Review Reports - Multi-Scale Analysis of Influencing Factors for Temporal and Spatial Variations in PM<sub>2.5</sub> in the Yangtze River Economic Belt

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Assessment of the Paper

This study offers a robust spatiotemporal analysis of PM2.5 (fine particulate matter) concentrations in the Yangtze River Economic Belt (YREB) from 2005 to 2020, using a multi-scale approach (municipal, watershed, and grid levels) to uncover patterns and drivers.

Key methodologies include time-series analysis, MK trend analysis, Geodetector for factor identification, and Geographically and Temporally Weighted Regression (GTWR) for heterogeneity.

The findings reveal an inverted U-shaped temporal trend peaking in 2013, spatial clustering with higher pollution in central/eastern regions, and a significant overall decline (faster in the east/central areas). Influencing factors like population factors (PFA, PISA, PD), vegetation (NDVI), and meteorological elements (MAT, PWA) are dissected, showing scale-dependent effects—e.g., human activities dominate at broader scales, while temperature and secondary industries impact at finer ones.

Strengths include the innovative multi-scale framework, which provides more granular insights than traditional administrative analyses, and the integration of models to handle spatiotemporal heterogeneity.

The emphasis on factors like NDVI (suppressing pollution) and PFA/PD (promoting it) adds practical value for environmental governance. Overall, it's a well-executed piece of environmental science, with clear conclusions that build on existing literature while advancing methodologies.

Opinion on the Importance of the Study

This study is highly important for several reasons. PM2.5 is a critical public health and environmental issue, especially in rapidly developing regions like the YREB (China's economic powerhouse). By quantifying spatiotemporal trends and factor influences at multiple scales, it bridges gaps between macro-level policies and micro-level interventions, offering "more precise scientific support for regional PM2.5 governance" as noted. Its novelty lies in the combined use of Geodetector and GTWR, which reveals nuanced interactions (e.g., scale-dependent effects of factors) that single-scale studies often miss. This could inform targeted pollution control strategies, such as prioritizing vegetation restoration in polluted areas or adjusting industrial policies based on regional differences. In a broader context, it contributes to global efforts against air pollution (e.g., aligning with UN Sustainable Development Goals on clean air and health), and its data-driven approach could inspire similar analyses in other polluted belts worldwide. The downward trend post-2013 is encouraging evidence of policy effectiveness (e.g., China's air quality campaigns), but the study underscores ongoing challenges in western regions, making it timely and actionable for researchers, policymakers, and environmental agencies.

What Does This Topic Lack in This Material?

While the paper excels in analysis and insights, it has some notable gaps that could enhance its completeness and impact:

Causal Depth and Mechanisms: The study identifies correlations (e.g., positive links between PD and PM2.5), but it doesn't deeply explore underlying mechanisms. For instance, how exactly do factors like PISA (population in secondary industries) drive pollution? Adding qualitative insights or mechanistic modeling (e.g., via process-based simulations) would strengthen causality.
Long-Term Projections and Future Scenarios: The analysis stops at 2020, but with climate change and economic shifts (e.g., post-COVID recovery), projections under different scenarios (e.g., policy interventions or extreme weather) are missing. This limits predictive value for long-term planning.
Socioeconomic and Policy Integration: While human factors are noted, there's limited discussion on socioeconomic drivers (e.g., income inequality's role in pollution exposure) or policy evaluations (e.g., specific regulations' effects on the 2013 inflection point). Broader integration with equity and justice aspects could make it more holistic.
Data Limitations and Uncertainty: The abstract/conclusion don't detail data sources, quality, or uncertainties (e.g., sensor biases in PM2.5 measurements or model assumptions). Acknowledging these would improve reliability, especially for grid-scale analyses.
Comparative or Broader Contextualization: It focuses on the YREB but lacks comparisons with other regions (e.g., global benchmarks or similar economic belts like the Pearl River Delta). This could highlight uniqueness or generalizability.
Practical Recommendations: The conclusion summarizes findings but could include more actionable advice, like prioritized interventions for western slow-decline areas or monitoring strategies.

In summary, this is a valuable study that advances PM2.5 research through multi-scale innovation, but expanding on causality, futures, and policy depth would elevate it further.

