Atmospheric Particulate Matter Pollution in the “U-C-S” Urban Agglomeration: Spatio-Temporal Distribution and Source Analysis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Comments
This paper explained several analyses of PM concentration behavior in western Chian. The results seem to be reasonable. But there are some mistakes in Figures ad text, and some corrections are necessary.

Detailed comments

Line41, 46, 49: Reference error

Line51 etc.:  reference number “910” -> “9, 10”

Line55: “..” -> “.”

Figure 1: 
It is better to show the locations of Urmqi city, Shihezi City, Wujiaqu city etc. in this map.

Figure 2, Figure3, Figure 4 , Figure 7, Figure 8:  Unit (micro g m^-3) should be added in the color scale bars. 

Figure 3: The color is incorrect in the figures of PM10. 

Figure5: The locations of the cities (areas) should be shown in Figure 1.

Figure 5, Figure 6: The names of the y axis are incorrect. “PM2.5”-> “PM10”

Figure 9: It is better to explain in the main text and/or figure caption that y axis is started from the middle of the concentration, but from not zero. 
Why is Wednesday highest concentration? If pollutants accumulate during the weekdays, Friday will become highest concentration.

Line406: You explained that the reason why PM10 is higher in weekend is ”economic and human activities”.  What human activities emit this enhanced PM10 actually?

Line412: “figure” -> “figure10”

Line423-424: In Figure5 winter PM10,   Urmuqi＞ChangiiPrefecture.

Figure11: ”The daily average concentrations of PM2.5 and PM10 in cities” should be removed.

Figure 12: The Trajectory numbers should be shown in the figures.
There are Chinese characters in the axis. Please replace to N or E.

Table2: 
The table is divided at not the end of the page. 
Horizontal lines should be added between cluster 6 and cluster 1 to clear the bordars of each season.

Figure13, Figure 14:  It is difficult to recognize the location of the map. It is better to shown the location in Figure 1.

Line529: 　“contributionFig” ?

Figure15, 16: Is the any reason to use different expressions for two correlation Matrixes?

Line577-579: The explanation about the relationship between temperature and concentration changes is reasonable. But if you use all data, the relationship is caused by seasonal change (high PM during cold winter, low PM during hot seasons).  If you want to show the controlling factor of high PM, it is better to make a plot of analyze for selected period. 

Figure17, random forest model ：It is nice trial to apply the random forest model to the analysis of PM controlling factors. But the results in figure 17 seem to be not very informative. I am not sure if this analysis is necessary here. How about analyzing only selected period (winter etc.)?

Author Response

Response to Reviewer 1

 

We sincerely appreciate the reviewer’s insightful comments and constructive suggestions, which have significantly improved the quality of our manuscript. All the issues raised have been carefully addressed as detailed below.

 

Comment: Line 41, 46, 49: Reference error

Response: Thank you for pointing this out. The reference formatting errors on these lines have been corrected.

 

Comment: Line 51 etc.: reference number “910” → “9, 10”

Response: Corrected. The reference numbers have been properly separated as “9, 10”.

 

Comment: Line 55: “..” → “.”

Response: Fixed. The punctuation has been revised to a single period.

 

Comment: Figure 1: It is better to show the locations of Urumqi city, Shihezi City, Wujiaqu city etc. in this map.

Response: We have added the geographic locations of Urumqi, Shihezi, Wujiaqu, and other key cities to Figure 1 for better spatial context.

 

Comment: Figure 2, Figure 3, Figure 4, Figure 7, Figure 8: Unit (μg·m^-3) should be added in the color scale bars.

Response: The unit “μg·m^-3” has now been included in the colorbars of all relevant figures.

 

Comment: Figure 3: The color is incorrect in the figures of PM₁₀.

Response: The colormap for PM₁₀ in Figure 3 has been corrected to ensure consistency and accuracy.

 

Comment: Figure 5: The locations of the cities (areas) should be shown in Figure 1.

Response: As suggested, the study areas referenced in Figure 5 are now clearly indicated in the updated Figure 1.

