Next Article in Journal
A Robust Sustainability Assessment Methodology for Aircraft Parts: Application to a Fuselage Panel
Previous Article in Journal
Assessment of the Saher System in Enhancing Traffic Control and Road Safety: Insights from Experts for Dammam, Saudi Arabia
 
 
Article
Peer-Review Record

Temporal–Spatial Evolution and Driving Mechanism for an Ecosystem Health Service Based on the GD-MGWR-XGBOOT-SEM Model: A Case Study in Guangxi Region

Sustainability 2025, 17(8), 3305; https://doi.org/10.3390/su17083305
by Zhenfeng Wei 1, Dong Chen 2,*, Qunying Huang 1, Qifeng Chen 2 and Chunxia Wei 3
Reviewer 2: Anonymous
Sustainability 2025, 17(8), 3305; https://doi.org/10.3390/su17083305
Submission received: 19 February 2025 / Revised: 28 March 2025 / Accepted: 31 March 2025 / Published: 8 April 2025

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article “Temporal-spatial Evolution and Driving Mechanism for Eco-system Health Service Based on GD-MGWR-XGBOOT-SEM model: A Case Study in Guangxi Region” is well-written and easy to understand. The manuscript is clear and relevant for diagnosing the regional ecosystem health status and the main factors which are affecting to it. The article is well-organized and contain four well-structured and well-developed sections: 1) Introduction, 2) Materials and Methods, 3) Results and 4) Discussion and Conclusions.

The authors do a good job of synthesizing the literature that concider monitoring, assessing and governing ecosystem health in order to achieve sustainable development and formulate ecosystem restoration strategies. The review is clear, comprehensive and doesn’t include any self-citations. The cited references are mostly recent publications (more than 71% within the last 5 years) and relevant for traditional analysis statistics-based research methods as well as machine and deep learning. These methods can only study relation, interaction and strength of the driving factors. However, the mechanism of driving factors on regional ecosystem health is missing and this was identified as a gap in knowledge. This paper introduces the adequate mechanism.

This study selects potential driving factors from four aspects, including vegetation cover, climatic conditions, topography and human activities and variable declaration was given in Table 2. Indicators and data sources for Guangxi Region are given in Table 1. The research methodology is clearly explained. This study adopted the improved vigor-organization-resilience-service (VORS) model and combined geographic detector (GD), multiscale geographically weighted regression model (MGWR), with XGBOOTS-SHAP model. Excellent technical workflow is presented in Figure 2. The Ecosystem health assessment model selected landscape indices such as Shannon diversity index and area-weighted average patch fractal index, determined the weights and calculated the ecosystem organization force. XGBOOST-SHAP model, which is known for its highly flexible parameter tuning and strong noise resistance, has demonstrated considerable effectiveness in nonlinear fitting applications. Improved MGWR model allowed different covariates to have different bandwidths, which is more sensitive to the spatial heterogeneity of geographical phenomena model. Structural equation model was used to establish, estimate and test causality model, and analyze the role of individual indicators on the whole and the mutual relationship between individual indicators. Combination of model is appropriate and the manuscript’s results are reproducible based on the details given in the methods section.

Is the manuscript scientifically sound  and results are convinced thanks to extremely vivid, comprehensive and temporal visual presentation. Characteristics of ecosystem health changes are presented from 2000 to 2020 separately via 1) Spatial evolution of ecosystem health; 2) Moran Index of ecosystem health regional aggregation and Evolution space of ecosystem health agglomeration. Spatial influence analysis of driving factors included impacts, interactions and coefficients of MGWR model. The analysis of driving factors influence mechanism presented the changes in the ordering of characteristics in a very effective manner. The critical value changes of the impact of driving factors on the health are thoroughly analyzed. Influence of driving factors on ecosystem health were analyzed by structural equation model.