In my opinion, the article is overloaded with illustrative material. Numerous abbreviations should be fully listed in the Abstract or presented on a separate sheet before the Introduction. However, the final decision rests with the editorial board and its technical services.

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions in track changes in the re-submitted files. ( All line numbers referenced below correspond to those displayed in the manuscript when the "Track Changes" mode is enabled)

Comments1: Causal Depth and Mechanisms: The study identifies correlations (e.g., positive links between PD and PM2.5), but it doesn't deeply explore underlying mechanisms. For instance, how exactly do factors like PISA (population in secondary industries) drive pollution? Adding qualitative insights or mechanistic modeling (e.g., via process-based simulations) would strengthen causality.

Response 1: Thank you very much for pointing out this issue. This study only analyzes the key influencing factors of PM2.5 concentrations and the variation patterns of their effects on PM2.5 concentrations, without in-depth exploration of the underlying mechanisms. This constitutes a limitation of the present research, which is discussed in detail in Section 4.3 (lines 689-693) of the manuscript.

Comments2: Long-Term Projections and Future Scenarios: The analysis stops at 2020, but with climate change and economic shifts (e.g., post-COVID recovery), projections under different scenarios (e.g., policy interventions or extreme weather) are missing. This limits predictive value for long-term planning.

Response 2: Thank you very much for pointing out this issue. Since this study is based on publicly available datasets obtained online and the analysis is conducted at the grid scale (the highest precision level adopted herein), the acquired data products cannot cover the most recent period. This also constitutes a limitation of the present research, which is described in Section 4.3 (lines 678-680) of the manuscript. Additionally, the prediction of PM2.5 concentrations is not the core research content of this study and thus is not addressed herein.

Comments3: Socioeconomic and Policy Integration: While human factors are noted, there's limited discussion on socioeconomic drivers (e.g., income inequality's role in pollution exposure) or policy evaluations (e.g., specific regulations' effects on the 2013 inflection point). Broader integration with equity and justice aspects could make it more holistic.

Response 3: Thank you very much for pointing out this issue. Since the highest research scale adopted in this study is the grid scale, most socioeconomic factors—predominantly statistical data—cannot be effectively disaggregated to the grid level. Consequently, the influencing factors investigated in this study are not exhaustive, with additional explanations provided in Section 4.3 (lines 673-676) of the manuscript. Regarding the evaluation of policy effects, supplementary content has been added, as detailed in Lines 339–341.

Comments4: Data Limitations and Uncertainty: The abstract/conclusion don't detail data sources, quality, or uncertainties (e.g., sensor biases in PM2.5 measurements or model assumptions). Acknowledging these would improve reliability, especially for grid-scale analyses.

Response 4: Thank you very much for pointing out this issue. Detailed information about the data sources is provided in Section 2.3 and Table 1 of the manuscript, while supplementary explanations regarding data quality are included in Section 4.3 (Lines 661–673).

Comments5: Comparative or Broader Contextualization: It focuses on the YREB but lacks comparisons with other regions (e.g., global benchmarks or similar economic belts like the Pearl River Delta). This could highlight uniqueness or generalizability.

Response 5: Thank you very much for pointing out this issue. This study focuses on the Yangtze River Economic Belt (YREB) as the research area, with the primary focus on comparative analysis of the differences among various regions within the YREB. Therefore, no supplementary comparative analysis between the YREB and other regions has been included in this manuscript.

Comments6: Practical Recommendations: The conclusion summarizes findings but could include more actionable advice, like prioritized interventions for western slow-decline areas or monitoring strategies.

Response 6: Thank you very much for pointing out this issue. The suggestions for the prevention and control of PM2.5 pollution can be found in Section 4.2 of the manuscript.

Reviewer 2 Report

Comments and Suggestions for Authors

General Comments :

This article is interesting, well-written, and the associated scientific treatment appears sound. However, it has several flaws, noted in the attached annotated file, which the authors should correct.

These can be grouped under four headings:

-1) This article uses a lot of acronyms, the meanings of which, as soon as they appear, must be carefully and clearly stated (for example, by referring to Table 1) while avoiding repetition.