 

Comment: Figure 5, Figure 6: The names of the y-axis are incorrect. “PM_2.5” → “PM₁₀”

Response: The y-axis labels in Figures 5 and 6 have been corrected from “PM_2.5” to “PM₁₀”.

 

Comment: Figure 9: It is better to explain in the main text and/or figure caption that y-axis starts from the middle of the concentration, not from zero. Why is Wednesday the highest concentration? If pollutants accumulate during weekdays, Friday should be highest.

Response: We have added clarification in both the main text (Line 452) and the caption of Figure 9, noting that the y-axis does not start at zero. Regarding the weekday pattern, we acknowledge the reviewer’s point and have revised the discussion to consider possible local emission patterns, meteorological conditions, or data aggregation effects that may lead to the observed Wednesday peak, rather than a simple weekday accumulation model.

 

Comment: Line 406: You explained that the reason why PM₁₀ is higher on weekends is “economic and human activities”. What human activities actually emit this enhanced PM₁₀?

Response: We have expanded the Introduction (Line 58) to specify likely weekend-related sources, such as increased construction activity, road dust resuspension from leisure travel, and agricultural burning, which are known to contribute to coarse particulate matter (PM₁₀) in the region.

 

Comment: Line 412: “figure” → “Figure 10”

Response: Corrected to “Figure 10”.

 

Comment: Line 423–424: In Figure 5 winter PM₁₀, Urumqi > Changji Prefecture.

Response: Thank you for the observation. We have verified the data and clarified in the text (Line XXX) that Urumqi indeed shows higher winter PM₁₀ levels than Changji Prefecture, consistent with its larger urban footprint and traffic emissions.

 

Comment: Figure 11: “The daily average concentrations of PM_2.5 and PM₁₀ in cities” should be removed.

Response: The redundant title has been removed from Figure 11.

 

Comment: Figure 12: Trajectory numbers should be shown in the figures. There are Chinese characters in the axis. Please replace with N or E.

Response: Trajectory cluster numbers are now labeled directly on Figure 12, and all Chinese characters on the axes have been replaced with standard directional labels (e.g., “N”, “E”).

 

Comment: Table 2: The table is split inappropriately across pages. Horizontal lines should be added between Cluster 6 and Cluster 1 to clarify seasonal boundaries.

Response: The table layout has been adjusted to avoid mid-table page breaks, and clear horizontal lines have been inserted to demarcate seasonal clusters.

 

Comment: Figure 13, Figure 14: It is difficult to recognize the location of the map. It is better to show the location in Figure 1.

Response: The regional context of Figures 13 and 14 is now linked to the base map in Figure 1, with an inset or annotation added for clarity.

 

Comment: Line 529: “contributionFig” ?

Response: This was a typo. It has been corrected to “contribution figure” (or the appropriate term based on context).

 

Comment: Figure 15, 16: Is there any reason to use different expressions for the two correlation matrices?

Response: The difference arose from software defaults. For consistency, both correlation matrices (Figures 15 and 16) now use the same visual style and annotation format.

 

Comment: Line 577–579: The explanation about temperature and concentration is reasonable, but the relationship may simply reflect seasonal variation. To identify true controlling factors, analysis over a selected period (e.g., winter) is recommended.

Response: We agree with the reviewer. Accordingly, we have added a focused analysis of winter months (Lines 681–708) to isolate meteorological influences from seasonal trends, strengthening our interpretation of temperature’s role.

 

Comment: Figure 17 (Random Forest model): The results seem uninformative. Consider analyzing only a selected period (e.g., winter).

Response: Following this valuable suggestion, we have re-run the Random Forest analysis using winter data only. The updated Figure 17 now provides clearer insights into the dominant drivers of PM during high-pollution periods, and the text has been revised accordingly (Lines 681–708).

