Research on Resilience Evaluation and Prediction of Urban Ecosystems in Plateau and Mountainous Area: Case Study of Kunming City

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for The subject of this research is highly significant, and the choice of Kunming as a representative case study, combined with the use of multiple methods, offers a valuable perspective for understanding urban sustainability in unique geographical settings.

However, to make this good paper even better, we have a few suggestions for your consideration:

1. After the detailed analysis of individual indicators within each factor layer in Section 3.2, please more explicitly link the changes in the individual indicators to the final comprehensive index values and low resilience scores in Section 3.3. 
2. The paper mentions that Appendix 1 contains the questionnaire indicators, but it was not included in the manuscript file. Please remember to attach it.
3. In Lines 360 and 455, the description mentions using "hyperbolic functions and encoded all indicators under the R language environment" for standardization. This description is somewhat brief. We suggest briefly stating which type of standardization method was used, to make the methods more transparent.

Author Response

Dear reviewer,

We sincerely appreciate your constructive comments on this study. The revised manuscript has been uploaded to the manuscript revision system, along with the attached questionnaire. Below, we provide detailed responses to each of your comments.

Comments 1: After the detailed analysis of individual indicators within each factor layer in Section 3.2, please more explicitly link the changes in the individual indicators to the final comprehensive index values and low resilience scores in Section 3.3.

Response 1: We sincerely thank you for your valuable suggestions, which have significantly improved the overall quality of our manuscript. In response, we have added a relevant analysis of the relationship between changes in individual indicators and the comprehensive index value in Section 3.3 (line 481-489). This addition highlights that the urban ecosystem resilience in the study area has not achieved a coordinated state across the five dimensions of “Driving force–Pressure–State–Impact–Response” (DPSIR). The results reflect a pattern of rapid urban development accompanied by growing tension between human activities and nature. On one hand, the region has maintained strong economic growth; on the other hand, it faces frequent natural disasters, declining ecological self-recovery capacity, and insufficient implementation of environmental response measures. These findings underscore the urgent need to develop strategies aimed at enhancing urban ecosystem resilience to ensure the study area’s sustainable development in the future.

Comments 2: The paper mentions that Appendix 1 contains the questionnaire indicators, but it was not included in the manuscript file. Please remember to attach it.

Response 2: Thank you for your suggestion. We apologize for our oversight and will upload the questionnaire to the revised manuscript submission system accordingly.

Comments 3: In Lines 360 and 455, the description mentions using "hyperbolic functions and encoded all indicators under the R language environment" for standardization. This description is somewhat brief. We suggest briefly stating which type of standardization method was used, to make the methods more transparent.

Response 3: Thank you for your valuable suggestion. We have supplemented Section 2.4, which discusses the all-permutation polygon indicator method, by adding the calculation process and underlying principle of the hyperbolic normalization function. According to Equations (1) to (3) (line 331-348), the normalization principle can be implemented by applying the hyperbolic normalization function to standardize all resilience indicators using R programming.

Sincerely,

Hui Li

Reviewer 2 Report

Comments and Suggestions for Authors

The article is thorough and covers various analytical methods. Its strength also lies in the fact that these methods are applied to the example of a specific city.

However, it is not clear which method provides better results or whether the outcomes obtained through different methods are coherent across the various factors. Moreover, the results of the analysis are rather trivial, such as “xxx” and “xxx.” Such conclusions could be drawn using common sense alone, and there seems to be no compelling reason to apply complex mathematical methods for this purpose.

Therefore, I would recommend significantly shifting the focus of the study—namely, to demonstrate whether and how such models could enable the prediction of urban development, and what potential benefits could be achieved through their application.

