentioned previously, the paper effectively lacks a discussion section. The authors must compare their findings with the results of other studies.
• It is necessary to comment on legislative and strategic materials that frame the topic under study (e.g. purchasing cropland quotas from underdeveloped areas – line 417). What other factors are behind the differences between cities?
6. Maps and Graphics: The article contains a large number of cartograms, but their quality and resolution are often low, hindering their readability. I recommend significantly reducing their number in the main text and focusing on the most essential ones. The remainder should be considered for inclusion in supplementary materials.

Conclusion

In summary, the manuscript has the potential to become a valuable contribution to the field. However, in its current form, it requires major revision. The authors should focus primarily on strengthening the theoretical framework, adding a robust discussion, critically evaluating their results and methodological limitations.

Author Response

Comments 1：

Abstract：The abstract lacks essential information about the data and methods used. The claim regarding the proposal of "differentiated policies" is not sufficiently developed or substantiated in the main body of the text.

Response 1：Thank you for this important observation. We have revised the abstract to clearly include key information about the data sources, methodological framework (including the use of DEP/CON indicators and geographically weighted regression), and the main findings. Additionally, we have strengthened the presentation of differentiated policy recommendations in the main text, particularly in Section 4.3, by providing region-specific strategies grounded in the spatial patterns and regression results. We believe these revisions improve both the clarity and the coherence between the abstract and the main content.

Comments 2：

Introduction：I recommend significantly expanding this section with a theoretical review. The note on China as the "most populous developing country" should be verified and, if necessary, updated.

Response 2：Thank you for this helpful recommendation. In the revised manuscript, we have rewritten the Introduction to provide a more substantial theoretical foundation. This includes a strengthened review of relevant literature on land-use transition, cropland protection, and the spatial dynamics of urban expansion, which situates our study more clearly within existing research. We have also revised the statement regarding China’s population status.

Comments 3：

Methods:

The abbreviation ACO is not defined in the text (line 127 -131), which impedes a full understanding of the methodology.

The selection of socioeconomic indicators, while supported by citations, lacks a clear justification for why these specific indicators were chosen and what their presumed influence on the studied phenomena is.

Response 3：Thank you for pointing out these important issues. To ensure clarity, we have added a statement in line 168 indicating that all abbreviations, including ACO, are defined in the Abbreviations section at the end of the manuscript. This should help readers unfamiliar with specific terms better navigate the text.

Additionally, to address the second point, we have expanded the explanation of the rationale behind the selection of socioeconomic and environmental indicators in lines 384–403. This section now outlines both the theoretical relevance and the expected influence of each factor on construction land expansion and cropland loss, based on established literature and policy context. We believe these changes enhance both the methodological transparency and the conceptual clarity of the manuscript.

Comments 4：

Results:

The presentation of the results is largely descriptive, confined to stating the trends shown in the maps. A deeper analysis and explanation of the causes of the observed phenomena are missing (e.g., why urbanization has opposing effects on DEP and CON).

Some findings are require detailed explanation. For instance, the reported decrease in built-up land in some provinces (with a minimum of -17.28%) is a highly unusual trend that must be carefully explained.

The authors should consider whether their results might be influenced by the initial share of cropland in each province. It is logical that regions with a high proportion of cropland will also exhibit a higher rate of its conversion to built-up areas.

Response 4：We thank the reviewer for these insightful observations. In response:

(1) Deeper explanation of observed patterns: We have substantially revised the Discussion (Section 4) and Conclusion (Section 5) to go beyond trend description and to provide deeper interpretation of the observed spatial patterns. In particular, we have provided a more detailed explanation of the seemingly contradictory role of urbanization in Section 4.2, specifically in lines 516–525.

(2) Unusual decrease in built-up land: We carefully reviewed the data related to provinces that show a negative change in construction land area (e.g., -17.28%). We suspect this phenomenon is linked to specific local events, such as large-scale flooding or administrative boundary adjustments, particularly in provinces with very small shares of construction land to begin with. We have acknowledged and discussed this in the Limitations and Future Directions (Section 6) to clarify its potential impact on the results.