-2) This article deals with the Chinese context but is aimed at an international audience and therefore the elements of this context that are unfamiliar to the reader must be explained: relocate the map of the region in relation to the whole of China, say what the 'Belt and Road Initiative cooperation' is, etc..

-3) The detailed description (mathematical formulas) of the Methods and technical processes used (2.4.1-2.4.3) must be returned to the appendix because it weighs down the text, and on the other hand the summary of the specificity of each of them must be briefly reported in the text.

-4) There is a lack of reflection and discussion on the correlation between certain interrelated explanatory variables (see 3.2.), hence their possible redundancy: for example, PISA and PD, CVO and GDP, etc. In this case, part of the variance of one can be statistically attributed to the other, without it being possible to clearly differentiate them. Therefore, we should only retain explanatory variables that are as independent of each other as possible. Similarly, your explanations concerning the VIF are incomplete and unclear.

-5) The space taken up by the illustrations in Appendix A1-24 takes up 9 pages, while the associated commentaries only take up 2 pages. These comments are also quite vague, without offering any real explanations for the observed changes. Wouldn't it be necessary to limit this section?

Detailed Comments :

See the attached annotated manuscript !

Comments for author File: Comments.pdf

Author Response

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions in track changes in the re-submitted files.（All line numbers referenced below correspond to those displayed in the manuscript when the "Track Changes" mode is enabled）

Comments 1: This article uses a lot of acronyms, the meanings of which, as soon as they appear, must be carefully and clearly stated (for example, by referring to Table 1) while avoiding repetition.

Response 1: Thank you very much for pointing out this issue. Regarding the concern you raised, corresponding revisions have been made in the manuscript: on Line 67, the full name of GDP is specified when it first appears; on Line 88, the full name of NDVI is indicated upon its initial occurrence; and on Line 107 and 123, the abbreviation GTWR is used directly when it appears for the second time.

Comments 2: This article deals with the Chinese context but is aimed at an international audience and therefore the elements of this context that are unfamiliar to the reader must be explained: relocate the map of the region in relation to the whole of China, say what the 'Belt and Road Initiative cooperation' is, etc..

Response 2: Thank you very much for raising this question. Regarding the concern you put forward, corresponding revisions have been made in the manuscript: on Line 91, specific explanations have been added to "the Belt and Road Initiative cooperation"; the specific location of the study area in China has been presented in Figure 1.

Comments 3: The detailed description (mathematical formulas) of the Methods and technical processes used (2.4.1-2.4.3) must be returned to the appendix because it weighs down the text, and on the other hand the summary of the specificity of each of them must be briefly reported in the text.

Response 3: Thank you very much for raising this question. In the manuscript, the content originally presented in Sections 2.4.1–2.4.3 has been moved to Appendix B, with a brief description of each item provided in the main text (see Lines 231–255 for details).

Comments 4: There is a lack of reflection and discussion on the correlation between certain interrelated explanatory variables (see 3.2.), hence their possible redundancy: for example, PISA and PD, CVO and GDP, etc. In this case, part of the variance of one can be statistically attributed to the other, without it being possible to clearly differentiate them. Therefore, we should only retain explanatory variables that are as independent of each other as possible. Similarly, your explanations concerning the VIF are incomplete and unclear.

Response 4: Thank you very much for raising this question. Indeed, the discussion on the correlations among variables is insufficient in Section 3.2 of the manuscript. The main discussion on the correlations among variables is provided during the implementation of the GTWR model (see Lines 487-494 and Table 2 for details), primarily by testing the multicollinearity among variables using the Variance Inflation Factor (VIF) to determine their correlations. For explanations of the VIF, refer to Lines 487-491 of the manuscript and Appendix B.4.

Comments 5:The space taken up by the illustrations in Appendix A1-24 takes up 9 pages, while the associated commentaries only take up 2 pages. These comments are also quite vague, without offering any real explanations for the observed changes. Wouldn't it be necessary to limit this section?

Response 5: Thank you very much for raising this question. To present the results more intuitively, the results of each factor at each scale have been visualized in the manuscript. After careful consideration, no optimal simplification method has been identified to better illustrate the findings; therefore, no streamlining of this section has been made in the current revision.

