Measurement report: Characterizing O<sub>3</sub>-NOₓ-VOC sensitivity and O<sub>3</sub> formation in a heavily polluted central China megacity using multi-methods during 2019&ndash;2021 warm seasons

Yu, Shijie; Liu, Hongyu; Wang, Hui; Su, Fangcheng; Yuan, Minghao; Zhang, Ruiqin

doi:10.5194/egusphere-2025-4519

Preprints

https://doi.org/10.5194/egusphere-2025-4519

Preprints

23 Oct 2025

| 23 Oct 2025

Measurement report: Characterizing O₃-NOₓ-VOC sensitivity and O₃ formation in a heavily polluted central China megacity using multi-methods during 2019–2021 warm seasons

Shijie Yu, Hongyu Liu, Hui Wang, Fangcheng Su, Minghao Yuan, and Ruiqin Zhang

Abstract. This study investigated the high ozone pollution in Zhengzhou City from 2019 to 2021 using observational data and model simulations, focusing on volatile organic compound (VOC) pollution and its impact on ozone formation. Using online VOC data and statistical analyses, the results showed that VOC concentration increased with ozone pollution level, with average values of 84.7±51.0, 96.6±53.4 and 105.3±59.4 µg/m³ for non-pollution, mildly polluted and moderately polluted periods, respectively. Source apportionment of ozone and its precursor VOCs was performed using CMAQ and PMF models. The results demonstrated that reducing vehicle emissions should be prioritized to mitigate ozone pollution in Zhengzhou, as ransportation emissions accounted for 64 % and 31 % of ozone and VOC emissions, respectively. In addition, local ozone production rates and HOx base budgets were calculated using an observation-based model (OBM). The ozone production rates on non-pollution, mildly polluted, and moderately polluted days were respectively 2.0, 4.5, and 6.9 ppbv/h on average. The HOx radical concentration on polluted days was 1.5–6.4 times higher than that on non-pollution days, which is indicative of more efficient radical cycling during photochemical pollution. The O3-NOx-VOC sensitivity was analyzed using the OBM model, CMAQ model and ratio method. The results showed that ozone generation in Zhengzhou was mainly limited by VOCs, suggesting that the reduction of VOCs should be focused on aromatic hydrocarbons and olefins. The optimal reduction ratio of anthropogenic VOCs to NOx was about 2.9:1. This study will offer deeper insights for formulating effective ozone pollution prevention and control strategies.

Received: 14 Sep 2025 – Discussion started: 23 Oct 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 2080 KB)

Supplement (1071 KB)

Download & links

Shijie Yu, Hongyu Liu, Hui Wang, Fangcheng Su, Minghao Yuan, and Ruiqin Zhang

Status: final response (author comments only)

RC1:
'Comment on egusphere-2025-4519', Anonymous Referee #1, 27 Nov 2025
This manuscript investigates O₃–NOx–VOC sensitivity and O₃formation mechanisms in Zhengzhou (2019–2021) using a combination of online VOC measurements, OBM, CMAQ-DDM, source apportionment, PMF, and machine-learning (ML)/SHAP interpretation. While the dataset is valuable and the research direction is meaningful, the manuscript suffers from inconsistent methodology, unclear descriptions of model configurations, uncertainties and machine learning, no comparisons between different analytical models for the O₃ formation mechanism, and very vague data interpretation. The manuscript is long and unreadable. Therefore, I suggest the manuscript be rejected.
Major Comment:
In Section 3, “Results and Discussion”, there is a substantial focus on O₃ source apportionment (CMAQ) and VOCs source apportionment (PMF). Unfortunately, how these contents are linked to the research question “O₃-NO_x-VOC sensitivity” is weak. For example, the authors claim that traffic and industry dominate both O₃ and VOCs. But there is no justification for “X% of O₃ from traffic” and “how O₃ responds if traffic NO_x/VOCs are reduced”.

In addition, there is no cross-method comparison. A combined table is highly recommended to show the section contributions across PMF, CMAQ-DDM, and OBM RIR/EKMA. As a result, the manuscript reads like a report by stacking results (sensitivity diagnostics + VOC and O₃ source apportionment + ML/SHAP), but with limited discussion.
The CMAQ simulation suggests that transportation and industry dominate MDA8 O₃ contributions, while the PMF result indicates that vehicle, solvent, and industrial sources contribute most VOCs. The current discussion is very vague: “transportation should be prioritized,” “more aggressive control is required,” etc. However, what is the take-home message for policy translation? There are no recommended emission-reduction scenarios (e.g. -30% traffic VOCs, -30% industrial NO_x, different VOC/NO_x ratios by sector) and no quantitative estimate of the expected O₃ under those scenarios.

Machine-learning setup and data leakage risk are not clearly addressed. The authors use XGBoost and RF with grid search and cross-validation (CV), and report training/test metrics in Table S4. However, essential methodological details are missing. How are training and test sets defined—random split or chronological split for this time-series dataset? What CV scheme is used (k-fold, blocked time-series CV)? Does CV account for temporal ordering?

What is the inherent relationship between the VOCs/NO_x ratio method for determining O₃sensitivity, modelling results from OBM and CMAQ, and radical budgets?

How do the results of XGBoost relate to those of other methods? Why is XGBoost analysis not mentioned at all in the Section 4 Summary and Conclusions? How do the authors come up with the claim that the ML method should be advised in the future in the subsection Limitations and future research directions in Line 986?

The role of biogenic VOCs is mentioned but not fully clarified in terms of O₃ formation and control. Biogenic VOCs are explicitly identified as a separate PMF factor with isoprene as tracer, and OBM RIR shows that biogenic VOCs have non-negligible reactivity, especially on polluted days. Emissions used in CMAQ also include biogenic VOCs via MEGAN. On the one hand, the authors claim that the contribution of biogenic sources to VOC is small (Line 724). On the other hand, they claim that the contribution of biogenic VOCs to local O₃ is high (Lines 840-850). So, I wonder what the overall role of biogenic VOCs in the O₃–NO_x–VOC sensitivity and O₃ formation mechanism is.

Minor Comment:
The period simulated by the CMAQ model was not mentioned in the methodology.

What is the rationale for studying warm seasons? Could you provide any justification and supporting references?

Lines 141-143: Could you provide any reference for the study city of Zhengzhou?

Lines 212-215: “PO₃^S(X) and PO₃^S (X-DX) refer to the simulated O₃ yields…” Are you sure “yield” is the right terminology here? Why does the author later state that “The net O₃ production rate (PO₃^S)” in Lines 215-216?

The CMAQ result shows transportation can contribute 64% to O₃(Fig.4), but the PMF result shows vehicles contribute only 31% to VOCs. This apparent discrepancy needs justification. Presumably, VOCs and NOₓ emissions from transportation are equally important for O₃ formation. Therefore, the role of NO_x emissions needs to be discussed.

Why does Table S5 show a negative correlation (Kendall’s, -0.305) between VOCs and O3, while the RIR indicates a positive correlation (Fig.9)? Shouldn’t this discrepancy require some explanations?

Table S4: The rationale for selecting XGBoost over RF is well justified.

A strong positive correlation between O₃ and alkenes is observed in SHAP (Fig.2), whereas Spearman's analysis reveals the opposite relationship (Table S6, -0.054**). Please standardise the terminology for “olefins” (Fig.2) or “Alkene” (Table S6)!

Line 427: “160 160 µg/m³ and 160 µg/m³ are categorised as light pollution and moderate pollution”? Could you please carefully check if this is CORRECT?

9: What are “AVOCs”, “ALKA”, “ALKE”, “ARO”? What is the relationship between them? Which subsets of compounds belong to AVOCs? Why is there no further discussion of the contribution and sensitivity of VOCs from different sources to O₃?

Technical Comment:
2 The font is too small to read.

Line 351: What is the full name of “XGBoost”?

Line 386: Should it be “SHAP” instead of “Shapley”?

Table S3: What do “DISP” and “BS” mean? What is the meaning of BS mapping＜80%?

S3: A full description needs to be provided for “E/X” in the manuscript for the first time and figure captions.

Line 712: “Zhengzhou comprises vehicle emissions (31%), solvent use (24%), and industrial processes (21%)”. But in Table S8, the corresponding values are 32.4%, 24.8% and 18.3%.

Lines 1135-1137: Make sure the font style is consistent throughout.

Please be consistent in the use of subscripts and full names or abbreviations for proper nouns.
Citation: https://doi.org/10.5194/egusphere-2025-4519-RC1
- AC3:
  'Reply on RC1', Yu Shijie, 01 Mar 2026
  Itemized Response to Editor's Comments
  Ms. Ref. No.: EGUSPHERE-2025-4519 | Measurement report
  Title: Measurement report: Characterizing O3-NOₓ-VOC sensitivity and O3 formation in a heavily polluted central China megacity using multi-methods during 2019–2021 warm seasons
  
  This manuscript investigates O₃–NOx–VOC sensitivity and O₃ formation mechanisms in Zhengzhou (2019–2021) using a combination of online VOC measurements, OBM, CMAQ-DDM, source apportionment, PMF, and machine-learning (ML)/SHAP interpretation. While the dataset is valuable and the research direction is meaningful, the manuscript suffers from inconsistent methodology, unclear descriptions of model configurations, uncertainties and machine learning, no comparisons between different analytical models for the O₃ formation mechanism, and very vague data interpretation. The manuscript is long and unreadable. Therefore, I suggest the manuscript be rejected.
  Response: We sincerely thank you for your careful reading of our manuscript and for providing valuable comments and constructive suggestions. In response to the concerns raised—including inconsistent methodology, unclear descriptions of model configurations, lack of uncertainty analysis, vague machine learning interpretation, absence of cross-model comparisons regarding O₃ formation mechanisms, and ambiguous data interpretation—we have thoroughly revised the manuscript.
  Below are our point-by-point responses to all comments (reviewer comments are in black font; our responses are in dark blue font). Major revisions made in response to these comments are highlighted in yellow in the marked-up version of the revised manuscript. Minor revisions made at our own initiative are indicated in red font. Please note that line numbers referenced below correspond to those in the revised manuscript.
  
  Major Comment:
  In Section 3, “Results and Discussion”, there is a substantial focus on O3 source apportionment (CMAQ) and VOCs source apportionment (PMF). Unfortunately, how these contents are linked to the research question “O3-NOx-VOC sensitivity” is weak. For example, the authors claim that traffic and industry dominate both O3 and VOCs. But there is no justification for “X% of O3 from traffic” and “how O3 responds if traffic NOx/VOCs are reduced”.
  
  In addition, there is no cross-method comparison. A combined table is highly recommended to show the section contributions across PMF, CMAQ-DDM, and OBM RIR/EKMA. As a result, the manuscript reads like a report by stacking results (sensitivity diagnostics + VOC and O3 source apportionment + ML/SHAP), but with limited discussion.
  Response: We sincerely appreciate the reviewer's critical feedback regarding the logical flow and the "stacking" of results. We agree that a stronger connection between source apportionment and sensitivity analysis is essential to elevate the manuscript's academic depth. Following your constructive suggestions, we have performed a fundamental restructuring of Section 3 to establish a clear "Source–Reactivity–Mechanism" narrative. Our revisions are detailed as follows:
  （1）Restructuring the Logical Framework (Source–Reactivity–Mechanism)
  To transform the manuscript from a descriptive report into a mechanistic study, we have reorganized the discussion to follow a more rigorous scientific logic:
     Observation as the Foundation (Section 3.1): We have removed the machine learning (ML/SHAP) components to reduce complexity and focus on atmospheric chemistry. Crucially, we added a "Weekend Effect" analysis as direct observational evidence. This "natural experiment" provides solid, model-independent proof that the study area is in a VOC-limited regime, serving as the basis for subsequent sensitivity diagnostics.
     Linking Sources to Impact (Section 3.2): We established a "Mass–Reactivity–Contribution" bridge. Previously, Section 3.2 only discussed VOC mass; we now incorporate Ozone Formation Potential (OFP) as a vital link. This explains why traffic and industry dominateO3 formation: while they contribute significantly to VOC mass (PMF), their high concentration of reactive species (e.g., alkenes and aromatics) leads to an even higher share of OFP (e.g., 35% for traffic). This justifies the finalO3 source apportionment results from CMAQ.
     Sensitivity as the Core (Sections 3.3 & 3.4): These sections have been merged to avoid result-stacking. We now use the source apportionment results to inform the sensitivity analysis, directly addressing "howO3 responds" to specific source reductions.
  （2）Cross-Method Comparison and Validation
  We have included a Combined Table in Section 3.2 as recommended. This table horizontally aligns the mass contributions (PMF), reactivity contributions (OFP), and O3 contributions (CMAQ-ISAM). By cross-validating results from these different tools, we have significantly enhanced the reliability of our conclusions and provided a clearer justification for the dominance of specific sectors.
  （3） Quantifying Policy-Relevant Information
  Regarding the reviewer's concern about "howO3 responds," we have replaced vague suggestions with a specific "Chemical Red Line." By extracting quantitative data from the EKMA isopleths and DDM sensitivity coefficients, we identified a critical VOC/NOx reduction ratio of 2.9:1. We clarify that to achieve a netO3 decrease, VOC emissions must be reduced at a rate at least 2.9 times that of NOx. This provides policymakers with a concrete, science-based target.
  We believe this restructured logic significantly strengthens the connection between our findings and addresses the reviewer's concerns regarding the depth of discussion.
  The CMAQ simulation suggests that transportation and industry dominate MDA8 O₃ contributions, while the PMF result indicates that vehicle, solvent, and industrial sources contribute most VOCs. The current discussion is very vague: “transportation should be prioritized,” “more aggressive control is required,” etc. However, what is the take-home message for policy translation? There are no recommended emission-reduction scenarios (e.g. -30% traffic VOCs, -30% industrial NOx, different VOC/NOx ratios by sector) and no quantitative estimate of the expected O3 under those scenarios.
  