 

Once again, we thank the reviewer for their time and thoughtful feedback. We believe the manuscript has been substantially improved through this revision process.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript submitted to the Editorial Office focuses on the issue of spatial and temporal variation of particulate matter pollution(PM₂.₅ and PM₁₀). The research was conducted  in the “Urumqi–Changji–Shihezi–Wujiaqu” agglomeration(known as U-C-S) in the Xinjiang region of northwestern China. This topic remains important and relevant, and the choice of region is crucial for the originality of the results. The Authors combine classical methods (trajectories and PSCF/CWT) with machine learning (random forest) to analyze sources and predict concentrations. The study used a relatively large data set (20 measuring stations, hourly data), but the observation period was short (4 years). Obviously, the longer the observation period, the more universal the results become.

The results confirmed the relationships known from other locations regarding differences in pollution levels on a seasonal, weekly, and hourly basis. The Authors also attempted to partially explain the observed patterns. The study contributes to the understanding of atmospheric pollution by particulate matter in the Xinjiang region. The results of this study may also have practical significance for the design of measures to reduce pollutant emissions in the region under study.

The weaknesses of the study include the lack of clear reference to environmental policy, the fact that some of the descriptive analyses are repetitive and could be shortened, and the lack of accurate validation of the RF model (e.g., R² coefficients, RMSE). A major shortcoming of the study is the lack of scientific discussion—a comparison of the results obtained with those of other researchers. 

The conclusions (lines 604-628) are a repetition of the main initial results. The work requires careful editing, as some references in the text contain technical errors (“Error! Reference source not found.”), and the figures also need correction. In the bibliography, some references are incompletely formatted (e.g., some DOI numbers and citation errors).

Please supplement the introduction with a broader international context and references to research results from other regions of the world; also, please supplement the basic information on typical anthropogenic sources of PMx.

I believe that due to the originality of the research area and the importance of the topic, the paper can be published in the journal after corrections have been made.

Specific comments:

- lines: 41, 46, 49: please remove “Error! Reference source not found.”

- please correct Figure 1, add the names of the areas and an explanation of what each of the individual figures shows

 - The 4-year study period is not a “long-term” period, as the Authors write (line 102);

- Please provide the exact start and end dates of the period from which the air pollution data originate (line 92)

 - Please supplement the information on meteorological data (line 110) – what is the temperature (air, average – maximum – minimum?, unit; measurement from what height above ground level?; what is the humidity – what type, unit. Why were the total precipitation and/or number of days with precipitation omitted?

- In Fig. 2 (two upper drawings) there are incomprehensible symbols.

Line 219 - “4. Results and Discussion” - please remove “Discussion” from the title, it is in Chapter 5.

 - Please specify what concentration levels were defined as high in a specific case? (from what threshold) – e.g., line 244. Based on what criteria?

- In the introduction or discussion, please refer to the observed dust concentration levels in the studied region in relation to WHO recommendations and concentration levels observed in other regions of the world. If there are any established standards for permissible or alert pollution levels in the studied region, please also add a note to the figures showing the concentration levels.

“Figure 2. Spatial distribution of average concentrations.”  – of what? where? Please complete the figure caption.

-  “Figure 3. Spatial distribution of annual mean concentrations.” / Figure 4. Spatial distribution of seasonal mean concentrations. – of what? where? Please complete the figure captions.

Figure 4 – please modify the figure captions or descriptions so that the same information is not repeated (e.g., “concentration” is used 8 times)

Fig. 2, Fig. 3, Fig. 4 – no units are given for pollutant concentrations; the color scales should be standardized across figures (the same color should denote the same concentration value). Mark the criteria for WHO and/or local standards on the scale

- line 244  “high PM2.5 concentration zones”, but also: lines 259; 276;  338; 634; – please explain the criteria for the separation and spatial extent of the described pollution concentration zones;

-Figure 5. Seasonal mean concentration changes."  - of what, where? Please standardize the vertical axes for all graphs in the figure (currently it is difficult to compare results between seasons).

“Figure 6. Monthly Mean Concentration Variation” – the caption should be made as for the previous figures, the scale on the OY axis should be standardized for the same pollutants, and the unnecessary description of the OY axis (“monthly average PM2.5”) should be removed. What types of particulate matter are shown in Figures e-h? The description of the OY axis (PM2.5) differs from the title of the figure (PM10).