 

Comments on the Quality of English Language

There are occasional grammatical issues and overly long sentences. A final round of language polishing would enhance the manuscript’s readability

Author Response

Dear Reviewer,

We sincerely appreciate your careful review and valuable suggestions regarding this study. In response to your comments, we have carefully considered and organized our replies as follows:

(1) On the differences and comparative advantages between methods
This study employs the DPSIR model as the core framework with the aim of systematically revealing the driving mechanisms behind urban ecosystem resilience formation and conducting a preliminary assessment. The all-permutation polygon diagram method is used as a visual representation of the urban ecosystem resilience evaluation results, primarily to intuitively display the temporal trends of various indicators and the system’s overall performance. Its advantage lies in eliminating dimensional differences and providing dynamic comparative analysis. In contrast, the grey system prediction model is applied to identify future development trends of urban ecosystem resilience in Kunming and to meet prediction needs under conditions of incomplete data. Each method targets different research objectives, serving the “current status assessment” and “future prediction” aspects respectively, and thus they are complementary rather than substitutes. Therefore, the relationship is not a simple “choice of the better method” but rather mutual supplementation.

(2) On the issue of some conclusions being common knowledge
We understand the reviewer’s concern regarding some conclusions appearing “obvious.” In our revision, we have strengthened quantitative support and causal logic as much as possible (for example, by removing controversial factors such as “consumer price growth rate” in the second conclusion and focusing on variables directly affecting the ecosystem, such as land-use change, energy consumption, and disasters. See line 629-631). Additionally, we have validated common-sense understandings through systematic quantification, aiming to enhance the scientific credibility and decision-making value of these findings.

(3) On the suggestion to emphasize research focus
We greatly appreciate your suggestion to further highlight “whether and how the model predicts urban development, as well as the model’s practical application value.” In Section 4 of the revised manuscript, we have included a theoretical discussion of model applicability and have strengthened the explanation of how prediction results can guide policy formulation (see Section 4.2, line 597-606). Future research will further expand this direction with the goal of deepening the model’s application scenarios in urban planning and governance.

Once again, thank you for your constructive criticism and valuable advice, which have greatly contributed to improving the quality of our research.

Sincerely,
Hui Li

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

In my opinion the Manuscript can not be published in such form in any journal. All sections of the MS require after major revision. The Language must be improved significantly. There are a lot of punctuation mistakes, and construction of most sentences through the text is not specific for English Language. The sentences are too long and unreadable, especially in conclusion. So, the poor English makes your work unclear. The aim of the work is unclear. The Introduction is too long. The sentences in Conclusion are too gramatically complicated and unreadable. All parameters and formulas should be explained. The Figure 2 is under great doubt. According this figure the wood area has increased significantly for 10 years. Is it really possible? Table 1 is more informative than the text and shoud be placed in the Initial Data Section. 

With best wishes,

Reviewer

 

 

Comments on the Quality of English Language

The Language must be improved significantly. There are a lot of punctuation mistakes, and construction of most sentences through the text is not specific for English Language. The sentences are too long and unreadable, especially in conclusion. So, the poor English makes your work unclear.

Author Response

Dear Reviewer,

Thank you for your detailed review and insightful comments on our manuscript. We greatly value the issues you raised regarding language expression, research structure, and figures, and have made systematic revisions and improvements throughout the manuscript accordingly. Our point-by-point responses are as follows:

Comments 1: On language expression and punctuation issues
Response 1: We fully accept your criticism regarding the language quality. We have conducted a sentence-by-sentence proofreading of the entire manuscript, correcting punctuation errors, simplifying lengthy sentences, and improving overall clarity and readability. In particular, the Introduction and Conclusion sections have been restructured to ensure clear and academically appropriate expression.

Comments 2: On the unclear structure of the introduction and research objectives
Response 2: We have reorganized the content of the Introduction to remove repetitive background information and enhance the focus on the core research problem. At the end of the Introduction (line 177-185), we have explicitly stated the research objectives: “Based on the DPSIR framework, this study aims to construct a resilience assessment indicator system for plateau and mountainous urban ecosystems to evaluate their resilience status, understand the overall health of the regional ecosystem, identify potential urban ecological crises, reveal the interactions between socioeconomic activities and the ecological environment in such cities, and predict future resilience trends. Ultimately, the goal is to provide strategic recommendations for ecosystem protection in similar urban contexts.”

Comments 3: On the complex and difficult-to-understand grammar in the conclusion
Response 3: We have thoroughly rewritten the Conclusion section, simplifying sentence structures, clarifying subject-verb relationships, and highlighting the key findings and policy implications to ensure that readers can clearly grasp the core significance and practical value of our research results (see line 607 onwards).