  Response: We sincerely appreciate the reviewer's suggestion regarding the need for more concrete policy-relevant insights. We agree that translating scientific findings into actionable management strategies is crucial.
  In response to your concerns, while we have not added new hypothetical emission-reduction scenarios (such as -30% reduction simulations), we have significantly refined the discussion in Section 3.2 and Section 4 to provide more specific and quantitative "take-home messages" based on our existing data-driven results. Our reasoning and the subsequent improvements are as follows:
  (1) Quantitative Prioritization via the "Mass-Reactivity-Contribution" Chain
  Instead of relying on vague recommendations, we now utilize the re-evaluated logic in Section 3.2 to provide a semi-quantitative basis for policy. By linking PMF (mass), OFP (reactivity), and CMAQ (contribution), we explicitly demonstrate why the transportation sector is the primary target. For instance, the fact that traffic VOCs account for the largest share of OFP (35%)—coupled with its high NOx emissions—provides a data-backed justification for prioritized control, rather than just a general suggestion.
  (2) Utilizing Sensitivity Coefficients (RIR) as Quantitative Indicators
  We have enhanced the discussion of the Relative Incremental Reactivity (RIR) and EKMA results in Section 3.3. These results already provide the quantitative sensitivity ofO3 to its precursors (i.e., theO3 response per unit change in precursors). By emphasizing these coefficients, we provide a quantitative measure of reduction benefits without the need for additional, and potentially uncertain, scenario simulations. This aligns with our focus on a "data-driven" investigation rather than a purely model-oriented sensitivity study.
  (3) Clarification of Research Scope and Focus
  We have clarified in the manuscript that this study is designed as a detailed diagnostic investigation based on high-resolution observations and source apportionment. Our goal is to provide a robust physical and chemical diagnosis of currentO3 pollution. We believe that identifying the "key drivers" through cross-method verification provides a more reliable foundation for policy-making in this specific study area than simulating hypothetical scenarios that may involve high degrees of uncertainty in emission inventory scaling.
  (4) Refined Policy Recommendations
  The revised Section 4 now provides more granular advice. For example, instead of "prioritizing transportation," we now specify that control efforts should focus on high-reactivity VOC species identified in the traffic and industrial sectors (e.g., alkenes and aromatics), as these are proven to be the most "cost-effective" targets based on our OFP and RIR analysis.
  We believe these revisions make the policy implications of our work much clearer and more grounded in the presented data. Thank you for pushing us to sharpen the practical value of our research.
  Machine-learning setup and data leakage risk are not clearly addressed. The authors use XGBoost and RF with grid search and cross-validation (CV), and report training/test metrics in Table S4. However, essential methodological details are missing. How are training and test sets defined—random split or chronological split for this time-series dataset? What CV scheme is used (k-fold, blocked time-series CV)? Does CV account for temporal ordering?
  
  We would like to express our sincere gratitude to the reviewer for these highly professional and insightful comments regarding our machine learning (ML) methodology. Your concerns regarding the potential for data leakage in time-series datasets and the necessity of a robust cross-validation (CV) scheme are technically well-founded and crucial for ensuring the reliability of such models.
  After careful consideration of your feedback and re-evaluating the core objectives of this manuscript as a "Measurement Report," we have decided to remove the machine learning (XGBoost and Random Forest) analysis from the revised manuscript. Our decision is based on the following considerations:
  (1) Refocusing on Observational Data and Chemical Mechanisms: As a "Measurement Report," the primary value of this study lies in its high-resolution observational dataset and the underlying atmospheric chemical processes. We found that while the ML models provided supplementary quantitative insights, they functioned as a "black box" that was less intuitive for explaining specific chemical mechanisms compared to the OBM, PMF, and OFP analyses.
  (2) Streamlining the Manuscript Logic: In line with the reviewer's suggestion to clarify the narrative, removing the ML section allows the manuscript to be more concise and focused. This "subtractive" approach eliminates the methodological uncertainties and potential risks associated with ML setups (such as the data leakage you correctly identified), thereby strengthening the overall transparency and physical interpretability of our findings.
  (3) Prioritizing Physico-chemical Diagnosis: By focusing on the "Mass-Reactivity-Contribution" bridge and the OBM-based sensitivity analysis, we ensure that our conclusions are grounded in robust, physically-based atmospheric chemistry rather than purely statistical associations.
  We believe that this major revision significantly improves the clarity and focus of the manuscript, ensuring it aligns more closely with the expectations for an atmospheric measurement investigation. Thank you again for your rigorous review, which helped us identify this opportunity to refine our work.
  What is the inherent relationship between the VOCs/NOx ratio method for determining O3 sensitivity, modelling results from OBM and CMAQ, and radical budgets?
  
  Response: We sincerely appreciate the reviewer's insightful comment. This question touches upon the theoretical core of our study—the consistency between microscopic chemical mechanisms and macroscopic atmospheric observations.
  In response, we have substantially restructured the manuscript by merging the original Section 3.3 (Radical chemistry) and Section 3.4 (O3 sensitivity) into a single, integrated discussion. This allows us to explicitly demonstrate the inherent relationships through a "Diagnostic Chain" moving from "Preconditions → Internal Mechanisms → Local Diagnostics → Regional Validation."
  The specific relationships are articulated as follows:
  （1） Radical Budgets: The Microscopic "Engine" (The Mechanistic Foundation)
  The radical budget is the fundamental driver of O3 formation. Our analysis shows that radical loss is dominated by the OH + NO2 reaction (L_N > L_R), which directly dictates the chemical regime. This microscopic imbalance—where NO2 outcompetes VOCs for OH radicals—is the primary reason why the system is sensitive to VOC increments.
  VOCs/NOx Ratio: The Macro-scale "Indicator" (The Environmental Background)
  
  The VOCs/NOx ratio represents the atmospheric chemical environment that sets the stage for radical competition. The observed low ratio (< 10 ppbC/ppbv) provides the macro-scale context for the NOx-saturated conditions found in the radical budget. It serves as a preliminary proxy that is consistent with the more detailed kinetic analysis.
  
  OBM (RIR/EKMA): The Local "Scalpel" (The Quantitative Diagnostic)
  
  The OBM translates the complex radical cycling (captured via MCM mechanism) into quantitative sensitivity metrics (RIR and EKMA). It acts as a local diagnostic tool that answers how the O3 "engine" responds to specific precursor changes based on the in-situ radical cycling efficiency.
  CMAQ (DDM): The Regional "Laboratory" (The Spatial Validation)
  
  While the above methods focus on local chemistry, the CMAQ-DDM provides a 3D perspective considering transport and regional emissions. The consistency between CMAQ results and our OBM/radical budget findings confirms that the radical-driven VOC-limited regime observed at the monitoring site is representative of the broader urban area of Zhengzhou.
  Summary of Integration:
  By restructuring Section 3.3, we emphasize that these four methods are not independent metrics but a coherent evidence chain. The consensus among the initial conditions (VOCs/NOx ratio), internal kinetics (radical budgets), local response (OBM), and regional manifestation (CMAQ) robustly supports our conclusion regarding the VOC-limited regime and the proposed 2.9:1 reduction ratio.
  We hope these clarifications and the integrated section better address the reviewer's concern. The revised text can be found in the updated Section 3.3.
  How do the results of XGBoost relate to those of other methods? Why is XGBoost analysis not mentioned at all in the Section 4 Summary and Conclusions? How do the authors come up with the claim that the ML method should be advised in the future in the subsection Limitations and future research directions in Line 986?
  
  Response: We are extremely grateful to the reviewer for these critical observations. Your assessment that the XGBoost analysis felt disconnected from the core chemical mechanisms and was not adequately integrated into the conclusions is very accurate.
  
  Upon careful reflection and in alignment with your previous comments regarding "data leakage" and methodological clarity, we have decided to completely remove the machine learning (ML) component (including XGBoost and Random Forest) from the revised manuscript. Our reasoning and the subsequent changes are as follows:
  （1）Resolving the Logical Disconnect:
  We acknowledge that the ML results in the previous version acted as a "separate layer" that did not harmonize well with the traditional physico-chemical models (OBM and PMF). As you noted, this lack of integration led to a narrative where the ML analysis seemed isolated. By removing this section, we have eliminated this logical "two-skin" problem, ensuring that the manuscript remains strictly focused on atmospheric chemical diagnosis and source apportionment.
  （2）Adhering to the "Measurement Report" Focus:
  The primary value of this study lies in its high-resolution observational data and the mechanistic insights derived from established chemical models. Removing the "black-box" ML analysis allows us to emphasize the empirical evidence and chemical pathways, which are the core requirements of an ACP Measurement Report.
  （3）Correcting the Conclusions and Future Research Directions:
  In response to your specific question regarding Line 986, we have deleted the claim recommending ML methods in the "Limitations and future research" section. We have also ensured that the Summary and Conclusions (Section 4) are now entirely consistent with the remaining data-driven and mechanism-based results.
  We believe these "subtractive" revisions have significantly strengthened the internal logic and professional focus of the paper. We sincerely thank the reviewer for identifying these inconsistencies, which allowed us to refine the manuscript into a more coherent and rigorous scientific report.
  The role of biogenic VOCs is mentioned but not fully clarified in terms of O3 formation and control. Biogenic VOCs are explicitly identified as a separate PMF factor with isoprene as tracer, and OBM RIR shows that biogenic VOCs have non-negligible reactivity, especially on polluted days. Emissions used in CMAQ also include biogenic VOCs via MEGAN. On the one hand, the authors claim that the contribution of biogenic sources to VOC is small (Line 724). On the other hand, they claim that the contribution of biogenic VOCs to local O3 is high (Lines 840-850). So, I wonder what the overall role of biogenic VOCs in the O3–NOx–VOC sensitivity and O3 formation mechanism is.
  
  Response: We sincerely thank the reviewer for this insightful observation. The perceived contradiction regarding the "small mass contribution" versus "high ozone contribution" of biogenic VOCs (BVOCs) is a critical point that requires further clarification. We have thoroughly revised the manuscript to address the "Mass vs. Reactivity" paradox and to better elucidate the role of BVOCs in local photochemistry. Our response and the corresponding revisions are detailed below:
  （1） Resolving the "Mass vs. Reactivity" Paradox
  We acknowledge that the previous version did not sufficiently distinguish between mass-based and reactivity-based contributions. While our PMF results indicate that BVOCs (primarily isoprene) contribute a relatively small fraction to the total VOC mass concentration (Line 724)—largely due to their extremely short chemical lifetimes and rapid consumption—their impact on ozone formation is disproportionately large.
  In the revised manuscript, we have introduced the Propylene-equivalent concentration (Prop-equiv) and Ozone Formation Potential (OFP) metrics to quantify this effect. Our new analysis demonstrates that although isoprene ranks low in mass, it exhibits the highest Prop-equiv concentration among all sources due to its exceptionally high OH radical reactivity (k_OH). This high reactivity explains why BVOCs are identified as major contributors to localO3 in our OBM and CMAQ simulations (Lines 840-850).
  （2） The Overall Role of BVOCs in O3 Formation and Sensitivity
  We have added a new discussion in Section 3.3 to clarify the role of BVOCs in theO3–NOx–VOC sensitivity:
  As a "Photochemical Fuel": BVOCs provide highly reactive substrates that significantly accelerate the photochemical cycle, especially during the high-temperature and high-radiation periods typical of polluted days.
  
  As a "Sensitivity Shifter": The presence of strong biogenic reactivity ensures that the atmosphere remains highly reactive even when anthropogenic VOCs (AVOCs) are relatively low. This background reactivity makes localO3 production more sensitive to further reductions in AVOCs, particularly high-reactivity species from industrial and transportation sources.
  
  （3） Clarification and Consistency in the Manuscript
  To ensure logical consistency, we have revised the language throughout the text:
  In Section 3.2 (formerly Line 724), we now specify that "BVOCs have a limited contribution to the total VOC mass burden, but play a dominant role in chemical reactivity."
  
  We have included a comparison in the revised manuscript (Table/Figure) showing the contrast between mass percentage and Prop-equiv percentage for different sources to provide a visual confirmation of this "low abundance, high activity" phenomenon.
  
  By removing the machine learning components as suggested in your other comments, we have dedicated more space to a rigorous chemical diagnosis of these biogenic-anthropogenic interactions.
  
  We believe these revisions provide a much clearer and more scientifically robust explanation of the biogenic contribution to ozone formation in the study area. We are grateful for the opportunity to refine this aspect of our work.
  
  Minor Comment：
  The period simulated by the CMAQ model was not mentioned in the methodology.
  