Fig. 6 and Fig. 7 – the designation of months should be the same (either 1-12 or Jan-Dec),

Fig. 7, Fig. 8.  – no units

“Figure 10. Daily mean concentration changes.” – where, when?, no units

Fig. 12 – please remove unknown symbols next to latitude and longitude markings

“Figure 12. Backward trajectory and air pressure.” – where, when?, what do the black lines in the figure mean? Similarly, Fig. 13.

Fig. 13 – no explanation of the symbol “WPSCF”?; units?

Fig. 14 – please explain “WCWT”

Chapter 5. Discussion: Fig. 15-Fig. 17 and text – this is a continuation of the results, no scientific discussion.

Author Response

Response to Reviewer 2

 

We sincerely thank the reviewer for their thorough and constructive comments, which have greatly enhanced the scientific rigor and clarity of our manuscript. All concerns raised have been carefully addressed as detailed below.

 

Comment: Lines 41, 46, 49: Please remove “Error! Reference source not found.”
Response: These reference errors have been corrected throughout the manuscript.

 

Comment: Please correct Figure 1: add the names of the areas and an explanation of what each panel shows.
Response: Figure 1 has been revised to include clear labels for Urumqi, Changji, Shihezi, Wujiaqu, and other key areas. A detailed caption now explains the content of each subfigure.

 

Comment: The 4-year study period is not a “long-term” period (Line 102).
Response: We agree. The term “long-term” has been removed.

 

Comment: Please provide the exact start and end dates of the air pollution data (Line 92).
Response: The data period is now explicitly stated as “from January 1, 2018, to December 31, 2021” (Line 104).

 

Comment: Please supplement information on meteorological data (Line 110): temperature (average/max/min, unit, measurement height), humidity type and unit, and explain why precipitation data were omitted.
Response: We have expanded Section 2.2 to specify that temperature (°C) and relative humidity (%) were measured at standard 1.5-m height above ground, with daily min/mean/max values provided. Precipitation data were not included in the main analysis because preliminary tests indicated a weak correlation between precipitation and PM variations in this region.

 

Comment: In Fig. 2 (upper panels), there are incomprehensible symbols.
Response: All ambiguous symbols in Figure 2 have been removed or replaced with standard, clearly defined markers.

 

Comment: Line 219 – “4. Results and Discussion”: please remove “Discussion” from the title, as discussion is in Chapter 5.
Response: The section title has been changed to “4. Results” to align with the structure.

 

Comment: Please specify what concentration levels were defined as “high” (e.g., Line 244). Based on what criteria?
Response: We now define “high PM2.5 concentration zones” as the Chinese Grade II standard (75 μg·m^-3 for 24-h mean). This threshold is explicitly stated in Lines 257–265 and referenced in relevant figure captions.

 

Comment: Please supplement the Introduction with broader international context, global PM research references, and basic information on typical anthropogenic PM sources.
Response: .The Introduction (Section 1) has been expanded to include the necessary information.

 

Comment: Refer to WHO guidelines and global/regional PM standards; add these benchmarks to figures showing concentrations.
Response: Visual markers have now been added to the color bars of Figures 2–4 and 6–8 in the text.

 

Comment: Figure captions (Figs. 2–4, etc.) are incomplete—specify pollutant, region, and time frame; avoid repetitive wording; standardize units and color scales.
Response: All figure captions have been rewritten for completeness and consistency.

 

Comment: Figures 5–10, 12: Axes lack units, inconsistent labeling.
Response: All figures have been harmonized:

Units (μg·m^-3) added to all concentration axes.

Month labels standardized to “Jan–Dec” format.

Y-axis titles corrected for accuracy .

Figure 10 now includes full caption: “Daily mean concentration changes in the "U-C-S" urban agglomeration.”.

 

Comment: Figure 12: Remove unknown symbols next to latitude/longitude; clarify meaning of black lines in “Backward trajectory and air pressure” plot.
Response: Extraneous symbols have been removed. The black lines in the figure represent China’s prefecture-level city boundaries and have been labeled accordingly.