Comments 4: On the concern regarding forest area change in Figure 2
Response 4: Thank you for raising this important point. We have added further explanation in the manuscript (see line 206-212). Due to space limitations, we did not originally include the interpretation process, which we now explain as follows:
We used Landsat 5 imagery (for 2006 and 2011) and Landsat 8 imagery (for 2016), all with 30-meter spatial resolution and cloud cover below 10%. Supervised classification was conducted using ENVI remote sensing software, and classification accuracy was verified using high-resolution Google Earth imagery and field surveys. Accuracy was assessed using the Majority analysis method, with classification accuracy exceeding 90% and Kappa coefficients above 0.7, meeting standard requirements.
The observed increase in forest area is primarily due to: (1) the transformation or degradation of grassland and cultivated land at the urban periphery into forest land, and (2) the afforestation of previously unused land through greening projects in suburban areas. These factors contributed to the increase in forest area compared to 2006.

Comments 5: On the suggestion regarding the position and purpose of Table 1
Response 5: We understand and accept your suggestion. Since Table 1 presents the standardized calculation results for each indicator—and given that the indicators for ecosystem resilience have already been described in detail in Section 2.3—we decided to place the table at the end of that section. Its primary purpose is to present the computed values of the DPSIR model indicators after standardization, thereby helping readers gain a comprehensive understanding of the indicator system and data structure.

Once again, we sincerely thank you for taking the time to review our manuscript and for your rigorous and constructive feedback. We look forward to your continued guidance.

Sincerely,
Hui Li

Reviewer 4 Report

Comments and Suggestions for Authors

Dear Authors,

 

The study analyses the changes in the ecological situation in Kunming city and the prospects for the conservation and development of urban ecosystems in the mountainous region.

Urban ecosystems are urban landscapes located in residential areas that have replaced native biogeocenoses. Residential areas, where human economic activity is a significant factor of influence, determine a narrow range of possibilities for the existence of urban ecosystems, depending on the support provided by humans. In fact, the state of urban ecosystems can be characterised as a climax, depending mainly on anthropogenic impacts. All other factors, both biotic and abiotic, are secondary in this case. Therefore, the authors' choice of five factors of anthropogenic impact on urban ecosystems, viz: 'five factors of the ecosystem, including driver, pressure, state, impact and response', seem perfectly acceptable. The paper is quite interesting for the development of methods for the conservation of urban landscapes in Kunming city, but some places in the text raise questions.

 

1. The conclusions do not quite fit in with the aim of the paper, which is described in the title as: 'Research on resilience assessment and prediction of urban ecosystems ...'. The first conclusion is: 'The DPSIR model framework can be integrated through the engagement of stakeholders across the city...', i.e. the outcome of the study was the applicability of the model, not 'assessment and prediction of urban ecosystems'. This conclusion should be restructured and rewritten.

 

2. In the second conclusion, attention is rightly drawn to the fact that the resilience of the urban ecosystem depends directly on anthropogenic impacts, but the factors mentioned are very strange: 'The resilience level of the urban ecosystem in the study area is significantly af-602 fected by urban development 'driver' factors such as the level of urbanisation, the natural growth rate of the urban population and the growth rate of consumer prices'. It may be possible to analyse each factor of economic activity in relation to the 604 urban landscape, but it is still necessary to select the key factors that have a direct impact. How does the factor 'growth rate of consumer prices' affect the state of the ecosystem? You are not talking about the ecosystem, so the factors that should be considered are those that directly affect the ecosystem. Please read carefully and reformulate the 2nd conclusion.

 

3. Figure 2 is not very informative. It would be helpful to indicate what the reader should be looking for in these figures, or to explain these changes in the figure legend, so that trends in environmental change in the area can be seen.

 

4. Lines 535-537 recommend: ‘we should undertake ecological restoration to enhance the capacity to resist natural disasters and improve the self-healing of the environment, thereby enhancing the self-healing ability of the environment, reduc-536 ing the negative impact of natural disasters, and improving the UER’. What kind of self-healing are you talking about in relation to the urban ecosystem?! It is immeasurably distant from natural ecosystems, and its recovery after a crisis may follow a completely different scenario that bears little resemblance to the previous one. Only human intervention can restore the urban ecosystem to its former form, and there is little hope for self-recovery!