  We apologize for the omission of the simulation period in the initial manuscript and thank the reviewer for this careful observation. We have now added the specific details to the Methodology section (Section 2.3).
  The CMAQ simulation was conducted for June 2019. The selection of this specific period and year is based on the following considerations:
  （1） Representativeness of Peak Pollution: June was chosen because it represents the most severe ozone (O3) pollution period in the study area (Zhengzhou). Simulating this window allows for a more robust analysis of the chemical mechanisms under peak pollution conditions.
  （2） Reliability and Consistency: 2019 was selected as the base year for the simulation. This year has been extensively validated and used as a benchmark in several peer-reviewed studies previously published by our research group. Utilizing this well-established baseline ensures the credibility and consistency of our modeling results with existing regional emission and meteorological datasets.
  In the revised manuscript, we have updated the "Model Configuration" part of the Methodology section as follows:
  "The CMAQ model simulations were performed for June 2019, which coincides with the peak of O3 pollution in Zhengzhou. This year was selected as the base year to maintain consistency with our group's previously validated regional studies, ensuring the reliability of the emission inventory and meteorological inputs."
  We hope this addition clarifies the scope of our modeling work.
  What is the rationale for studying warm seasons? Could you provide any justification and supporting references?
  
  Response: We sincerely appreciate the reviewer's constructive suggestion regarding the temporal scope of our study. The rationale for focusing on the warm season (May to September) is based on the following considerations:
  （1） Peak Period ofO3 Pollution in Zhengzhou
  The selection of May to September is primarily driven by the seasonal characteristics of air quality in Zhengzhou. Statistical monitoring data and our previous research indicate that O3 exceedances and high-concentration events are heavily concentrated during these five months, which coincide with high solar radiation and temperatures—key drivers for photochemical reactions.
  （2） Foundation of Previous Research
  This study builds upon a series of investigations conducted by our research group in the North China Plain. Our prior published works have consistently identified May–September as the most critical window for O3 mitigation in Zhengzhou. Focusing on this period ensures that our findings are highly relevant to the most severe pollution episodes, providing a solid scientific basis for the current analysis.
  Categorization of Pollution Levels
  
  The core objective of this study is to explore the dynamic relationship between VOCs andO3 under varying atmospheric conditions. By focusing on the high-frequencyO3 season, we can effectively categorize the data into different pollution levels (e.g., clean vs. polluted days). This allows for a more nuanced understanding of how precursor sensitivity shifts as pollution intensifies, which is a central theme throughout the manuscript.
  We have incorporated these justifications and the suggested references into the Introduction section as follows:
  "Although the Central Plains urban cluster centered on Zhengzhou has achieved significant progress in controlling primary pollutants, the paradoxical situation of persistently rising ozone levels still exists in this area, particularly characterized by the high frequency of O3 pollution episodes from May to September (Yu et al., 2020 and 2021). This persistent challenge necessitates an urgent investigation into its formation mechanisms during these critical periods (Jia et al., 2024; Li et al., 2020; Min et al., 2022; Yu et al., 2021)."
  We hope these clarifications and the added references sufficiently address the reviewer's concern.
  Lines 141-143: Could you provide any reference for the study city of Zhengzhou?
  
  Response: We thank the reviewer for this reminder. The air quality ranking mentioned in the manuscript is based on the official statistics released by the Ministry of Ecology and Environment (MEE) of the People's Republic of China. We have added the formal citation for this report.
  Additionally, to provide a more comprehensive academic context for Zhengzhou's severe pollution status, we have also cited peer-reviewed literature characterizing the historical and persistent air quality challenges in this region (e.g., Wang et al., 2021; Yu et al., 2021).
  The text has been updated as follows:
  Zhengzhou... faces severe air pollution, with its air quality ranking among the bottom twenty of 168 major cities in China from January to September 2024 (MEE, 2024; Wang et al., 2021).
  Lines 212-215: “PO3S(X) and PO3S (X-DX) refer to the simulated O3 yields…” Are you sure “yield” is the right terminology here? Why does the author later state that “The net O3 production rate (PO3S)” in Lines 215-216?
  
  Response: We apologize for the imprecise terminology. "Yield" has been replaced with "net O₃ production rate" (P(O₃), ppbv/h) throughout the manuscript to maintain consistency with the OBM results and Equation (2).
  The CMAQ result shows transportation can contribute 64% to O3 (Fig.4), but the PMF result shows vehicles contribute only 31% to VOCs. This apparent discrepancy needs justification. Presumably, VOCs and NOₓ emissions from transportation are equally important for O₃ formation. Therefore, the role of NOx emissions needs to be discussed.
  
  Response: The discrepancy arises because PMF measures mass contribution, while CMAQ calculates O3 impact. As shown in the newly added OFP and Prop-equiv analysis, transportation-derived VOCs have disproportionately high reactivity. Furthermore, transportation is a major source of NOx. In the VOC-limited regime of Zhengzhou, the synergistic impact of NOx and reactive VOCs from traffic leads to the higher O₃ contribution (64%).
  Why does Table S5 show a negative correlation (Kendall's, -0.305) between VOCs and O3, while the RIR indicates a positive correlation (Fig.9)? Shouldn't this discrepancy require some explanations?
  
  We appreciate the reviewer's insightful observation regarding the apparent discrepancy between the negative statistical correlation (Table S5) and the positive RIR values (Fig. 9). This distinction is indeed critical for a nuanced understanding of atmospheric photochemistry. We have added a detailed explanation in the revised manuscript to clarify this point.
  Response: Our response and the rationale for this phenomenon are summarized below:
  （1） Statistical Correlation (Concentration) vs. Chemical Sensitivity (Production Rate)
  The negative Kendall's correlation in Table S5 reflects the relationship between the observed concentrations of VOCs and O3, which are influenced by the complex interplay of physical transport, boundary layer dynamics, and chemical transformation. In contrast, the RIR (Fig. 9) represents the chemical sensitivity of the O3 production rate to its precursors. It specifically answers how the net photochemical formation of O3 changes in response to a perturbation in precursor levels, independent of physical dilution.
  （2） The Influence of Boundary Layer Dynamics (Physical Factor)
  The negative correlation in concentration is largely driven by the diurnal evolution of the planetary boundary layer (PBL). During the early morning, the shallow PBL traps primary emissions, leading to peak VOC concentrations, while O3 remains low due to the lack of sunlight and NO titration. Conversely, in the afternoon, the rising PBL height dilutes VOC concentrations, while intense solar radiation triggers rapid photochemical O3 formation, leading to an O3 peak. This "seesaw" effect inherently results in a negative statistical correlation between their ambient concentrations.
  （3） Photochemical Consumption of Precursors (Chemical Factor)
  During periods of high O3 production, the concentration of OH radicals typically reaches its peak. These radicals rapidly consume VOCs through oxidation. Consequently, the period of maximum O3 production often coincides with a significant depletion of VOCs. This chemical "sink" further contributes to the observed negative correlation between the two species in ambient air.
  （4） Consistency with the VOC-limited Regime
  The positive RIR values indicate that the study area is in a VOC-limited regime, where increasing VOCs enhances O3 formation potential. The negative correlation does not contradict this; rather, it highlights that the observed concentrations are dominated by physical processes and consumption, while the RIR captures the underlying chemical driving force.
  We believe these clarifications address the reviewer's concern and strengthen the interpretation of our results.
  （7）&(8) Table S4: The rationale for selecting XGBoost over RF is well justified.
  A strong positive correlation between O3 and alkenes is observed in SHAP (Fig.2), whereas Spearman's analysis reveals the opposite relationship (Table S6, -0.054). Please standardise the terminology for “olefins” (Fig.2) or “Alkene” (Table S6)
  Response: We appreciate the reviewer's insightful observation. 1. Regarding the discrepancy between SHAP and Spearman's results: To ensure the internal consistency and maintain the focus of a Measurement Report, we have decided to entirely remove the Machine Learning (ML) section (including the XGBoost model and SHAP analysis). We agree that the contradiction between SHAP and statistical analysis highlights the uncertainties of the ML model in this context. Removing this section allows the manuscript to focus on more robust observational and OBM-RIR results. 2. Regarding terminology standardization: We have standardized the terminology to "Alkene" throughout the manuscript and supplementary materials, replacing "olefins" as suggested. Thank you for your constructive guidance, which has helped us simplify and strengthen the manuscript.
  Line 427: “160 160 µg/m³ and 160 µg/m³ are categorised as light pollution and moderate pollution”? Could you please carefully check if this is CORRECT?
  
  Response: We apologize for the typo. The text has been corrected according to the GB 3095-2012 standard: MDA8 O₃ > 160 μg/m³ is light pollution, and > 215 μg/m³ is moderate-to-severe pollution.
  What are “AVOCs”, “ALKA”, “ALKE”, “ARO”? What is the relationship between them? Which subsets of compounds belong to AVOCs? Why is there no further discussion of the contribution and sensitivity of VOCs from different sources to O3?
  
  Response: We appreciate the opportunity to clarify these definitions and the logical structure of our analysis. 1. Definitions and Relationships: We have added explicit definitions in the manuscript. AVOCs refers to Anthropogenic Volatile Organic Compounds, which in this study is the sum of three chemical groups: ALKA (Alkanes), ALKE (Alkenes), and ARO (Aromatics). 2. Discussion on VOC Contribution to O3: To better address the contribution and sensitivity of different VOC groups toO3 formation, we have added the Ozone Formation Potential (OFP) analysis in Section 3.2. This addition bridges the gap between VOC sources andO3 production, resulting in a more coherent logical flow: VOC Source Apportionment → OFP Analysis →O3 Sensitivity (RIR analysis). This sequence provides a comprehensive evaluation of how various VOC precursors and their sources contribute to ozone formation. The corresponding sections and figures have been updated to reflect these clarifications. Thank you for your constructive feedback.
  Technical Comment:
  The font is too small to read.
  
  Line 351: What is the full name of “XGBoost”?
  Line 386: Should it be “SHAP” instead of “Shapley”?
  Response: As detailed in our general response, to streamline the manuscript and focus on the core chemical diagnostics, we have completely removed the machine learning and SHAP analysis sections (including the former Figure 2). Therefore, these specific terminological and formatting issues are no longer applicable.
  Table S3: What do “DISP” and “BS” mean? What is the meaning of BS mapping＜80%?
  Response: We have added full descriptions for these PMF diagnostic terms in the Supplemental Information. BS stands for Bootstrap and DISP stands for Displacement. A BS mapping < 80% (specifically for certain factors) indicates potential rotational ambiguity or instability in those factors; however, we have retained the 6-factor solution because it provided the most physically meaningful source profiles and passed the DISP error estimation without any factor swaps. We have clarified this in the revised Table S3 caption.
  S3: A full description needs to be provided for “E/X” in the manuscript for the first time and figure captions.
  Response: Thank you for your valuable suggestion. We have added the full description of "E/X" at its first mention in the manuscript and in the corresponding figure captions. Specifically, E and X denote ethylbenzene and m,p-xylene, respectively. The E/X ratio is employed as a diagnostic tracer to estimate the photochemical age of air masses, based on the differing reaction rates of these species with OH radicals. We appreciate your careful reading.
  Line 712: “Zhengzhou comprises vehicle emissions (31%), solvent use (24%), and industrial processes (21%)”. But in Table S8, the corresponding values are 32.4%, 24.8% and 18.3%.
  Response: We sincerely apologize for this discrepancy. The values in Table S8 were incorrectly based on a different timeframe (May 2019–September 2021). We have updated Table S8 to ensure full consistency with the results reported in the main text. Thank you for your meticulous review.
  Lines 1135-1137: Make sure the font style is consistent throughout.
  Please be consistent in the use of subscripts and full names or abbreviations for proper nouns.
  Response: We have thoroughly revised the manuscript to ensure consistent font styles (including Lines 1135-1137). A comprehensive check was conducted to standardize the use of subscripts (e.g., O3, NOx) and the consistent application of abbreviations and full names for all proper nouns and chemical species
  
  Citation: https://doi.org/10.5194/egusphere-2025-4519-AC3

RC2: 'Comment on egusphere-2025-4519', Anonymous Referee #2, 15 Dec 2025

In “Measurement report: Characterizing O₃-NOₓ-VOC sensitivity and O₃ formation in a heavily polluted central China megacity using multi-methods during 2019–2021 warm seasons” Yu et al. investigate the sensitivity of O₃ formation in Zhengzhou City between 2019 and 2021 based on VOC in-situ observations, several models and machine learning tools. The authors find that O₃ formation in the Chinese megacity is limited by the availability of VOCs and recommend a focus on VOC emission reductions with simultaneous NO_x control.

While the investigation of O₃ pollution generally remains a highly important topic, I have major concerns regarding the implementation and results presented in this study. The large number of methods, that often seem redundant, the use of many abbreviations (often not defined) and changes between units makes it difficult to follow the line of argument. The lack of presenting NO_x measurements and a detailed discussion on the role of NO_x in O₃ formation makes it additionally challenging to understand how the authors reach their scientific conclusions. Unfortunately, I therefore cannot recommend this manuscript for publication in ACP.