 

Comment: Figures 13 & 14: Explain acronyms “WPSCF” and “WCWT”; add units.
Response: “WPSCF” (Weighted Potential Source Contribution Function) and “WCWT” (Weighted Concentration Weighted Trajectory) are now defined at first use in the text and in figure captions. Units for both are dimensionless probability-weighted indices, as per standard PSCF/CWT methodology.

Comment: Chapter 5 (“Discussion”) mainly repeats results (Figs. 15–17); lacks scientific comparison with other studies.
Response: I have added the following references:

51. Jiang L, Liu YS, Yang YJ. Spatial and temporal characteristics of PM2.5 and O3 pollution in Jiangxi Province from 2015 to 2023 and their relationship with meteorological elements [J]. Environmental Science Research, 2025,38 (3): 460 .

52.Zhang ZY,Zhang XL,Gong DY,et al.Evolution of surface O3 and PM2.5 concentrations and their relationships with meteorological conditions over the last decade in Beijing[J].Atmospheric Environment,2015,108:67-75.

 

Comment: Conclusions (Lines 604–628) repeat initial results.
Response: The Conclusions section has been rewritten to emphasize novel insights, policy implications, and future research directions, rather than restating results.

Comment: References contain formatting errors (missing DOIs, incomplete entries).
Response: DOIs have been added for all references that possess them. Some references—such as government bulletins, books, and Chinese-language publications—do not include DOIs because their publication type or source does not assign DOIs.

 

We are grateful for the reviewer’s thoughtful critique, which has significantly strengthened the manuscript’s scientific contribution and presentation. We believe all concerns have been adequately addressed.

 

 

 

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

see attached file

Comments for author File: Comments.pdf

Author Response

Response to Reviewer (from atmosphere-3971464)

We sincerely thank the reviewer for their thorough and constructive feedback, which has helped us significantly improve the clarity, accuracy, and scientific rigor of our manuscript. All comments have been carefully addressed as detailed below.

 

General Comments

“The topic is scientifically not new, but the thorough investigation including some seldom used tools like cluster analysis favours publication... readability could be considerably improved.”

 

Response:

 

We appreciate the reviewer’s positive assessment of our methodological approach and agree that enhancing readability is essential. We have implemented all suggested improvements regarding figure labeling, caption clarity, reference formatting, and textual consistency throughout the revised manuscript.

 

Detailed Responses

Comment: Before line 35: A title must be introduced (preferably “Introduction”).

 

Response: The section heading “Introduction” has been added before line 35.

 

Comment: Lines 41–49: References not correctly displayed.

 

Response: Reference formatting in this section has been corrected according to journal style.

 

Comment: Line 51 etc.: Link to references should be “9, 10” and not “910”.

 

Response: All concatenated reference numbers (e.g., “910”) have been properly separated as “9, 10”.

 

Comment: Figure 1: Maps must be described in legend or caption; cities mentioned in the text must be indicated on the map. Consider using a topographic base map.

 

Response: Figure 1 now includes labeled locations of Urumqi, Changji, Shihezi, Wujiaqu, and other key areas. A topographic background has been added for geographic context, and the caption fully describes the map content.

 

Comment: Line 219: Delete “and Discussion”, as Section 5 is titled “Discussion”.

 

Response: The section title has been changed from “4. Results and Discussion” to “4. Results”.

 

Comment: Line 226: What is meant by “thin atmosphere”?

 

Response: The ambiguous phrase “thin atmosphere” has been removed and replaced with a precise meteorological description (“lower air density”).

 

Comment: All figures must be referenced in the text before they appear.

 

Response: We have ensured that every figure (Figures 1–17) is explicitly cited in the main text prior to its first appearance.

 

Comment: Indicate concentration units (likely µg m⁻³) in all relevant figure captions.

 

Response: The unit “µg m⁻³” has been added to the captions and colorbars of all PM concentration figures (Figures 2–11, 14).

 

Comment: Replace “annual” with “total” average and specify study years in figure captions (e.g., Figure 2).

 

Response: Captions for Figures 2 now state “Spatial distribution of the total average concentrations and ratio of PM2.5 to PM10 over the study period (2019–2022) in the "U-C-S" urban agglomeration” instead of “annual average”.