(3) Initial cropland share as a variable: We attempted to incorporate the initial share of cropland as a control variable to test its influence. However, we observed high multicollinearity between this variable and average slope. When we replaced slope with the initial cropland share, the regression results were similar in both spatial distribution and magnitude of coefficients. This suggests that the spatial structure captured by slope may also reflect underlying patterns in cropland distribution. We have briefly discussed this in the revised methodology and limitation sections to clarify our modeling choices.

Comments 5：

Discussion and Implications:

As mentioned previously, the paper effectively lacks a discussion section. The authors must compare their findings with the results of other studies.

It is necessary to comment on legislative and strategic materials that frame the topic under study (e.g. purchasing cropland quotas from underdeveloped areas – line 417). What other factors are behind the differences between cities?

Response 5：Thank you for your valuable comment. In response, we have comprehensively redrafted the Discussion section (Section 4) to provide deeper analytical insight and explicitly compare our findings with those of relevant studies. This includes clarifying the dual role of urbanization in driving opposing trends in DEP and CON, and interpreting these dynamics in the context of China’s differentiated development strategies. In addition, we discuss the spatial mechanisms behind city-level differences by introducing explanatory factors such as administrative level, economic structure, and land-use autonomy.

Comments 6：

Maps and Graphics：

The article contains a large number of cartograms, but their quality and resolution are often low, hindering their readability. I recommend significantly reducing their number in the main text and focusing on the most essential ones. The remainder should be considered for inclusion in supplementary materials.

Response 6：Thank you for this helpful suggestion. In response, we have regenerated all maps and graphics using higher resolution and improved layout design to ensure visual clarity and readability. We carefully reviewed the figures and retained only the most essential ones in the main text to support the core findings.

Reviewer 3 Report

Comments and Suggestions for Authors

Overall, this is a very good article which has explored the topic of land-use conversion in different regions across Mainland China. I command the authors for the very detailed methodology. 

Here are a few minor edits which I believe are necessary for the article's development.

1) Please separate the conclusion from the discussion. The discussion should be a single section, addressing the strengths and weaknesses of the method. The conclusion should encompass the entirety of the article and state, in a concise manner, what was developed, what the findings were and to propose ways forward.

2) In the discussions "subsection" I haven't seen much of a discussion regarding the weaknesses of the process. I believe those should be better incorporated into the article.

3) Also, the authors need to address the fact that such regions differ in geographical area. How does this affect the regressions and indices used? I believe it doesn't affect much, given the methodology presented, but this might be a question other readers may have.

4) The article uses a plethora of socioeconomic data. Are they available throughout the last decades? What if they were to study similar events prior to 2000? I say that because China's impulse into industrialization happened a few decades before, and certainly the trends might show a few interesting results. Will the data be available or will the regression need to use less variables?

5) Finally, please make sure all maps are sufficiently large to be readable. Perhaps taking an entire page per figure would help enlarge the maps of China a bit more and improve readability.

Kindly make those adjustments and resubmit the article for a second evaluation. Best regards.

Author Response

Comments 1：

Please separate the conclusion from the discussion. The discussion should be a single section, addressing the strengths and weaknesses of the method. The conclusion should encompass the entirety of the article and state, in a concise manner, what was developed, what the findings were and to propose ways forward.

Response 1：Thank you for this valuable structural recommendation. In the revised manuscript, we have fully separated the Discussion and Conclusion into two distinct sections—Section 4 (Discussion) and Section 5 (Conclusion)—to improve the clarity and organization of the paper.

In addition, we have rewritten the Introduction to clarify the study’s theoretical grounding and better position its contributions within the broader literature.

Comments 2：

In the discussions "subsection" I haven't seen much of a discussion regarding the weaknesses of the process. I believe those should be better incorporated into the article.