Major Comments:

Manuscript type: The manuscript type “Measurement Report” does not seem appropriate to me. A measurement report should “present substantial new results from measurements of atmospheric properties and processes from field and laboratory experiments” (https://www.atmospheric-chemistry-and-physics.net/about/manuscript_types.html). While it can be accompanied by model results, the focus should be on the presentation of a unique dataset rather than a pool of different methods, including modeling and machine learning tools.
Number of methods: The authors use a large number of different methods to investigate O₃ formation sensitivity. It is not clearly established why all these methods are needed and what their added value is. The questions posed could be answered with a reliable set of in-situ observations and an appropriate model to simulate missing trace gases. Instead, the introduction of all these methods is confusing and makes it difficult to follow the central line of argument.
Measurements: From the results section, I understand that besides VOCs, measurements of NO_x and O₃ are available. However, there are no details presented in the methods section.
NOx: Besides VOCs, Nitrogen Oxides are important precursors to tropospheric O₃. However, the role of NO_x seems to be mostly neglected in this study. A detailed description and discussion of the role of NO_x are missing and I wonder how the authors reached their conclusions on O₃ formation sensitivity, without accounting for NO_x. E.g. the authors state that chemistry is VOC-limited in Zhengzhou, but suggest that VOC control is more important than NO_x. VOC-limited chemistry is characterized by a large excess of NO_x, which requires drastic emission cuts to improve long-term air quality.
Correlation analysis: The presentation of correlations between different parameters in Section 3.1.1 seems random and does not follow a clear strategy, e.g. hypothesis – method – result – discussion. The set of in-situ observations is much more powerful than this: I recommend presenting trace gas levels (and if possible a longer time series), the characteristics of each season, diurnal cycles and the weekend effect for a sensitivity analysis. The application of machine learning tools is not necessary here or needs to be better justified.
Abbreviations: Many abbreviations are used in this manuscript, and they are often not defined upon first use, which makes it difficult to follow. It is further concerning that the authors are in some cases not consistent with the abbreviations, e.g. “OBM” is an “observation-based model” in Line 41 and an “Ozone Box Model” in Line 128.
Units: Many different units for trace gases are used throughout the text, including ppbv, ug/m3 and molecules/cm3. This makes it difficult to compare trace gas levels and I recommend choosing one unit (preferably mixing ratios) and using it throughout the entire manuscript.
PM2.5: Why is PM2.5 relevant to this study? I recommend focusing on O₃ and its precursors to avoid overloading this study.

Minor comments:

33 f.: This sounds like VOCs increase in response to O₃ increases, while VOCs are precursors to O₃.
34 f.: Do these values refer to VOC or O₃ concentrations?
37: Please define abbreviations upon first use.
39: What is meant by “ozone emissions”? Ozone is not emitted but formed photochemically.
46: What’s the “ratio method”?
47: If ozone generation is limited by VOCs, it is highly important to control NO_x. Of course, it remains important to reduce VOCs simultaneously, but long-term air quality improvements can only be reached through NOx reductions in that case.
55 ff.: Several things are missing in the introduction, i.a. how O₃ is formed from its precursors and particularly what the role of NO_x is.
65 ff.: Are the authors saying that they are the first to investigate O₃ formation from increasing anthropogenic sources?
76 f.: This sounds like the range of VOC mixing ratios in China is 27 – 92 ppbv.
81 ff.: This section is difficult to follow due to the jumps between countries and continents.
83: Is BB the major VOC source throughout the entire year?
93 ff.: Are the authors talking about concentrations, emissions, formation rates or sensitivities?
115 ff.: Why not use a set of in-situ observations?
128: In the Abstract the authors state the OBM to be an observation-based model.
188: How were other trace gases and meteo parameters measured?
215: More details on the OBM are required.
223: The reaction of OH and NO₂ does not destroy O₃ but limits its formation. It should therefore be accounted for in Equation (3), rather than (4).
242: More details are needed on the WRF/CMAQ model.
248 ff.: What are all these abbreviations: FNL, SAPRC-99, AERO6, IC/BC, MEIC, REAS2?
262 f.: What are first- and second-order sensitivities of O₃?
265 ff.: What exactly do these equations show?
277: Why is PM2.5 needed in this study?
283: Why would the model be better at simulating emitted species?
284: Only a small fraction of NO2 is emitted directly, most is formed photochemically from NO.
319: What’s MDL?
348: Why exactly is machine learning needed in this study?
425 / Fig. 1: Why is a smoothing applied? What exactly does it involve? Why is the time series not just averaged to the desired resolution?
426 ff.: All three numbers are the same. How exactly are the pollution levels defined?
432: What exactly do the percentage values relate to? If it’s years, the time period is too short for a trend analysis.
437: Why is O₃ positively correlated with wind speed? Usually, higher wind speeds lead to less accumulation?
438: Because H₂O contributes to O₃ loss? These correlations need to be discussed in more detail.
476 / Fig. 2: Why is half the figure upside down?
482 f.: It should be specified what is meant by “affecting boundary layer structure” – the current term is very generic.
494: HO₂ can be lost on aerosol surfaces, which inhibits O₃ formation (opposite effect!)
522 ff.: I cannot follow this logic. NO is lower for high O₃ days because (a) O₃ generation is VOC-limited or because (b) of the titration effect close to NO_x sources (NO + O₃ --> NO₂ + O₂)
556 ff.: It is important to control NO_x when chemistry is VOC-limited.
559 ff: What exactly are these different phases? Is titration meant by suppression?
571: Is there a specific reason to investigate midnight concentrations? Maybe the analysis should be limited to daylight values.
652: These sources emit both VOCs and NO_x.
656 ff.: How exactly were these factors identified? Why are there six factors? Additional explanations are needed here.
696 ff. / Fig. 6: There does not seem to be a relevant difference between the three cases. What’s the uncertainty? Are the differences even significant?
706 ff.: Please provide an explanation for speculations.
799 ff.: Sillman et al. suggested the HCHO to NO₂ ratio for O₃ sensitivity analysis.
803: What is MEM?
827: What is RIR?
944: It is not clear why the slope of the ridge could indicate the ratio at which VOCs and NO_x need to decline. Wouldn’t it be important to reduce NO_x as quickly as possible to move towards NO_x limited chemistry?
961: What is meant by “high-resolution observations” – the hourly measurements?

Citation: https://doi.org/10.5194/egusphere-2025-4519-RC2

AC2: 'Reply on RC2', Yu Shijie, 01 Mar 2026

Itemized Response to Editor’s Comments

Ms. Ref. No.: EGUSPHERE-2025-4519 | Measurement report

Title: Measurement report: Characterizing O3-NOₓ-VOC sensitivity and O3 formation in a heavily polluted central China megacity using multi-methods during 2019–2021 warm seasons

In “Measurement report: Characterizing O3-NOₓ-VOC sensitivity and O3 formation in a heavily polluted central China megacity using multi-methods during 2019–2021 warm seasons” Yu et al. investigate the sensitivity of O3 formation in Zhengzhou City between 2019 and 2021 based on VOC in-situ observations, several models and machine learning tools. The authors find that O3 formation in the Chinese megacity is limited by the availability of VOCs and recommend a focus on VOC emission reductions with simultaneous NOx control.

While the investigation of O3 pollution generally remains a highly important topic, I have major concerns regarding the implementation and results presented in this study. The large number of methods, that often seem redundant, the use of many abbreviations (often not defined) and changes between units makes it difficult to follow the line of argument. The lack of presenting NOx measurements and a detailed discussion on the role of NOx in O3 formation makes it additionally challenging to understand how the authors reach their scientific conclusions. Unfortunately, I therefore cannot recommend this manuscript for publication in ACP.

Response: Thank you for your careful reading of our paper and the valuable comments and constructive suggestions. We sincerely appreciate the reviewer's critical assessment, which has helped us identify key areas for improvement. We acknowledge the concerns regarding the large number of seemingly redundant methods, the frequent use of undefined abbreviations, and the lack of clarity in unit conversions, which may have made the line of argument difficult to follow. We also recognize the importance of presenting NOx measurements and providing a detailed discussion on the role of NOx in O3 formation, as these are essential for understanding how we reached our scientific conclusions.

In response to these concerns, we have thoroughly revised the manuscript. Below are the point-to-point responses to all the comments (the comments are marked in black font and the responses are marked in dark blue font). The major changes that have been made according to these responses are marked in yellow in the highlighted copy of the revised manuscript, and our own minor changes are marked in red font. Note that the following line numbers refer to those in the corrected version.

We hope that the revised version now meets the standards for publication in ACP. Thank you again for your time and expertise.

Major Comments:

Manuscript type: The manuscript type “Measurement Report” does not seem appropriate to me. A measurement report should “present substantial new results from measurements of atmospheric properties and processes from field and laboratory experiments” (https://www.atmospheric-chemistry-and-physics.net/about/manuscript_types.html). While it can be accompanied by model results, the focus should be on the presentation of a unique dataset rather than a pool of different methods, including modeling and machine learning tools.

Response: We sincerely appreciate the reviewer’s critical guidance regarding the definition and scope of a “Measurement Report” in Atmospheric Chemistry and Physics (ACP). We fully agree that the previous version relied too heavily on a “pool of methods,” which inadvertently overshadowed the primary value and uniqueness of our observational data.

In light of your suggestions and in strict accordance with ACP’s guidelines, we have fundamentally restructured the manuscript to realign its focus on our high-quality, long-term dataset. The specific revisions are summarized below:

(1) Highlighting Substantial New Data in an Understudied Region:

As per the ACP guidelines requiring “substantial new results from measurements,” we have emphasized the significance of our 3-year (2019–2021) warm-season dataset of VOCs and O₃ precursors with 1-hour temporal resolution. This dataset represents a rare and substantial record from Zhengzhou, a megacity in Central China that is significantly understudied compared to the North China Plain or the Yangtze River Delta. These high-resolution observations provide essential insights into the inter-annual evolution of urban photochemistry in this region.

(2) Streamlining Methods (“Subtraction” Strategy):

To address the concern regarding methodological redundancy and to refocus on the measurements, we have entirely removed the machine learning (XGBoost/SHAP) sections. We recognized that these "black-box" tools, while innovative, diverted attention away from the empirical evidence provided by the measurements themselves.

Deepening Data Interpretation with New Observational Evidence:

Following the requirement for “preliminary interpretation” within a Measurement Report, we have incorporated new analyses derived directly from the observations:

O₃ Weekend Effect: We added a detailed comparison of VOCs, NOxand O₃ responses between weekdays and weekends. This provides direct, measurement-based evidence of O₃ sensitivity without relying on complex model assumptions.

Chemical Characterization (OFP and Prop-equiv): We now calculate the Ozone Formation Potential (OFP) and Propylene-equivalent concentrations based on the measured VOC species. This allows for a deeper chemical characterization of the dataset, directly linking precursor mass to atmospheric reactivity.

(4) Logical Restructuring:

We have refined the logical flow to follow a natural progression of data interpretation: Observational Characteristics (including the Weekend Effect)→Source Apportionment (Mass vs. Reactivity) →Photochemical Mechanisms (simplified OBM) →Quantitative Control Strategies. In this revised structure, the OBM and CMAQ models are no longer the primary focus; instead, they serve as supporting tools to fulfill the ACP requirement for “discussing the potential significance” of the measured data.

We believe that this restructured version strictly adheres to the spirit of a “Measurement Report” by showcasing a substantial and unique dataset through rigorous, measurement-centric analysis.

Number of methods: The authors use a large number of different methods to investigate O3 formation sensitivity. It is not clearly established why all these methods are needed and what their added value is. The questions posed could be answered with a reliable set of in-situ observations and an appropriate model to simulate missing trace gases. Instead, the introduction of all these methods is confusing and makes it difficult to follow the central line of argument.

Response:We sincerely thank the reviewer for this insightful comment. We understand that presenting multiple methodologies could potentially cause confusion if their individual roles and collective value are not clearly articulated. We have carefully reflected on this and have fundamentally restructured the manuscript to present these methods not as a redundant collection, but as a hierarchical, multi-dimensional diagnostic framework.

We believe that the integration of multiple approaches is a key strength of this study. Our rationale for employing this "multi-method" framework is based on the following three aspects:

（1） Cross-verification and Robustness:

As highlighted by recent studies (e.g., Chu et al., 2024), O3 sensitivity analysis is often subject to significant uncertainties stemming from emission inventories, chemical mechanisms, or observation errors. By using independent methods (Observation-based vs. Model-based), we can cross-verify our findings. When different methods point to the same conclusion (in this case, VOC-limited conditions), it greatly enhances the reliability of the results and provides a more solid foundation for policy recommendations.

（2） Hierarchical and Multi-scale Characterization:

O3 formation sensitivity is characterized by complex spatio-temporal dynamics. No single method can capture all dimensions. We have organized our methods into a logical progression:

Preliminary Diagnosis (Weekend Effect): We have added an analysis of the “weekend effect” as an initial, empirical reality check. It provides direct, measurement-based evidence of how O3responds to real-world precursor fluctuations.
Temporal Dynamics (Ratio Method): The VOCs/NOx ratio is used specifically to capture the diurnal (hour-by-hour) evolution of sensitivity, providing a rapid screening of regime shifts during the day.
Refined Mechanism Analysis (RIR & EKMA): These OBM-based methods are driven by in-situ observations and are independent of emission inventory uncertainties. RIR is used to identify the "key controlling factors" (which specific species to reduce), while EKMA defines the "local reduction thresholds" (the specific proportions needed to reach the O3ridge line).
Regional/Spatial Context (CMAQ-DDM): While the above methods are site-specific, the 3D model provides the regional perspective, accounting for transport and meteorology that point-based observations cannot capture.