Supplementary Notes:

Regarding the specific issues you raised, corresponding revisions have been made in the manuscript:
The content originally in Line 72 of the initial manuscript has been revised and now appears in Lines 80–83 of the revised version.
"Level 7" originally in Line 133 of the initial manuscript has been deleted, which corresponds to Line 162 of the revised version.
A citation to Table 1 has been added in Section 2.3 (see Line 173 of the revised manuscript).
The content in Lines 205–208 of the revised manuscript has been deleted.
The explanation for "CLCD" in Line 212 of the revised manuscript can be found in Line 193: "China Land Cover Dataset (CLCD)".
Revisions have been made to the content corresponding to Line 275 of the initial manuscript.
The color scheme of Figure 4 has been standardized for consistency.

Reviewer 3 Report

Comments and Suggestions for Authors

The problem addressed in this study is relevant because it examines the dependence of the environmental situation on many factors. In the abstract, the authors explain the relevance of the problem and briefly describe the research methods and results. The article title and keywords adequately reflect the article's content.

In the introduction, the authors justify the relevance of the topic and its study in studies by various authors, the methods used, and the influence of various factors on urban air pollution. The second section is devoted to a description of the research materials and methods. In the third section, the authors describe the results of the applied research methods. The fourth section contains a discussion, including limitations and recommendations. In the concluding section, the authors summarize the results of the completed research.

The article has been prepared in accordance with the instructions for authors and is relevant to the topic it explores and publishes. The theoretical and practical conclusions are illustrated with figures and tables of sufficient quality. The list of references is sufficient in terms of the number of sources.

In our opinion, the article is relevant to the topic of "Ecological State of the Air Basin of Urbanized Areas" and corresponds to the type of Preliminary Study.

Comment.

Despite the relevance and timeliness of the study, there are some comments:

In our opinion, the authors need to strengthen the scientific component of the article: specify the purpose of the study, the tasks addressed in the article to achieve this goal, how the achieved results can be assessed, and how the proposed approach differs from the work of other authors.
The authors should clarify whether the methodology is universal, who will implement it, and how it will help improve the environmental situation.
The conclusion should identify the limitations of the study and recommendations for future work.
There are some shortcomings in the formatting. For example, Section 3.2.1 is listed twice.
It is recommended to expand the list of references by including more recent articles, as both the topic of ecology and measurement and multivariate analysis methods have been developing in recent years, especially with advances in the field of AI.

Author Response

Comments1: In our opinion, the authors need to strengthen the scientific component of the article: specify the purpose of the study, the tasks addressed in the article to achieve this goal, how the achieved results can be assessed, and how the proposed approach differs from the work of other authors.

Response 1: Thank you very much for pointing out this issue. The revised parts regarding the research objectives and the specific tasks addressed in the manuscript to achieve these objectives can be found in Lines 115–126. The differences between the research method proposed in this study and those adopted by other scholars are detailed in Lines 107–114.

Comments2: The authors should clarify whether the methodology is universal, who will implement it, and how it will help improve the environmental situation.

Response 2: Thank you very much for pointing out this issue. The generalizability of the research method proposed in this study can be demonstrated from two aspects. Firstly, all the research data used in this manuscript are publicly available (see Table 1 for details). Secondly, the research methods employed are relatively mature. The implications of this method for environmental improvement can be found in Section 4.2 of the manuscript, which presents policy recommendations for mitigating PM2.5 pollution.

Comments3: The conclusion should identify the limitations of the study and recommendations for future work.

Response 3: Thank you very much for pointing out this issue. Detailed explanations regarding the limitations of this study and suggestions for future research work are provided in Section 4.3 of the manuscript.

Comments4: There are some shortcomings in the formatting. For example, Section 3.2.1 is listed twice.

Response 4: Thank you very much for pointing out this issue. Regarding the concern you raised, the necessary revision has been made: the duplicated section number "3.2.1" has been corrected to "3.2.2".

Comments5: It is recommended to expand the list of references by including more recent articles, as both the topic of ecology and measurement and multivariate analysis methods have been developing in recent years, especially with advances in the field of AI.

Response 5: Thank you very much for pointing out this issue. Following your suggestion, supplementary references have been added to the reference section, as detailed in References 20–24.