Considering the influence factors (including vegetation, terrain, climate and human activities), the mechanism of driving factors on Guangxi ecosystem health from 2000 to 2020 was analyzed by using combining model to assess ecosystem health and reveal the main driving factors. The conclusions are: 1) frequent human activities have a greater impact on the ecological environment and damage in the early stage of urbanization, and most of them have negative effects which results in a relatively low ecosystem health; 2) influence characteristics of climate variability is relatively weak, and spatial differences are caused by terrain complexity; 3) natural factors has positive effect on ecosystem health especially the vegetation elements have a dominant role.

This article contains ten figures and they are easy to interpret and understand. The data are properly shown in three tables. The five appropriate equations are used in methodology section. 

 

In line 352 and line 398 are the same subtitles “The dependent characteristics of driving factors“. Please change the unappropriate one.

The errors are in lines 254, 265, 276, 295, 323, 351 and 397. There are misspelling Firue 3-10.  Please change those to Figure 3-10.

Author Response

RESPONSE TO REVIEWER 1

General Comments (Reviewer 1)

Response: We greatly appreciate the reviewer for the insightful comments, which help us improve the quality of the paper significantly. We have carefully revised the manuscript according to your valuable suggestions. The revised contents are highlighted in yellow in the manuscript.

Comment 1 (Reviewer 1)

In line 352 and line 398 are the same subtitles “The dependent characteristics of driving factors”.

Response: We sincerely appreciate the reviewer’s insightful comments. We changed the title of line 352 to “The impact characteristics of driving factors”. Line 398 remains the same.

Comment 2 (Reviewer 1)

The errors are in lines 254, 265, 276, 295, 323, 351 and 397. There are misspelling Firue 3-10.  Please change those to Figure 3-10.

Response: We sincerely appreciate the reviewer’s insightful comments. Because we didn't check carefully enough, we made some spelling mistakes. The erroneous content has been revised according to the suggestions of the reviewers. In addition, we corrected the grammatical errors and polished the representations carefully. The manuscript was refined the words to improve the readability.

Reviewer 2 Report

Comments and Suggestions for Authors

Review for “Temporal-spatial Evolution and Driving Mechanism for Ecosystem Health Service Based on GD-MGWR-XGBOOT-SEM 3model: A Case Study in Guangxi Region”

 

The authors have analyzed the relationship between ecosystem service value and urbanization, which holds significant academic value. The article effectively integrates both theoretical and empirical perspectives, utilizing practical case studies to explore the application of ecosystem health assessment in regional development. However, there remains room for improvement in the data analysis and the rigor of the conclusions. Certain aspects of the logic and expression could benefit from further refinement to enhance clarity and precision.

  1. section 1: This section highlights the link between ecosystem health and urbanization but should further emphasize the significance of ecosystem service value for policy, economic development, and social stability. A brief overview of current research would also identify gaps and showcase the study’s innovative contributions.
  2. Section 2.1: L134~144, the description of the research area is rather simplistic and lacks detailed information on the ecological characteristics of Guangxi. The natural environment and socio-economic background of Guangxi could significantly influence the research findings. It is recommended to elaborate on Guangxi's geographical location, climatic features, and socio-economic structure, explaining how these factors may impact the research conclusions.
  3. Section 2.2: The article employs a geographical detector model to analyze ecological health, but the methodology section is too brief. It is necessary to provide a detailed explanation of the principles of the geographical detector, its application methods, and its suitability, particularly how it is applied in this study.
  4. Section 4: The discussion section should further analyze the practical implications of the results and address potential limitations of the study. For example, the assumptions underlying the model and potential sources of error could impact the accuracy of the results.

 

Author Response

RESPONSE TO REVIEWER 2

General Comments (Reviewer 2)

Response: We greatly appreciate the reviewer for the insightful comments, which help us improve the quality of the paper significantly. We have carefully revised the manuscript according to your valuable suggestions. The revised contents are highlighted in yellow in the manuscript.

Comment 1 (Reviewer 2)

Section 1: This section highlights the link between ecosystem health and urbanization but should further emphasize the significance of ecosystem service value for policy, economic development, and social stability. A brief overview of current research would also identify gaps and showcase the study’s innovative contributions.