Response 2：Thank you for this important observation. In response, we have added a dedicated Section 6 (Limitations and Future Directions) to explicitly address the methodological and data-related limitations of the study. This includes discussion of issues such as abnormal land-use fluctuations caused by flood events, the absence of pre-2000 data, and the lack of interaction terms in the GWR model. We also acknowledge the potential influence of initial land-use composition and the limited integration of national spatial planning documents, which we highlight as areas for future research. We believe this addition provides a more balanced and transparent account of the study’s methodological scope.

Comments 3：

Also, the authors need to address the fact that such regions differ in geographical area. How does this affect the regressions and indices used? I believe it doesn't affect much, given the methodology presented, but this might be a question other readers may have.

Response 3：Thank you for this thoughtful observation. To clarify this point for readers, we have added a statement in lines 206–210 of the revised manuscript. Specifically, we note that the geographical size of the study units does not affect the regression results or the calculation of the DEP and CON indicators, since both are area-normalized metrics. Furthermore, the GWR model estimates coefficients based on local spatial relationships rather than absolute area. We hope this clarification helps address any potential confusion regarding the influence of spatial scale.

Comments 4：

The article uses a plethora of socioeconomic data. Are they available throughout the last decades? What if they were to study similar events prior to 2000? I say that because China's impulse into industrialization happened a few decades before, and certainly the trends might show a few interesting results. Will the data be available or will the regression need to use less variables?

Response 4：Thank you for this insightful comment. We agree that examining land-use change prior to 2000, particularly during China’s early stages of industrialization, could reveal valuable historical trends. However, many of the socioeconomic indicators used in this study are not consistently available at the city or county level before 2000. As a result, extending the temporal scope backward would require either reducing the number of explanatory variables or relying on more aggregated and less detailed data.

We have acknowledged this limitation in Section 6 (Limitations and Future Directions) and noted that future research could consider alternative historical datasets or supplement quantitative analyses with qualitative or policy-focused approaches when long-term data are unavailable.

Comments 5：

Finally, please make sure all maps are sufficiently large to be readable. Perhaps taking an entire page per figure would help enlarge the maps of China a bit more and improve readability.

Response 5：Thank you for the helpful suggestion. In response, we have redrafted all maps using higher-resolution layouts and adjusted their dimensions to improve clarity and readability. We hope these updates ensure that spatial information is more accessible and visually interpretable to readers.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Firstly, the topic selection of the article is conventional and lacks innovation. Secondly, the research in the article is overly simplistic and lacks sufficient workload. Finally, the revision of the article did not address core issues such as the value and innovation of the topic selection.

Comments on the Quality of English Language

The English proficiency of the article can be improved.

Author Response

Comment 1:

Firstly, the topic selection of the article is conventional and lacks innovation. Secondly, the research in the article is overly simplistic and lacks sufficient workload. Finally, the revision of the article did not address core issues such as the value and innovation of the topic selection.

Response 1: 

We sincerely thank the reviewer for this important comment. In the revised manuscript, we have made a conscious effort to clarify the novelty and significance of our study. Specifically:

In the last paragraph of the Introduction, we now clearly explain the theoretical and methodological contributions of our study, including the introduction of the DEP and CON indicators and the use of a multiscale spatial analysis framework that links provincial and city-level dynamics.

In Section 4.3 (Discussion), we added a final paragraph that interprets how our findings enhance the understanding of regionally differentiated land-use conflicts and support more adaptive land governance.

In the Conclusion, we added a closing paragraph that summarizes how our approach contributes to both academic research and practical cropland protection strategies, particularly under rapid urbanization.

We hope these additions more clearly demonstrate the originality, theoretical relevance, and policy value of our work.

 

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors, thank you for the changes you have made. The text is now ready for publication.

 

Author Response

Comment 1:

Dear authors, thank you for the changes you have made. The text is now ready for publication.

Response 1: 

We sincerely thank you for your positive and encouraging feedback. We are grateful for your time and constructive comments, which greatly helped improve the quality of the manuscript.

As suggested, we have reduced the number of figures and tables. Specifically, Table 2, Table 3, Figure 3, Figure 6, Figure 9, Figure 10, and Figure 12 have been moved to the Supplementary Materials. We have also reordered the remaining figures and tables to ensure clarity and consistency in the main text.