 

5. On line 553 you will find the following recommendation: 'It is necessary to rationally delineate ecological red lines, increase the proportion of nature reserves in the study area, maintain the balance of the ecosystem, and comprehensively improve the protection intensity and management level of protected areas'. Can you tell me which nature reserves you are talking about in relation to the urban area of the high mountain city of Kunming? To what extent and where is it possible to respond under the reserve from urban areas? If you do not explain this thesis, it will 'hang in the air' and have little meaning in your work.

 

I think you should reflect on the findings in your paper and show the importance and promise of situational modelling research to address the environmental challenges facing the Kunming urban region. It is important to show how your research is able to focus public attention on specific measures to create conservation of urban ecosystems, with the understanding that their condition is directly dependent on anthropogenic factors rather than environmental factors. With a slight revision of the above comments, the paper can be recommended for publication.

 

Author Response

Dear Reviewer,

Thank you very much for your meticulous review and constructive feedback on our research. Your suggestions have provided important guidance for improving the overall quality of the manuscript. Based on your specific comments, we have carefully reflected on and addressed the identified issues, and we now respond to each point as follows:

Comments 1: The conclusions do not quite fit in with the aim of the paper, which is described in the title as: 'Research on resilience assessment and prediction of urban ecosystems ...'. The first conclusion is: 'The DPSIR model framework can be integrated through the engagement of stakeholders across the city...', i.e. the outcome of the study was the applicability of the model, not 'assessment and prediction of urban ecosystems'. This conclusion should be restructured and rewritten.

Response 1: Thank you for pointing this out. As per your suggestion, we have rewritten the first conclusion to better summarize and align with the evaluation and prediction results, while de-emphasizing the general applicability of the DPSIR model. The revised text (line 608-628) is as follows:

The UER assessment system constructed based on the DPSIR model enables a comprehensive evaluation and quantification of the interactions among “Driving forces,” “Pressures,” “State,” “Impact,” and “Response” in highland mountainous cities. Using Kunming as a case study, the assessment results show that from 2006 to 2016, the resilience of the urban ecosystem remained consistently low, with composite index values all below 0.25. This indicates that the expansion of construction land, the reduction of cultivated land, and increased landscape fragmentation driven by urbanization were the dominant “pressure” factors contributing to resilience decline. Additionally, a high proportion of disaster-affected areas and elevated energy consumption per unit of GDP further weakened the ecosystem’s self-regulating capacity. To enhance resilience, it is essential to regulate “driving forces” at the source, mitigate “pressures,” improve the ecological “state,” and strengthen recovery pathways through targeted “response” policies. Given that highland mountainous urban ecosystems are complex mosaics composed of diverse landscape types, the evaluation indicators within the DPSIR framework reflect different functional roles: “driving force” and “pressure” indicators represent active underlying causes and direct influences; “state” indicators refer to the dynamic stability and resilience condition of the urban ecosystem; “impact” indicators reflect the system’s condition when facing external stress; and “response” indicators capture the policy and management actions taken to enhance the system’s resistance capacity. This framework highlights that urban ecosystems, over the course of long-term development, possess the ability to self-organize, adapt, learn, and innovate, exhibiting nested, multi-scalar resilience cycles.

Comments 2: In the second conclusion, attention is rightly drawn to the fact that the resilience of the urban ecosystem depends directly on anthropogenic impacts, but the factors mentioned are very strange: 'The resilience level of the urban ecosystem in the study area is significantly af-602 fected by urban development 'driver' factors such as the level of urbanisation, the natural growth rate of the urban population and the growth rate of consumer prices'. It may be possible to analyse each factor of economic activity in relation to the 604 urban landscape, but it is still necessary to select the key factors that have a direct impact. How does the factor 'growth rate of consumer prices' affect the state of the ecosystem? You are not talking about the ecosystem, so the factors that should be considered are those that directly affect the ecosystem. Please read carefully and reformulate the 2nd conclusion.