（3） Policy Support:

For local environmental management, identifying the "what" (key precursors) and "how much" (reduction thresholds) requires multi-dimensional evidence. Our approach provides a comprehensive "toolbox" for stakeholders to develop dynamic, localized control strategies.

To improve clarity, we have added a dedicated table in the revised manuscript that explicitly compares the principles and specific added value of each method:

Method	Main Principle	Advantages	Limitations
Ratio Method (VOCs/NOx)	Uses the concentration ratio of VOCs to NOx to determine ozone formation sensitivity (typically <8-10 indicates VOC-limited, >15-20 indicates NOx-limited).	Simple operation, quick and intuitive; Suitable for preliminary screening using routine monitoring data; Enables rapid identification of sensitivity regimes.	Thresholds based on U.S. experiences (NRC, 1991), lacking regional universality; Ignores reactivity differences among VOC species; Does not account for pollutant dispersion and chemical evolution; Cannot predict future trends (Sistla et al., 2002; Wolff & Korsog, 1992).
EKMA Curves	Uses box models or observation constraints to plot ozone isopleths, visualizing the non-linear response of ozone to precursor reductions; the "ridge line" divides sensitivity regimes.	Intuitively displays non-linear ozone-precursor relationships; Applicable for scenario simulations and policy development; Local thresholds can be adjusted; Clearly distinguishes VOC-limited, NOx-limited, and transition regimes (Dodge, 1977; Chen et al., 2019).	Relies on input parameters (e.g., VOC speciation, meteorology); Does not consider regional transport; Thresholds lack regional universality; Applicable only for daily maximum ozone assessment; Morning VOCs/NOx ratio often not significantly correlated with daily maximum O3 (Sillman, 1999; Shafer & Seinfeld, 1986).
RIR (Relative Incremental Reactivity) (OBM-based)	Uses observation-constrained zero-dimensional photochemical models to simulate the impact of precursor changes on ozone production rates, calculating relative incremental reactivity.	Independent of emission inventories, avoiding inventory uncertainties; Can simulate detailed gas-phase chemical mechanisms (e.g., MCM); Quantitatively identifies key precursors and key VOC species; Suitable for mechanistic studies at point scale (Cardelino & Chameides, 1995).	Requires high time-resolution and high-precision observational data (especially VOC speciation); Results represent only the vicinity of the observation site; Cannot predict long-term trends or regional transport; Chemical mechanisms still involve uncertainties; Cannot directly guide reduction amounts (Russell & Dennis, 2000; Zhao et al., 2020).
CMAQ-DDM (Decoupled Direct Method) (3D model-based)	Embeds the decoupled direct method within regional air quality models to quantitatively calculate the sensitivity coefficients of ozone concentrations to precursor emission changes.	Wide spatial coverage, enables assessment of spatial and temporal distribution; Can predict future emission scenarios; Accounts for complete processes including meteorology, transport, and deposition; Results applicable for policy formulation (Dunker et al., 2020; Luecken et al., 2018).	Relies on high-precision emission inventories and meteorological fields; Computationally resource-intensive; Gas-phase chemical mechanisms relatively simplified (to save computation); Uncertainties arise from inventories, meteorology, and chemical mechanisms (Kitayama et al., 2019; Liu et al., 2020).

In the revised manuscript, we have also improved the transitions between these sections to ensure the "central line of argument" is logical and easy to follow. We believe these changes clarify that these methods are complementary and essential for a robust assessment of O3 sensitivity in a complex urban environment.

Measurements: From the results section, I understand that besides VOCs, measurements of NOx and O3 are available. However, there are no details presented in the methods section.

Response: We would like to express our sincere gratitude to the reviewer for pointing out the omission of measurement details for NOx and O3. This information is indeed essential for a complete understanding of our data foundation. Following your suggestion, we have added a description of these measurements in the revised "Methods" section and provided comprehensive technical details, including instrument specifications and QA/QC protocols, in the Supplementary Materials.

The key details regarding these measurements are as follows:

Data Source and Co-location:

The NOx and O3 data were obtained from the Zhengzhou National Air Quality Monitoring Station. To ensure high data comparability, this station is co-located within the same monitoring area as our VOC sampling and meteorological observation sites. Furthermore, one of our co-authors, Minghao Yuan, who is a staff member at this monitoring station, directly supervised the data acquisition, ensuring the highest level of data integrity.

Instrumentation and Standards:

All gaseous pollutants were measured using high-precision analyzers compliant with China’s National Environmental Protection Standards (HJ 193-2013):

NOx was measured using a chemiluminescence analyzer (e.g., Thermo Fisher Scientific 42i). O3 was measured using a UV photometric analyzer (e.g., Thermo Fisher Scientific 49i).

QA/QC and Maintenance:

The monitoring process strictly adhered to the Technical Specifications for Operation and Quality Control of Ambient Air Quality Continuous Automated Monitoring System (HJ 817-2018).

Professional Maintenance: Routine maintenance, including daily zero/span checks and weekly precision audits, was performed by qualified third-party professional institutions.
Data Validation: Multi-point linear calibrations were conducted monthly to ensure measurement accuracy. Any data affected by power fluctuations or localized site interference were strictly screened and excluded.

NOx: Besides VOCs, Nitrogen Oxides are important precursors to tropospheric O3. However, the role of NOx seems to be mostly neglected in this study. A detailed description and discussion of the role of NOx are missing and I wonder how the authors reached their conclusions on O3 formation sensitivity, without accounting for NOx. E.g. the authors state that chemistry is VOC-limited in Zhengzhou, but suggest that VOC control is more important than NOx. VOC-limited chemistry is characterized by a large excess of NOx, which requires drastic emission cuts to improve long-term air quality.

Response: We have carefully revised the manuscript to include a more comprehensive analysis of NOx and to clarify our control strategy logic. Our detailed responses are as follows:

(1) Clarification of NOx Inclusion in Sensitivity Models

We would like to clarify that the role of NOx was fully considered in our sensitivity simulations. The Observation-Based Model (OBM) was constrained by high-resolution, hourly observed NOx data. The conclusions regarding O3 formation sensitivity (e.g., Relative Incremental Reactivity (RIR), EKMA, and DDM results) were derived based on the current high-titration environment. The results indicated that while both precursors are present, the O3 production rate is currently much more sensitive to changes in VOCs than to NOx in the urban area of Zhengzhou.

(2) Scientific Basis for Prioritizing VOCs Controls (Short-term vs. Long-term)

We fully concur with the reviewer’s assessment that VOCs-limited regimes are typically characterized by an excess of NO. In Zhengzhou, the extremely high NO concentrations exert a strong "titration effect" (NO +O3→NO2 + O2), which locally suppresses O3 levels.

Short-term Strategy: Under these strong VOC-limited conditions, moderate reductions in NOxcan lead to an "O3 disbenefit"—where the weakened titration effect causes O3 concentrations to rise. Therefore, to achieve rapid compliance with air quality standards in the short term, prioritizing VOCs reductions is the most direct and effective approach to suppress O3
Long-term Strategy: We completely agree with the reviewer that long-term, sustainable air quality improvement requires deep and synergistic cuts in NOx Only through substantial NOx reduction can the chemical regime eventually transition across the "ridgeline" into a NOx-limited regime. We have added a discussion in the "Policy Implications" section to emphasize that VOCs control is a current priority, while synergistic NOx reduction remains the ultimate long-term goal.

(3) Evidence from the "Weekend Effect" Analysis

To further address the reviewer’s concern, we have added a detailed analysis of the "Weekend Effect" in the revised manuscript. The observed increase in O3 concentrations during weekends, despite lower NOx emissions from reduced heavy-duty traffic, provides empirical evidence of the strong NOx titration effect and the VOCs-limited nature of the study area. This analysis helps to validate the sensitivity results obtained from our models.

We have incorporated these detailed descriptions and discussions into the revised manuscript. We believe these additions clarify the indispensable role of NOx in our study and strengthen the logical consistency of our proposed control strategies.

Correlation analysis: The presentation of correlations between different parameters in Section 3.1.1 seems random and does not follow a clear strategy, e.g. hypothesis – method – result – discussion. The set of in-situ observations is much more powerful than this: I recommend presenting trace gas levels (and if possible a longer time series), the characteristics of each season, diurnal cycles and the weekend effect for a sensitivity analysis. The application of machine learning tools is not necessary here or needs to be better justified.

Response: We sincerely thank the reviewer for these constructive comments and the strategic suggestions for restructuring our manuscript. We have carefully revised Section 3.1 and the overall analysis framework to better utilize the richness of our three-year in-situ dataset and to ensure a clearer scientific logic. The specific modifications are as follows:

（1） Restructuring the logic of Section 3.1

We have reorganized Section 3.1 to follow a "macro-to-micro" strategy, moving from an overall assessment of pollution trends to specific chemical mechanisms. The new structure follows a clear logic of: General Overview – Diurnal Evolution – Empirical Sensitivity Evidence.

Section 3.1.1 (Macro Overview): We now present the full 2019–2021 warm season (May–September) time series. We have added a comparative analysis between non-polluted and polluted days to establish the links between meteorological drivers (e.g., temperature, humidity, and radiation) and ozone accumulation. This addresses the reviewer’s concern regarding the previously "random" presentation of correlations.
Section 3.1.2 (Dynamic Diurnal Evolution): We merged the previous diurnal analysis into this subsection. By comparing the diurnal cycles of O3, NOx, and VOCs under different pollution levels, we highlight the "chemical fingerprints" of ozone episodes, such as the rapid accumulation of NO in the morning and its titration effects at night.

Deepening the analysis of observational data (Weekend Effect)

Following the reviewer’s suggestion to better exploit the in-situ observations, we have added a new section (Section 3.1.3) focusing on the "Weekend Effect."

We compared the concentrations of NOx, VOCs, and O3 between weekdays and weekends. Our statistical findings show that while NOx levels decrease during weekends, O3 concentrations remain high or even increase. This "weekend effect" serves as robust, purely observational evidence that the study area is in a VOCs-limited regime. This provides a solid foundation for the subsequent source apportionment and OFP calculations without relying on black-box models.

Addressing the necessity of Machine Learning

We agree with the reviewer that for a "Measurement Report," the focus should remain on chemical mechanisms and observational facts. Therefore, we have removed the machine learning tools (XGBoost and SHAP analysis) from the manuscript. This removal allows the paper to be more concise and ensures that the conclusions are directly supported by the three-year high-resolution monitoring data.

We believe these changes significantly improve the logical flow and the scientific weight of the observational analysis. Again, we thank the reviewer for guiding us toward a more rigorous presentation of our findings.

Abbreviations: Many abbreviations are used in this manuscript, and they are often not defined upon first use, which makes it difficult to follow. It is further concerning that the authors are in some cases not consistent with the abbreviations, e.g. “OBM” is an “observation-based model” in Line 41 and an “Ozone Box Model” in Line 128.

Response: We sincerely apologize for the oversight regarding the inconsistent and undefined abbreviations in our original manuscript. We appreciate the reviewer's meticulous attention to detail, which is crucial for the clarity and professional standing of our work.

In response to your suggestion, we have conducted a comprehensive review of the entire manuscript and implemented the following improvements:

Standardization and Consistency: We have meticulously checked all abbreviations to ensure they are defined upon their first mention. Specifically, the term "OBM" has been unified as "Observation-Based Model" throughout the text (e.g., corrected in Line 128) to eliminate any ambiguity.
Global Correction: Every abbreviation used in the manuscript, including those in figures and tables, has been cross-verified for consistency in both formatting and meaning.
Addition of an Abbreviation List: To further enhance readability and provide a convenient reference for readers, we have added a comprehensive "List of Abbreviations" . This table summarizes all technical terms and their corresponding full names used in the study.

We are truly sorry for any confusion caused by our previous presentation and hope that these systematic corrections ensure the manuscript is now clear and easy to follow.

Units: Many different units for trace gases are used throughout the text, including ppbv, ug/m3 and molecules/cm3. This makes it difficult to compare trace gas levels and I recommend choosing one unit (preferably mixing ratios) and using it throughout the entire manuscript.

Response: We sincerely appreciate the reviewer's constructive suggestion regarding the unification of units. We apologize for the inconsistency in the original manuscript, which indeed created difficulties for the readers.

Following your advice to maintain a single unit throughout the study, we have carefully considered the most appropriate choice. While we acknowledge the reviewer's recommendation to use mixing ratios (ppbv), we have decided to unify all gas concentration units to µg/m³ for the following reasons:

（1） Consistency with Local Standards: The majority of our data analysis and the National Ambient Air Quality Standards (NAAQS) in the study region are based on mass concentrations (µg/m³). Using this unit allows for a more direct comparison with regulatory limits and previous regional studies.

（2） Internal Logic and Calculations: Since the bulk of our original discussion, source apportionment, and health-related impact assessments were conducted using µg/m³, adopting this unit as the standard ensures the internal consistency of our calculations and avoids potential rounding errors or confusion during extensive conversions.

We have now meticulously revised the entire manuscript, including all text, tables, and figures, to ensure that all trace gas concentrations (previously in ppbv or molecules/cm³) are converted and presented consistently in µg/m³. We hope this modification meets with your approval and enhances the clarity of our report.

5: Why is PM2.5 relevant to this study? I recommend focusing on O3 and its precursors to avoid overloading this study.