Response: We sincerely thank the reviewer for their constructive suggestion. We added the research results of ecosystem health on economy, policy, and human social activities in the introduction. The research literature was summarized and commented. At the same time, we also carefully analyzed the bridges and gaps between our research and these research results. We further presented our research perspectives and main contributions. All revisions could be found in the introduction.

Comment 2 (Reviewer 2)

Section 2.1: L134~144, the description of the research area is rather simplistic and lacks detailed information on the ecological characteristics of Guangxi. The natural environment and socio-economic background of Guangxi could significantly influence the research findings. It is recommended to elaborate on Guangxi's geographical location, climatic features, and socio-economic structure, explaining how these factors may impact the research conclusions.

Response: We sincerely thank the reviewer for their constructive suggestion for regarding the clarity of study object. In the revised manuscript, we have greatly improved the introduction and analysis of the study area of Guangxi, including geographical location, temperature, vegetation conditions, water resources, soil conditions, crops, population distribution, and socio-economic structure. In the study, we selected vegetation, climatic conditions, geographical shape and human activities as the main objects. Because in our preliminary analysis, these several factors are very significant, in order to more accurately reflect the actual situation of regional ecological environment evolution mechanism. Meanwhile, the research pays attention to the comprehensive analysis of regional characteristics and multi-dimensional analysis to improve the rationality and practicability. All revisions could be found in Section 2.1.

Comment 3 (Reviewer 2)

Section 2.2: The article employs a geographical detector model to analyze ecological health, but the methodology section is too brief. It is necessary to provide a detailed explanation of the principles of the geographical detector, its application methods, and its suitability, particularly how it is applied in this study.

Response: We greatly appreciate the reviewer for the insightful comments, which help us improve the quality of the paper significantly. We added 2.2.4 Geodetector(GD) to introduce in detail the principle of GD, the use conditions, the use skills, especially for the usage instructions of this paper.

Comment 4 (Reviewer 2)

Section 4: The discussion section should further analyze the practical implications of the results and address potential limitations of the study. For example, the assumptions underlying the model and potential sources of error could impact the accuracy of the results.

Response: We sincerely thank the reviewer for their constructive suggestion. We discussed potential limitations of this study. in Discussions and Conclusions. We introduced the research method and scope in detail, and also pointed out the shortcomings of the research. Finally, some constructive ideas and directions for future research are put forward. All modifications have been marked in yellow in Discussions and Conclusions.

Reviewer 3 Report

Comments and Suggestions for Authors

This paper presents a study on developing a diagnostic methodology for regional ecosystem health from 2000 to 2020 in Guangxi and identifying the main factors affecting ecosystem health. The study uses EHI, an ecosystem health index, which is estimated by the authors as the geometric mean of EV, EO, ER, and ES, respectively, the ecosystem vitality index, ecosystem organization index, ecosystem resilience index, and ecosystem service index. EHI is divided into five levels.

The influencing mechanism of driving factors on regional ecosystem health was analyzed using the XGBOOTS-SHAP model and the SEM structural equation model. These are mainly methods based on game theory, and they are used to explain the performance of machine learning models.

Structural Equation Modeling (SEM) is a statistical method that combines various mathematical models and computer algorithms to analyze complex relationships between variables. However, it is worth noting that the model requires a large amount of data, and its results are difficult to interpret. The method is widely used in social sciences, business, and scientific research to test theoretical models.

The authors use a systems approach to diagnose the health of regional ecosystems from 2000 to 2020 in Guangxi and identify the main factors affecting ecosystem health.

The authors note that a significant result of the study is the confirmation that the vegetation factor has a dominant positive effect on ecosystem health. This is indeed the case, although it is not an original discovery.

Weaknesses and remarks:

The study object, Guangxi Zhuang Autonomous Region, is very large for modeling: its area is 235,000 km2, and the population is more than 50 million people.
The study used non-specialized and non-standardized mathematical models for ecological research, which require verification and validation.
The authors do not discuss what they consider to be the confirmation of their study results and the limitations of the methodology. What is required here is not a formal presentation of the processes and statistics given, but an analysis of the final quantitative results with possible ways of regulating them using an objective and reliable analytical database.
Fig. 2 has little information value without a detailed review and comments from the authors.
To understand the leading processes and factors influencing the state of the ecosystem, it is necessary to select and present data on the most studied key areas (such data are components of the information model).