Response 2: We fully understand and accept your recommendation. We have refined the logical pathway between 'driving force' and 'pressure' indicators, removing indirect metrics such as the consumer price index growth rate, and emphasizing key disturbance variables such as urbanization, land use change, energy intensity, and natural disasters. The revised second conclusion now reads (line 629-652):
The resilience level of urban ecosystems in highland mountainous cities is significantly influenced by urban development “driving forces,” such as the level of urbanization, the natural population growth rate, and the growth rate of the consumer price index. The expansion of urban construction land and the continuous decline of arable land indirectly challenge the ecological carrying capacity. Changes in “pressure” factors—such as energy consumption per unit of GDP and the proportion of areas affected by natural disasters—further weaken the carrying and recovery capacities of urban ecosystems, serving as direct causes of resilience degradation. Landscape pattern metrics such as the Shannon Diversity Index, aggregation index, and landscape shape index reflect the diversity and fragmentation of ecological patches, thus representing the “state” of resilience at the landscape level. Socio-economic “impact” indicators—including per capita GDP, economic density, residents’ per capita disposable income, and per capita public green space—offer varying degrees of support to ecological services and can influence the direction of ecosystem resilience. Meanwhile, indicators such as the proportion of protected area coverage, the efficiency of industrial solid waste utilization, and the area of artificial afforestation reflect “response” measures under human intervention, including policies and regulations that affect the future resilience of urban ecosystems in highland mountainous areas. Particularly under external pressures, these responses help improve recovery capacity and maintain overall ecosystem resilience. Under the combined influence of these factors, the composite resilience index values of the study area in 2006, 2011, and 2016 were 0.1796, 0.2432, and 0.2186, respectively, all indicating a low-resilience state. Therefore, restoring urban ecosystem resilience in highland mountainous cities requires a comprehensive strategy that includes regulating “driving forces,” reducing “pressures,” improving the “state,” exerting constructive “impacts,” and enhancing “response” measures.

Comments 3: Figure 2 is not very informative. It would be helpful to indicate what the reader should be looking for in these figures, or to explain these changes in the figure legend, so that trends in environmental change in the area can be seen.

Response 3: Thank you for the suggestion. Figure 2 presents the results of remote sensing classification and serves as a key source for the 'state' indicators in our study. We have added clarification on the function of Figure 2 in the data source section (line 205-211). Furthermore, we have elaborated on how landscape pattern indices used as state indicators were derived using FRAGSTATS software, and have included relevant definitions and formulas in Section 2.3 (Line 280-291).

Comments 4: Lines 535-537 recommend: ‘we should undertake ecological restoration to enhance the capacity to resist natural disasters and improve the self-healing of the environment, thereby enhancing the self-healing ability of the environment, reduc-536 ing the negative impact of natural disasters, and improving the UER’. What kind of self-healing are you talking about in relation to the urban ecosystem?! It is immeasurably distant from natural ecosystems, and its recovery after a crisis may follow a completely different scenario that bears little resemblance to the previous one. Only human intervention can restore the urban ecosystem to its former form, and there is little hope for self-recovery!

Response 4: We fully accept your criticism and have removed the original expressions related to 'self-repair'. Instead, we now refer to enhancing the buffering and response capacity of the ecosystem through ecological engineering interventions. The revised sentence (line 563-567) reads:
Additionally, ecological engineering restoration measures—such as wetland rehabilitation and vegetation restoration on slopes—can enhance the disaster resilience and ecological buffering capacity of urban ecosystems. These efforts ultimately improve the urban ecosystem's capacity to respond to and recover from external disturbances.

Comments 5: On line 553 you will find the following recommendation: 'It is necessary to rationally delineate ecological red lines, increase the proportion of nature reserves in the study area, maintain the balance of the ecosystem, and comprehensively improve the protection intensity and management level of protected areas'. Can you tell me which nature reserves you are talking about in relation to the urban area of the high mountain city of Kunming? To what extent and where is it possible to respond under the reserve from urban areas? If you do not explain this thesis, it will 'hang in the air' and have little meaning in your work.