Response: We genuinely appreciate the reviewer’s insightful suggestion regarding the scope of our study. We agree that focusing on O3 and its precursors provides a more concise and coherent narrative, preventing the manuscript from becoming unnecessarily complex.

In accordance with your recommendation, we have performed the following modifications:

(1) Removal of PM2.5 Content: We have removed the descriptions and data related to PM2.5 throughout the manuscript. This includes the approximately seven instances where PM2.5 was mentioned, primarily in the Methodology section and Section 3.1.

(2) Refined Focus: The revised manuscript now focuses exclusively on O3, its precursors, and the associated photochemical mechanisms. This adjustment ensures that the core objectives of the study are emphasized more clearly.

We believe these changes have significantly streamlined the paper and thank you for helping us improve its clarity and focus.

Minor comments:

33 f.: This sounds like VOCs increase in response to O3increases, while VOCs are precursors to O3.

34 f.: Do these values refer to VOC or O3 concentrations?

Response: We appreciate the reviewer’s insightful comments regarding the causal relationship between VOCs and ozone. We have revised the text to clarify that VOCs, as precursors, contribute to the varying levels of ozone pollution. We also explicitly stated that the numerical values provided refer to the mass concentrations of VOCs.

The revised sentence in the manuscript now reads:

"Mean VOC mass concentrations were found to be higher during more severe ozone episodes, with values of 84.7±51.0, 96.6±53.4, and 105.3±59.4 µg/m³ for non-polluted, mildly polluted, and moderately polluted periods, respectively, reflecting the role of these precursors in ozone formation."

37: Please define abbreviations upon first use.

Response: We thank the reviewer for this suggestion. The full names for CMAQ (Community Multiscale Air Quality) and PMF (Positive Matrix Factorization) have been provided at their first mention in the revised manuscript. We have also carefully checked the text to ensure all other abbreviations are properly defined upon initial use.

39: What is meant by “ozone emissions”? Ozone is not emitted but formed photochemically.

We are grateful to the reviewer for pointing out this oversight. We fully agree that ozone is a secondary pollutant formed through photochemical reactions rather than being directly emitted. To rectify this, we have replaced "ozone emissions" with "ozone formation" throughout the revised manuscript to ensure scientific accuracy.

46: What’s the “ratio method”?

Response: We thank the reviewer for pointing out the lack of clarity regarding the “ratio method.” We have now explicitly defined this term in the revised manuscript. Specifically, the “ratio method” refers to the diagnostic approach of using the ratio of precursor concentrations (VOCs/NOx) to determine the sensitivity of ozone formation. We have also ensured that the definition is clearly presented upon its first mention to avoid any ambiguity.

47: If ozone generation is limited by VOCs, it is highly important to control NOx. Of course, it remains important to reduce VOCs simultaneously, but long-term air quality improvements can only be reached through NOx reductions in that case.

Response: We sincerely appreciate the reviewer’s insightful comment regarding the long-term importance of NOₓ reduction. We fully agree that NOₓ control is the fundamental driver for sustained, long-term air quality improvements on a regional scale.

However, in the specific urban context of Zhengzhou, our findings indicate that the city is currently located in a strongly VOC-limited regime. Under such conditions, prioritizing the synergistic reduction of VOCs is more direct and effective for controlling O₃ peak concentrations and reducing the number of exceedance days in the short term. To reflect this perspective, we have added a discussion in the revised manuscript to clarify the distinction between long-term NOₓ-focused goals and short-term, VOC-oriented mitigation strategies for peak ozone control.

55 ff.: Several things are missing in the introduction, i.a. how O3is formed from its precursors and particularly what the role of NOx

Response: We appreciate the reviewer’s constructive suggestion. In the revised manuscript, we have expanded the first paragraph of the Introduction to provide a concise yet comprehensive description of the O₃ photochemical formation mechanism. Specifically, we have incorporated the radical cycle involving VOCs, NOₓ, and OH radicals. Furthermore, we have clarified the dual role of NOₓ in ozone chemistry, emphasizing both the NO₂ photolysis cycle, which leads to O₃ production, and the NO titration effect, which acts as a sink for O₃. We believe these additions provide the necessary theoretical foundation for the subsequent discussion.

65 ff.: Are the authors saying that they are the first to investigate O3formation from increasing anthropogenic sources?

Response: We apologize for the imprecise expression in the original manuscript, which may have led to a misunderstanding regarding the novelty of our work. We did not intend to claim that we are the first to investigate the general relationship between anthropogenic sources and O3 formation.

Instead, our goal was to provide a systematic investigation specifically focused on Zhengzhou—a representative megacity experiencing rapid urbanization and significant shifts in emission characteristics in recent years. In the revised version, we have adjusted the tone to emphasize that this study offers a "comprehensive assessment through integrated multi-model analysis and systematic monitoring" tailored to this specific region and period. We have revised the relevant sentences to more accurately reflect our contribution.

Thank you for your valuable feedback, which helped us present our findings more humbly and precisely.

76 f.: This sounds like the range of VOC mixing ratios in China is 27 – 92 ppbv.

Response: We sincerely thank the reviewer for pointing out this ambiguity. We apologize for the imprecise phrasing in the original manuscript, which may have inadvertently suggested that the VOC mixing ratios across the entire country were confined to a narrow range.

Our intention was to highlight the significant spatial variability of VOC concentrations across different Chinese cities. To clarify this, we have revised the sentence to specify that these values represent specific observations from different regions. The revised text now reads: "In China, VOC pollution exhibits complex spatial and temporal patterns; for instance, previous studies have reported average summertime VOC concentrations ranging from 27.0 ppbv in Nanjing to 92.0 ppbv in Tianjin."

We have updated this section to ensure the description is more accurate and clear. Thank you for your careful reading.

81 ff.: This section is difficult to follow due to the jumps between countries and continents.

Response: We sincerely apologize for the lack of clarity in the original geographical transitions. Following the reviewer’s constructive suggestion, we have reorganized this section to improve the logical flow. We now follow a 'from-global-to-regional' and 'from-characteristic-to-source' structure. Specifically, we first summarize the global diversity in VOC compositions and sources, and then transition to a detailed discussion of the complex patterns within China. This ensures a smoother transition before identifying the specific research gaps that this study aims to address.

This paragraph has been revised and rewritten as follows.Recent research has advanced our understanding of VOCs, key precursors to ozone. As revealed by global monitoring data, there are substantial geographical differences in the chemical composition and source profiles of VOCs. Globally, VOC signatures are highly region-specific. For instance, alkanes dominate the VOC pool in Colorado, USA (>80%), whereas oxygenated VOCs (OVOCs) prevail in Athens, Greece (Abeleira et al., 2017; Kaltsonoudis et al., 2016). Similarly, source contributions shift from LPG and solvents in Paris to biomass burning in Punjab, India (Baudic et al., 2016; Pallavi et al., 2019). Within China, VOC pollution manifests through complex spatiotemporal patterns, with concentrations varying widely—from 27.0 ppbv in Nanjing to 92.0 ppbv in Tianjin (An et al., 2017; Han et al., 2015). While fossil fuel combustion and solvent use are the primary drivers in the North China Plain, the petrochemical industry remains a dominant contributor in the Yangtze River Delta (Mozaffar et al., 2020).

83: Is BB the major VOC source throughout the entire year?

Response: We sincerely thank the reviewer for this insightful question, which has helped us improve the temporal precision of our description.

We agree that the dominance of biomass burning (BB) as a VOC source is highly seasonal rather than year-round. The study cited (Pallavi et al., 2019) was conducted during the winter months, a period characterized by intensive crop residue burning and increased heating demands in the Punjab region.

93 ff.: Are the authors talking about concentrations, emissions, formation rates or sensitivities?

Response: We thank the reviewer for identifying this ambiguity. In Lines 93 ff., we are primarily discussing O₃–NOx–VOCs sensitivity (i.e., the response of O₃ concentrations to changes in precursor levels).

We have clarified the text to state that while precursor emissions and concentrations are the drivers, and formation rates are the kinetic outputs, the focus of this section is on the diagnostic methods (like EKMA and OBM) used to determine the sensitivity regime (VOC-limited vs. NOₓ-limited). The revised text now explicitly uses the O₃–NOx–VOCs sensitivity to avoid confusion

115 ff.: Why not use a set of in-situ observations?

Response: The reviewer’s point is well-taken. In fact, this entire study is fundamentally built upon a comprehensive 3-year (2019-2021) dataset of high-resolution in-situ observations (including 108 VOC species, O₃, NOx, and meteorological parameters).

We chose the multi-method approach (OBM, PMF, and CMAQ) precisely to maximize the diagnostic value of these in-situ observations:

OBM: Directly utilizes in-situ concentrations to calculate real-time radical budgets and O₃production rates.
PMF: Uses the observed chemical fingerprints to trace the physical sources of VOCs.
Statistical Analysis: We also included direct analysis of observed trends, ratios, and correlations (e.g., Section 3.1).

By combining these, we provide a more robust chemical interpretation than what could be achieved by simple statistical descriptions of in-situ data alone. We have updated the text in Line 115 to more clearly emphasize that our analysis is observation-driven."

128: In the Abstract the authors state the OBM to be an observation-based model.

Response: It was my oversight; it is indeed "an observation-based model" here, and it has been corrected.

188: How were other trace gases and meteo parameters measured?

Response: We appreciate the reviewer’s inquiry regarding the measurement methods for trace gases and meteorological parameters. In response, we have added detailed descriptions in the revised manuscript. Specifically, these data were obtained from the Zhengzhou National Air Quality Monitoring Station. To ensure high data comparability, this station is co-located within the same monitoring area as our VOC sampling and meteorological observation sites. Furthermore, one of our co-authors, Minghao Yuan, who is a professional staff member at this monitoring station, directly supervised the data acquisition process. This involvement ensured the highest level of data integrity and quality control. We hope these clarifications address your concerns.

215: More details on the OBM are required.

242: More details are needed on the WRF/CMAQ model.

Response: We have added more comprehensive details for both models. For the OBM, we have clarified the chemical constraints (MCM v3.3.1) and the 1-hour time step. For the WRF/CMAQ, we added a new table in the Supplementary Information (Table S1) detailing the nested grid settings (36/12/4/1 km) and physical schemes to ensure model stability.

223: The reaction of OH and NO2does not destroy O3 but limits its formation. It should therefore be accounted for in Equation (3), rather than (4).

Response: The reviewer is absolutely correct. The reaction of OH and NO₂ is a termination step that removes precursors and radicals, thereby limiting O₃ formation rather than directly destroying O₃ molecules. We have relocated this term from the destruction equation (Eq. 4) to the discussion of formation limitation in Section 2.2.1 and updated the equations accordingly to ensure kinetic rigor.

248 ff.: What are all these abbreviations: FNL, SAPRC-99, AERO6, IC/BC, MEIC, REAS2?

Response: We have provided the full names for all technical abbreviations upon their first mention in the revised manuscript to ensure clarity for the readers.

FNL: Final Operational Global Analysis (NCEP)

SAPRC-99: Statewide Air Pollution Research Center 1999 mechanism

AERO6: Sixth-generation CMAQ aerosol module

IC/BC: Initial Conditions and Boundary Conditions

MEIC: Multi-resolution Emission Inventory for China

REAS2: Regional Emission inventory in ASia version 2

262 f.: What are first- and second-order sensitivities of O3?

265 ff.: What exactly do these equations show?

Response: We have clarified the physical meaning of the DDM equations:

First-order sensitivities (S_V,S_N): Represent the local linear response (the slope) of O₃ concentration to changes in VOCs and NOx emissions.

Second-order sensitivities (S_VV,S_NN,S_VN): Capture the non-linearity of the O₃ response surface. For instance, S_VV describes the curvature of the O₃-VOCs relationship, which is critical for understanding why the effectiveness of emission controls changes as reductions intensify.

Equations (5)-(9) collectively allow us to reconstruct the Taylor expansion of the O₃ response, providing a robust diagnostic of the sensitivity regime.

277: Why is PM2.5 needed in this study?

Response: We appreciate the reviewer's insightful suggestion. Upon careful consideration, we agree that the inclusion of PM_2.5 data might deviate from the primary focus of this study. To enhance the clarity of the manuscript and maintain a more concentrated discussion on the research theme, we have removed all PM_2.5 related content in the revised version. We believe this modification makes the study more concise and better aligned with our core objectives. Thank you for your professional guidance.

283: Why would the model be better at simulating emitted species?

Response: Primary species (e.g., SO₂ or NOx) primarily depend on the accuracy of emission inventories and transport/dispersion, which are relatively linear. In contrast, secondary pollutants like O₃ involve complex, non-linear chemical transformations and radical cycling. The accumulation of uncertainties in reaction mechanisms, precursor levels, and meteorological feedbacks typically makes secondary species more challenging to simulate than primary ones.

284: Only a small fraction of NO2 is emitted directly, most is formed photochemically from NO.

Response: The reviewer is correct. NO is indeed the dominant primary emission, and most NO₂ is formed via NO + O₃. We have revised the text to clarify that the model performs better for NOx as a primary precursor group compared to the purely secondary O₃, as the former is more directly constrained by the emission inventory.

319: What’s MDL?

Response: MDL stands for Method Detection Limit, defined as the lowest concentration of a substance that can be identified with 99% confidence to be greater than zero. We have added the full name to the text.

348: Why exactly is machine learning needed in this study?