Conclusion:

The article represents a certain contribution to the development of the methodology for diagnosing the health of ecosystems. However, it can be recommended for publication only after taking into account the above comments and making the appropriate clarifications and adjustments.

Author Response

RESPONSE TO REVIEWER 3

General Comments (Reviewer 3)

Response: We greatly appreciate the reviewer for the insightful comments, which help us improve the quality of the paper significantly. We have carefully revised the manuscript according to your valuable suggestions. The revised contents are highlighted in yellow in the manuscript.

Comment 1 (Reviewer 3)

The study object, Guangxi Zhuang Autonomous Region, is very large for modeling: its area is 235,000 km2, and the population is more than 50 million people. The study used non-specialized and non-standardized mathematical models for ecological research, which require verification and validation.

Response: We sincerely appreciate the reviewer’s insightful comments and valuable questions. In this study, we drew on the research ideas and research methods of scholars who have published their research results (as the following references [1-4]) on regional ecosystem health assessment in journals and gained attention. On this basis, we also made an improvement and applied them to the study of Guangxi ecosystem health assessment.

[1]Li, Y. Y.; Qin, L.; Wang, Y.H.; et al. Ecosystem Health Assessment of the Largest Lake Wetland in the Yellow River Basin Using An Improved Vigor-Organization-Resilience-Services Model. Ecological Indicators 2024, 166, 112539.

[2] Zhang, X.P.; Wu, T.X.; Du, Q.Q.; et al. Spatiotemporal Changes of Ecosystem Health and the Impact of Its Driving Factors on the Loess Plateau in China. Ecological Indicators 2025, 170, 113020.

[3] Shen, W.; Li, Y.; Qin, Y. C. Research on the Influencing Factors and Multi-Scale Regulatory Pathway of Ecosystem Health: A Case Study in the Middle Reaches of the Yellow River, China. J. Clean. Prod. 2023, 406, 137038.

[4] Song, F.; Su, F. L.; Mi, C. X., Sun, D. Analysis of Driving Forces on Wetland Ecosystem Services Value Change: A Case in Northeast China. Sci. Total Environ. 2021, 751, 141778.

Comment 2 (Reviewer 3)

The authors do not discuss what they consider to be the confirmation of their study results and the limitations of the methodology. What is required here is not a formal presentation of the processes and statistics given, but an analysis of the final quantitative results with possible ways of regulating them using an objective and reliable analytical database.

Response: We sincerely thank the reviewer for their constructive and insightful suggestion. We added the limitations and shortcomings of the study in revised manuscript. We also looked forward to future research interests and directions. Meanwhile, based on the discussions of research data, some meaningful conclusions are given. All modifications have been marked in yellow in Discussions and Conclusions.

Comment 3 (Reviewer 3)

Fig. 2 has little information value without a detailed review and comments from the authors. To understand the leading processes and factors influencing the state of the ecosystem, it is necessary to select and present data on the most studied key areas (such data are components of the information model).

Response: Many thanks to the authors for their suggestions. In 2.2 Research method, we introduced the research steps in framework diagram (Figure 2) in detail, and explained the meaning of variables, data sources and detailed process. This helps the readers to understand the research work. Since the data is relatively large, we have not been able to present it in the paper for the time being. 

 

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed all my concerned, and I suggest the manuscript should be accepted as current format.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors briefly commented on my remarks, but only on three points, and made a number of additions directly to the text of the article. Although these additions, despite their lack of information, are formally reasonable, I do not believe that they remove all the problems I have emphasized.

Nevertheless, I can assume that the Editorial Board may take responsibility and decide to publish the article.

I have no objection and will be sympathetic to this possible decision of the Editorial Board.

I agree to the publication of the article.

Back to TopTop