Response 5: We sincerely appreciate your reminder. We have added spatial information regarding the overlap and adjacency between Kunming’s urban fringe and key nature reserves such as the Dianchi Lake watershed and Jiaozi Snow Mountain. Additionally, we elaborated on the ecological co-functioning mechanisms between the urban ecosystem and protected areas. The revised content (line 591-582) reads:
The urban development edge of Kunming city overlaps and borders to varying degrees with ecological functional areas such as the Dianchi Lake Basin Protection Zone and the Jiaozixueshan Nature Reserve. These nature reserves serve not only as critical barriers for regional ecological security but also directly influence key ecosystem services, including urban water supply, biodiversity maintenance, and air quality regulation. By delineating and connecting ecological redlines more effectively, increasing the proportion of nature reserves within the study area, and strengthening the ecological connectivity between urban green infrastructure and protected areas, spatial integration and functional complementarity between the city and nature reserves can be promoted. This integration enhances the overall ecosystem’s responsiveness and adaptive capacity. 

Once again, we sincerely thank you for your insightful and comprehensive comments. Your suggestions have significantly deepened the theoretical and practical dimensions of our manuscript. We have thoroughly revised the content in response to your feedback and look forward to your continued guidance.

Sincerely,
Hui Li

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

I am suprised by so quick correction of your manuscript. It was improved significantly but some figures need to be corrected. The Language was improved properly as well. My minor questions you may find in the attached file.

All best wishes,

Reviewer 

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

Thank you very much for your valuable suggestions regarding the revision of our manuscript. We have carefully addressed all the issues you raised concerning language, figures, and other aspects. Below are our point-by-point responses to your specific comments:

Comments 1: Regarding the description of indicators corresponding to the five parts of the DPSIR framework in the abstract.
Response 1: Thank you for your helpful suggestion. We have revised the abstract to explicitly list the 25 indicators used in this study and clearly assign them to the five dimensions of the DPSIR framework: driving forces, pressures, states, impacts, and responses (Line 16-27).

Comments 2: Regarding the readability of figure legends and the need to improve several figures.
Response 2: We sincerely appreciate your detailed comments on the figures. In response, we have revised the “Study Area” figure, enlarged the legends and labels (Line 228), and ensured that all figures are embedded in the manuscript at high resolution and original size. Additionally, we will request the editorial office to provide the source files of the figures to the reviewers for clearer examination.

Comments 3: Also, I suggest you to mark plateau areas and mountain areas in a separate figure.
Response 3: Thank you for this constructive suggestion. We have added a new figure titled “Range of Plateau and Mountainous Area of Kunming” at the end of the Introduction (Line 197). The identification of plateau and mountainous regions is based on Chinese scholars’ geomorphological classification of the Central Yunnan region (including Kunming), with references as follows:

• Z.Y., Liu, J.L., Wang. Geomorphological characteristics of Central Yunnan based on SRTM DEM [J]. Yunnan Geographic Environment Research, 2017, 29(06): 9-15+27.

• B.Y., Li, B.T., Pan, J.F., Han. A discussion on basic landform types and classification indicators in China [J]. Quaternary Sciences, 2008, (04): 535–543.

In these definitions, a plateau is an area with an altitude greater than 1000 meters and a terrain relief less than 30 meters, while mountainous areas are defined by a terrain relief greater than 200 meters.

Comments 4: I advise you to replicate the cap of the table on the next pages Nos 9 and 10 as well.
Response 4: Thank you for the suggestion. We have revised the tables to ensure that the headers are repeated on each page for consistency and readability (Line 323-334).

Comments 5: If it possible, try to do the conclusions Nos 1 and 2 some shorter.
Response 5: We appreciate your advice. We have revised conclusions (1) and (2), shortening the text while preserving the original meaning as much as possible (Line 621-654).

Comments 6: The final question. Is the method you use in this work adapted for described ecological system or this method is proposed here for the first time? In the abstract, I advise you to mark the novelty more clear in short words.
Response 6: The methods used in this study are adapted from existing ecological assessment approaches and tailored to the specific characteristics of plateau-mountainous urban ecosystems. Following your suggestion, we have added the following sentence to the abstract to highlight the novelty:"This study is the first to integrate the DPSIR framework, Entire-array Polygon method, and Grey System Forecasting Model into the evaluation and prediction of urban ecosystem resilience in plateau-mountainous cities." (Line 33-35)

Once again, we sincerely thank you for your insightful comments and look forward to receiving more of your valuable suggestions.

Sincerely
[Hui Li]