Response: Upon reflection and in response to the reviewer's concern, we have realized that the machine learning component, while providing some quantitative weightings, introduced unnecessary complexity and was not as physically intuitive as our OBM and CMAQ results. To make this Measurement Report more focused on chemical mechanisms and observational data, we have entirely removed the machine learning section from the revised manuscript.

425 / Fig. 1: Why is a smoothing applied? What exactly does it involve? Why is the time series not just averaged to the desired resolution?

Response: Savitzky-Golay smoothing was applied in Fig. 1 solely for visualization purposes to reduce high-frequency noise and better reveal the seasonal and inter-annual patterns of pollutants. We have clarified this in the figure caption.

426 ff.: All three numbers are the same. How exactly are the pollution levels defined?

Response: We apologize for the clerical error. The levels are defined based on China's National Ambient Air Quality Standard (GB 3095-2012) for MDA8 O₃: Non-polluted (≤160μg/m³), Lightly polluted (160<MDA8≤215μg/m³), and Moderately polluted (>215μg/m³). The text has been corrected.

432: What exactly do the percentage values relate to? If it’s years, the time period is too short for a trend analysis.

Response: The percentage values refer to the frequency of exceedance days within the sampling period of each year. We agree that a 3-year period is insufficient for a climatological 'trend' analysis. We have replaced 'downward trend' with 'inter-annual variation' to accurately reflect the year-to-year changes in the proportion of polluted days during our study.

437: Why is O3positively correlated with wind speed? Usually, higher wind speeds lead to less accumulation?

Response: In Zhengzhou, the positive correlation between O₃ and wind speed is primarily due to: (1) Regional transport, where strong winds bring O₃ or its precursors from upwind polluted areas; and (2) Meteorological coupling, as higher wind speeds during the warm season in this region often coincide with clear skies and intense solar radiation, which are the primary drivers for O₃ formation.

438: Because H2O contributes to O3loss? These correlations need to be discussed in more detail.

Response: The negative correlation with RH is twofold: (1) Chemical loss: High water vapor concentrations enhance the sink of O₃ through the reaction O¹D + H₂O → 2OH; (2) Radiation attenuation: High RH is typically associated with increased cloud cover, which attenuates the UV radiation required for photolysis. We have expanded this discussion in the revised Section 3.1.1.

476 / Fig. 2: Why is half the figure upside down?

Response: This figure was part of the SHAP interpretability analysis for the machine learning model. As we have removed the entire machine learning section (as explained in the General Response), this figure has been deleted from the revised manuscript.

482 f.: It should be specified what is meant by “affecting boundary layer structure” – the current term is very generic.

494: HO2 can be lost on aerosol surfaces, which inhibits O3 formation (opposite effect!)

Response: We are grateful to the reviewer for pointing out the importance of HO₂ heterogeneous loss on aerosol surfaces and its inhibitory effect on O₃ formation. We acknowledge that the previous description was imprecise. In accordance with the reviewer’s suggestions and to improve the overall quality of the paper, we have removed the machine learning-related content and the corresponding discussion in this part. This revision helps to avoid potential inaccuracies and ensures that the manuscript remains focused on the core findings. Thank you for your professional and rigorous guidance.

522 ff.: I cannot follow this logic. NO is lower for high O3days because (a) O3 generation is VOC-limited or because (b) of the titration effect close to NOx sources (NO + O3 --> NO2 + O2)

Response: The lower NO concentration on high O₃ days is primarily attributed to the titration effect (NO + O₃→ NO₂+ O₂). Under high O₃ conditions, NO is rapidly consumed and converted to NO₂. This explains why we observe a negative correlation between NO and O₃ levels in an urban environment like Zhengzhou. The text has been revised to clarify this chemical feedback.

556 ff.: It is important to control NOxwhen chemistry is VOC-limited.

Response: We thank the reviewer for this important comment. We fully agree that controlling NOx emissions when the atmospheric chemistry is in a VOC-limited regime is indeed critical. In such regimes, reducing NOx without corresponding reductions in VOCs may actually increase O3 levels due to reduced titration, as we have discussed in relation to the "ozone disbenefit" concept. This highlights the necessity of coordinated emission control strategies. We have now emphasized this point in the revised manuscript to better guide policy implications. Thank you again for your valuable insight.

559 ff: What exactly are these different phases? Is titration meant by suppression?

Response:We thank the reviewer for this clarifying question. By "different phases," we refer to the distinct diurnal stages of O₃ behavior: (1) the suppression phase (P1) during midnight and early morning, where O₃ is suppressed by fresh NO emissions via titration (NO + O₃ → NO₂ + O₂); (2) the photochemical generation phase (P2) during daytime, where O₃ accumulates due to VOC/NOₓ reactions under sunlight; and (3) the titration phase (P3) prior to the evening peak, where O₃ is again consumed by reaction with NO from rush-hour emissions. Titration is indeed a form of suppression, but we have delineated these phases to capture the full diurnal cycle.

571: Is there a specific reason to investigate midnight concentrations? Maybe the analysis should be limited to daylight values.

Response:While midnight concentrations provide insights into the residual background and nocturnal NOx titration, we agree that the daylight period is more critical for O₃ formation analysis. We have shifted the primary focus of our radical and reactivity discussions to the peak photochemical hours.

652: These sources emit both VOCs and NOx.

656 ff.: How exactly were these factors identified? Why are there six factors? Additional explanations are needed here.

Response:The 6-factor solution was selected based on both statistical criteria (Q/Q(exp)) ratio, stable BS and DISP results) and physical interpretability. While we acknowledge these sources also emit NOx, the PMF analysis is inherently based on VOC species fingerprints. We have added a more detailed justification for the 6-factor selection in Section 2.3.2 and Section 3.2.1.

696 ff. / Fig. 6: There does not seem to be a relevant difference between the three cases. What’s the uncertainty? Are the differences even significant?

Response:We thank the reviewer for this insightful comment. Although the overall VOC composition in a region is relatively stable, Figure 6 reveals that as pollution levels increase, the contributions from biogenic sources, as well as anthropogenic sources such as vehicles and combustion, exhibit an upward trend. This pattern provides valuable insights for ozone pollution control strategies.

706 ff.: Please provide an explanation for speculations.

Response:We have replaced the term 'speculation' with a more detailed kinetic explanation. During O₃ pollution periods, enhanced solar radiation and radical levels (OH) significantly accelerate the oxidation of highly reactive aromatics from solvent use. We have provided evidence by comparing the ratios of reactive species (e.g., xylenes) to stable tracers (e.g., ethane), which shows a clear decrease during peak hours, confirming their rapid chemical consumption.

799 ff.: Sillman et al. suggested the HCHO to NO2ratio for O3 sensitivity analysis.

Response:We thank the reviewer for the suggestion regarding the HCHO/NO₂ indicator. We fully acknowledge the importance of the HCHO/NO₂ ratio in diagnosing ozone sensitivity, as highlighted in classic studies such as Sillman (1995). It should be noted that in the present study, we did not conduct field observations of HCHO, nor did we use satellite-based HCHO data products. Therefore, our ozone sensitivity analysis primarily relies on ground-based VOCs/NOx ratios and diagnostic methods such as RIR and EKMA. In future work, we plan to incorporate HCHO observations or satellite-retrieved HCHO data to further validate and extend the findings of this study.

803: What is MEM?

827: What is RIR?

Response:We have clarified these abbreviations: MEM stands for the Municipal Environmental Monitoring station, and RIR stands for Relative Incremental Reactivity. We have ensured they are defined upon first mention.

944: It is not clear why the slope of the ridge could indicate the ratio at which VOCs and NOxneed to decline. Wouldn’t it be important to reduce NOx as quickly as possible to move towards NOx limited chemistry?

Response:This is a critical point for policy formulation. In VOC-limited urban areas like Zhengzhou, prematurely reducing NOx without concurrent VOCs control can lead to an 'Ozone Disbenefit' (an increase in O₃ due to weakened titration). The 'ridge line' in the EKMA plot represents the transition boundary; its slope (2.9:1) provides the optimal 'pathway' for synergistic reduction. This ratio ensures that O₃ levels decline steadily while moving the system toward a NOx-limited regime without causing transient spikes.

961: What is meant by “high-resolution observations” – the hourly measurements?

Response:By 'high-resolution observations,' we refer to both the hourly temporal resolution of the measurements and the comprehensive suite of chemical species (108 VOC species) monitored. We have revised the text to 'hourly, multi-species observations' to be more precise.

Citation: https://doi.org/10.5194/egusphere-2025-4519-AC2

RC3: 'Comment on egusphere-2025-4519', Anonymous Referee #3, 23 Dec 2025

This work discussed concerns ozone formation within ZhengZhou, China, a mega city with considerable anthropogenic activity. The authors use observations (2019-2021), a box model, CMAQ, and machine learning tools to investigate the emissions sectors and species contributing to ozone during the period.

Unfortunately, this manuscript is rambling and does not present a clear message. “Conclusions” discussed are not traceable, and basic concepts well established in the community about ozone formation are clearly not well understood by the authors. Inadequate discussion of key factors to the analysis, along with generic findings does not lend well to this work, in its current form, being useful to the community and therefore I do not recommend this manuscript for publication in ACP.

Section comments:

Methods: The manuscript uses a variety of models, and methods, many of which are confusing for the reader to understand and follow. The authors need to shorten the observations and methods section (10 pages is excessive), and more succinctly describe the methodology used for this analysis. Make use of supplemental information for details that are not as critical. The paper should convey the main points, the big concepts of this work.

Results: Many of the findings presented in the results section are not groundbreaking, or new, and rely on machine learning techniques which not well described and therefore are questionable at best for accurately accessing relationships between ozone and precursor species. Additionally, the authors mention PM2.5, like a buzz word – the scope of this paper needs to be made smaller and more impactful. The units should be mixing ratios when discussing gaseous species, not the variety of units that are currently mentioned in the paper. What are the criteria to classify non-polluted, lightly polluted, and moderately polluted days? Discussion of the ozone diurnal cycle are not fully accurate, basic knowledge of ozone formation is clearly lacking by the authors. No clear policy relevant message is conveyed – it seems the authors do not have a great handle on the differing results from the different methods used.

Conclusions: Nothing of consequence discussed here. No policies recommended, again, nothing novel presented.

Citation: https://doi.org/10.5194/egusphere-2025-4519-RC3

AC1: 'Reply on RC3', Yu Shijie, 01 Mar 2026

Itemized Response to Editor’s Comments

Ms. Ref. No.: EGUSPHERE-2025-4519 | Measurement report

Title: Measurement report: Characterizing O3-NOₓ-VOC sensitivity and O3 formation in a heavily polluted central China megacity using multi-methods during 2019–2021 warm seasons

Response:Thank you for your careful reading of our paper and the valuable comments and constructive suggestions. Below are the point-to-point responses to all the comments (the comments are marked in black font and the responses are marked in dark blue font). The major changes that have been made according to these responses are marked in yellow in the highlighted copy of the revised manuscript, and our own minor changes are marked in red font. Note that the following line numbers refer to those in the corrected version.

In response to the reviewer's concerns regarding the rambling structure, unclear core message, untraceable conclusions, and misunderstandings of fundamental concepts, we have thoroughly revised the manuscript. Specifically, we have:

Restructured the manuscript to present a clear, logical narrative, eliminating redundant descriptions and improving overall readability.
Refined the core message by establishing a coherent "Source–Reactivity–Mechanism" framework that directly links our observational data, source apportionment results, and sensitivity analyses.
Made all conclusions fully traceable by ensuring each finding is explicitly supported by the corresponding data, figures, tables, or model results presented in the manuscript.
Strengthened the discussion of fundamental concepts related to ozone formation chemistry, ensuring accurate representation of well-established mechanisms and providing clearer explanations of key processes such as radical chemistry, titration effects, and the non-linear response of ozone to precursor reductions.
Reorganized the Methods and Results sections to improve coherence, with technical details moved to the Supplementary Information where appropriate.

We hope that these substantial revisions address the reviewer's concerns and that the revised manuscript now meets the standards for publication in ACP. Thank you again for your time and expertise.

Section comments:

Methods: The manuscript uses a variety of models, and methods, many of which are confusing for the reader to understand and follow. The authors need to shorten the observations and methods section (10 pages is excessive), and more succinctly describe the methodology used for this analysis. Make use of supplemental information for details that are not as critical. The paper should convey the main points, the big concepts of this work.

Response: We sincerely appreciate the reviewer’s constructive suggestion regarding the structure and clarity of the Methods section. We agree that a more concise and focused presentation of our methodology will significantly improve the readability of the manuscript. In response, we have performed a comprehensive revision of Section 2: Removal of Redundant Content: We have completely removed the sections related to machine learning (ML) methods. Upon further reflection, we agreed with the reviewer that these methods were not entirely consistent with the observational focus of this report and added unnecessary complexity.

Streamlining and Restructuring: To reduce the length and improve the flow, we have significantly condensed the descriptions of observations and modeling. Technical details and secondary parameters that are not critical for understanding the main results have been moved to the Supplemental Information.
Enhancing Methodological Clarity: We have supplemented the descriptions of the Observation-Based Model (OBM) and the Community Multiscale Air Quality (CMAQ) model with key data and essential parameters to ensure the scientific rigor of our analysis while maintaining brevity.
Standardization: We have carefully reviewed the entire section to unify all abbreviations and terminology, ensuring consistency throughout the manuscript. We believe these changes have made the "Observations and Methods" section much more succinct and better aligned with the core objectives of our work. Thank you for your insightful guidance.