 

Comment: Use decimal-ordered concentration intervals (e.g., 10, 20, 30…) and arrange from lowest to highest in all concentration maps. Harmonize scales across subfigures.

 

Response: The concentration color scale has been consistently annotated across all figures, clearly indicating the threshold values for PM2.5 and PM10 corresponding to light and heavy pollution levels.

 

Comment: The identified errors have been corrected. Text in lines 242–262 must be corrected accordingly.

 

Response: We identified a data processing error in the original PM₁₀ interpolation for Figure 3. The map has been regenerated using consistent methodology, and the revised spatial pattern now aligns with Figure 2 and physical expectations (higher concentrations in urban centers). The corresponding text (Lines 306–307) has been updated.

 

Comment: Figure 5 caption: Delete “changes”; harmonize y-axis scales across subpanels.

 

Response: The word “changes” has been removed from the caption. Y-axis scales are now uniform across all seasonal subplots for direct comparability.

 

Comment: Lines 382–389 & Figure 9: Text discusses PM₂.₅ weekday patterns but omits PM₁₀ (which peaks on Sundays). This is noted later (Lines 442) but should be consistent.

 

Response: Relevant content has been added at line 442.

 

 

Comment: Line 412: Write “As illustrated in Fig. 10”.

 

Response: Corrected as requested.

 

Comment: Line 441: Add “(Fig. 11)” at the end of the sentence.

 

Response: Added.

 

Comment: Figure 12: Add a legend explaining trajectory colors (which color corresponds to which cluster number).

 

Response: A clear color-to-cluster legend has been added to Figure 12.

 

Comment: Line 493: Indicate “(Fig. 13)” at the end of the sentence.

 

Response: Added.

 

Comment: Lines 495–496: Clarify what “trend” means and explain what low/high WPSCF values indicate.

 

Response: The term "trend" has been removed and replaced with a more accurate expression, and the meaning of high WPSCF values has been clarified in the Methods section (line 191)..

 

Comment: Figure 13: Use the same WPSCF scale across all panels.

 

Response: The WPSCF color scale is now identical in all subfigures of Figure 13.

 

Comment: Line 529: Sentence disrupted; “Fig” must be deleted. A reference to Fig. 14 is necessary.

 

Response: The typo “contributionFig” has been corrected, and a proper citation to Figure 14 has been inserted.

 

Comment: Lines 531–533: Clarify “consistent pattern”.

 

Response: It has been replaced with a more precise expression to avoid ambiguity.

 

 

Comment: Line 534: WCWT appears without definition. What is the difference between CWT and WCWT?

 

Response: We now clarify: “WCWT (Weighted Concentration Weighted Trajectory) is an enhanced version of CWT that incorporates frequency weighting to reduce bias from low-trajectory-density regions.”

 

Comment: Figure 14 caption: Specify whether it’s CWT or WCWT; explain why units are µg m⁻³; interpret low/high values.

 

Response: Caption revised to: “Concentration-weighted trajectory (WCWT) distribution (unit: µg m⁻³), where higher values indicate air mass pathways associated with elevated PM concentrations at the receptor site.”

 

Comment: Line 571: “Wushaer”—please explain or correct.

 

Response: Thank you for pointing this out—this was a typographical error and has now been corrected..

 

Comment: Line 572: Refer to Figs. 15 and 16; explain different symbols.

 

Response: It was originally formatted that way for better distinction, but has now been standardized to a consistent format.

 

Comment: Line 595: “the third figure” → likely Figure 17.

 

Response: Corrected to “Figure 17”.

 

Comment: Conclusions: Include a summary table showing highest or range of concentrations at each spatial/temporal scale.

 

Response: A new Table 3 has been added to the Conclusions section.

 

Comment: Check references for consistency; e.g., “[J]” in refs. 34 and 35—what does this mean?

 

Response: The “[J]” markers (indicating journal article type in some Chinese reference styles) have been removed. All references now strictly follow the journal’s required format.

 

Once again, we deeply appreciate the reviewer’s meticulous reading and valuable suggestions. The manuscript has been substantially improved thanks to this feedback.