Results:

Many of the findings presented in the results section are not groundbreaking, or new, and rely on machine learning techniques which not well described and therefore are questionable at best for accurately accessing relationships between ozone and precursor species.

Response: We sincerely appreciate the reviewer’s critical assessment regarding the novelty of our findings and the clarity of our methodology. We have taken these concerns into serious consideration and have implemented substantial revisions to better highlight the unique contributions of this work and to ensure the scientific rigor of our approach.

（1） Regarding the novelty and significance of the findings

While we acknowledge that ozone (O3) studies are extensive, we believe this research provides distinct scientific and practical value through the following aspects:

Unique and High-Resolution Dataset: Unlike studies based on short-term observations, this work utilizes a continuous, three-year (2019–2021) high-resolution online dataset covering 108 VOC species in Zhengzhou, a core megacity in the Central Plains Urban Agglomeration. This long-term dataset provides stronger statistical significance and temporal representation, which is crucial for capturing inter-annual variations and informing stable policy-making.
Robust Multi-Method Cross-Validation: Rather than relying on a single model, we integrated multiple independent approaches, including OBM-MCM (observation-based), CMAQ-DDM (grid-based), EKMA curves, and Ratio methods. By synthesizing results from tools with different strengths in chemical mechanism description, regional transport, and observational constraints, we reached highly consistent conclusions. This cross-methodological rigor significantly enhances the reliability of our findings beyond that of single-method studies.
Practical Policy Value: A key "groundbreaking" aspect of this study is the derivation of a specific, quantifiable synergistic reduction ratio (VOCs:NOx = 2.9:1). This provides a science-based, "ready-to-implement" target for local environmental departments to develop forward-looking emission control strategies, bridging the gap between academic discussion and regulatory application.

（2） Regarding the Machine Learning (ML) techniques

We agree with the reviewer that in the context of a mechanistic investigation, the "black-box" nature of certain machine learning techniques may obscure the underlying atmospheric chemistry and complicate the interpretation of precursor relationships.

In response to your suggestion and to ensure the manuscript remains focused on rigorous chemical mechanisms, we have completely removed the machine learning (ML/SHAP) components from the revised manuscript. We have instead strengthened the discussion using traditional, well-established atmospheric chemistry models (as mentioned in Point 1) to provide a more transparent and physically consistent analysis of O3-precursor relationships.

We believe these changes have significantly sharpened the focus and increased the scientific value of the manuscript. Thank you for your guidance.

Additionally, the authors mention PM2.5, like a buzz word – the scope of this paper needs to be made smaller and more impactful.

Response: We sincerely appreciate the reviewer’s insightful suggestion regarding the scope of our study. We entirely agree that narrowing the focus to O3 and its precursors provides a more concise and coherent narrative, preventing the manuscript from becoming unnecessarily complex and enhancing its overall impact.

In accordance with your recommendation, we have performed the following modifications in the revised manuscript:

（1）Removal of PM2.5 Content: We have removed the descriptions and data related to PM2.5 throughout the manuscript. Specifically, we have deleted approximately seven instances where PM2.5 was mentioned, primarily in the Methodology section and Section 3.1, to ensure the narrative remains focused.

（2）Refined Research Focus: The revised manuscript now focuses exclusively on O3, its precursors, and the associated photochemical mechanisms. This adjustment ensures that the core objectives and scientific findings of the study are emphasized more clearly and effectively.

We believe these changes have significantly streamlined the paper and made the scientific contribution more impactful. Thank you once again for helping us improve the clarity and focus of our work.

The units should be mixing ratios when discussing gaseous species, not the variety of units that are currently mentioned in the paper.

What are the criteria to classify non-polluted, lightly polluted, and moderately polluted days?

Response: We thank the reviewer for this important inquiry regarding the classification of pollution levels. In this study, the classification of non-polluted, lightly polluted, and moderately polluted days was strictly based on the Chinese National Ambient Air Quality Standard (GB 3095-2012) and the associated Technical Regulation on Ambient Air Quality Index (on trial) (HJ 633-2012).

Specifically, the levels are categorized according to the daily maximum 8-hour average (MDA8) concentration of O3 as follows:

Non-polluted days (Excellent and Good air quality): MDA8 O3≤ 160 μg/m³.
Lightly polluted days: 160 μg/m³ < MDA8 O3≤ 215 μg/m³.
Moderately polluted days: 215 μg/m³ < MDA8 O3≤ 265 μg/m³.

We have added these specific criteria to the revised manuscript to provide better clarity for the readers. We appreciate the opportunity to make our methodology more transparent and rigorous.

Discussion of the ozone diurnal cycle are not fully accurate, basic knowledge of ozone formation is clearly lacking by the authors. No clear policy relevant message is conveyed – it seems the authors do not have a great handle on the differing results from the different methods used.

Response: We are deeply grateful for the reviewer's critical and constructive feedback. We acknowledge that the previous version of the manuscript was overly descriptive and lacked a sufficiently rigorous exploration of the underlying atmospheric chemistry mechanisms. In response to your concerns, we have fundamentally restructured the logical framework of the study—moving from a descriptive report to a mechanistic analysis—and have thoroughly revised the discussion on O3 formation and policy implications.

The major revisions are summarized as follows:

Restructuring the Logical Framework (Source–Reactivity–Mechanism)

To ensure the manuscript is grounded in atmospheric chemistry principles, we have reorganized the discussion to follow a more rigorous scientific logic:

Observational Evidence (Section 3.1): We have removed the machine learning (ML/SHAP) components to focus on core chemical processes. Critically, we have added a "Weekend Effect" analysis. This serves as a "natural experiment" providing solid, model-independent observational evidence that the study area is in a VOC-limited regime, establishing a robust theoretical foundation for subsequent sensitivity diagnostics.
Linking Sources to Impact (Section 3.2): We have established a "Mass–Reactivity–Contribution" bridge. Beyond discussing VOC mass (PMF), we now incorporate Ozone Formation Potential (OFP). This explains the mechanistic link between sources and O3formation: for instance, while traffic and industry contribute significantly to VOC mass, their high content of reactive species (e.g., alkenes and aromatics) leads to an even higher share of OFP (35% for traffic), which justifies the final O3 source apportionment results from CMAQ.
Mechanism and Sensitivity (Sections 3.3 & 3.4): We merged these sections to focus on the response of O3to specific source reductions, using source apportionment results to directly inform the sensitivity analysis.

（2）Integration and Cross-Validation of Methodologies

To address the concern regarding the handling of different methodological results, we have included a Combined Table in Section 3.2. This table provides a horizontal comparison of mass contributions (PMF), reactivity contributions (OFP), and O3 contributions (CMAQ-ISAM). By cross-validating results from these distinct tools, we have significantly enhanced the reliability of our findings and provided a clearer scientific justification for the dominance of specific sectors.

（3）Quantifying Policy-Relevant Conclusions

We have replaced vague suggestions with a concrete, science-based "Chemical Red Line." By synthesizing quantitative data from EKMA isopleths and DDM sensitivity coefficients, we identified a critical VOC/NOx reduction ratio of 2.9:1. We now explicitly state that to achieve a net O3 decrease, VOC emissions must be reduced at a rate at least 2.9 times that of NOx. This provides policymakers with a specific, quantifiable target for effective O3 control.

（4）Systematic Comparison of Sensitivity Methods

To directly address the reviewer's concern regarding the integration of results from different methods, we have added a comprehensive comparative analysis of the ozone sensitivity diagnostic techniques employed in this study. Specifically, we now include a summary table that systematically compares the four main methods used in this work—the VOCs/NOx ratio method, EKMA curves, RIR (OBM-based), and CMAQ-DDM—across key dimensions including their underlying principles, applicability, advantages, limitations, and their specific results in the context of Zhengzhou. This cross-method comparison not only demonstrates the consistency of our core finding (VOC-limited regime) across different diagnostic tools but also highlights the complementary strengths of each approach. The table provides readers with a clear, at-a-glance understanding of why multiple methods were necessary and how their collective application strengthens the robustness of our conclusions.

Method	Main Principle	Advantages	Limitations
Ratio Method (VOCs/NOx)	Uses the concentration ratio of VOCs to NOx to determine ozone formation sensitivity (typically <8-10 indicates VOC-limited, >15-20 indicates NOx-limited).	Simple operation, quick and intuitive; Suitable for preliminary screening using routine monitoring data; Enables rapid identification of sensitivity regimes.	Thresholds based on U.S. experiences (NRC, 1991), lacking regional universality; Ignores reactivity differences among VOC species; Does not account for pollutant dispersion and chemical evolution; Cannot predict future trends (Sistla et al., 2002; Wolff & Korsog, 1992).
EKMA Curves	Uses box models or observation constraints to plot ozone isopleths, visualizing the non-linear response of ozone to precursor reductions; the "ridge line" divides sensitivity regimes.	Intuitively displays non-linear ozone-precursor relationships; Applicable for scenario simulations and policy development; Local thresholds can be adjusted; Clearly distinguishes VOC-limited, NOx-limited, and transition regimes (Dodge, 1977; Chen et al., 2019).	Relies on input parameters (e.g., VOC speciation, meteorology); Does not consider regional transport; Thresholds lack regional universality; Applicable only for daily maximum ozone assessment; Morning VOCs/NOx ratio often not significantly correlated with daily maximum O3 (Sillman, 1999; Shafer & Seinfeld, 1986).
RIR (Relative Incremental Reactivity) (OBM-based)	Uses observation-constrained zero-dimensional photochemical models to simulate the impact of precursor changes on ozone production rates, calculating relative incremental reactivity.	Independent of emission inventories, avoiding inventory uncertainties; Can simulate detailed gas-phase chemical mechanisms (e.g., MCM); Quantitatively identifies key precursors and key VOC species; Suitable for mechanistic studies at point scale (Cardelino & Chameides, 1995).	Requires high time-resolution and high-precision observational data (especially VOC speciation); Results represent only the vicinity of the observation site; Cannot predict long-term trends or regional transport; Chemical mechanisms still involve uncertainties; Cannot directly guide reduction amounts (Russell & Dennis, 2000; Zhao et al., 2020).
CMAQ-DDM (Decoupled Direct Method) (3D model-based)	Embeds the decoupled direct method within regional air quality models to quantitatively calculate the sensitivity coefficients of ozone concentrations to precursor emission changes.	Wide spatial coverage, enables assessment of spatial and temporal distribution; Can predict future emission scenarios; Accounts for complete processes including meteorology, transport, and deposition; Results applicable for policy formulation (Dunker et al., 2020; Luecken et al., 2018).	Relies on high-precision emission inventories and meteorological fields; Computationally resource-intensive; Gas-phase chemical mechanisms relatively simplified (to save computation); Uncertainties arise from inventories, meteorology, and chemical mechanisms (Kitayama et al., 2019; Liu et al., 2020).

We believe these comprehensive revisions address your concerns regarding the scientific rigor and the clarity of our conclusions. Thank you for pushing us to improve the depth and impact of this work.

Citation: https://doi.org/10.5194/egusphere-2025-4519-AC1

AC4:
'Comment on egusphere-2025-4519', Yu Shijie, 01 Mar 2026
Dear Editor,
Thank you for the opportunity to revise our manuscript for consideration as a Measurement Report in Atmospheric Chemistry and Physics. We are also grateful to the three reviewers for their critical yet highly constructive feedback, which has been instrumental in reshaping this work.
In this revision, we have fundamentally restructured the manuscript to strictly align with the requirements of an ACP Measurement Report. Our primary strategy was "distillation and focus"—moving away from overly complex modeling and returning to the core strength of our study: a high-quality, three-year, high-time-resolution dataset from a representative region of central Henan Province.
Key substantial changes include:
Refinement of Scope: We have removed the machine learning component included in the previous version. While technically interesting, we agree with the reviewers that it detracted from the observational focus of the report. This has allowed us to provide a more rigorous and in-depth analysis of the measured data itself.

Structural Reconstruction of Results:

(1) Section 3.1 now focuses on a detailed analysis of the "weekend effect," replacing the ML analysis to better characterize local ozone (O3) precursor dynamics.
(2) Section 3.2 has been redesigned to integrate VOC source apportionment, chemical reactivity (OFP), and O3 source attribution into a cohesive framework, providing a clearer picture of photochemical pollution in this specific region.
(3) Sections 3.3 and 3.4 were overhauled to investigate O3 formation mechanisms, including a comparative analysis of different sensitivity diagnostic methods to ensure the robustness of our conclusions.
Enhanced Significance: A new section on policy implications (Section 3.4) has been added. By bridging the gap between high-resolution observations and mitigation strategies, we believe the manuscript now offers greater value to both the scientific community and regional air quality management.

Conciseness: The methodology has been streamlined, with redundant technical details moved to the Supporting Information, ensuring the main text remains focused on the scientific narrative.

This study presents a rare, multi-year perspective on one of the most photochemically polluted regions in China. We believe the revised manuscript, with its emphasis on high-quality observational evidence and multi-method validation, now meets the high standards of ACP.
We have addressed all reviewer comments in the attached point-by-point response. Thank you for your time and for managing this submission.
Sincerely,
Shijie Yu
Citation: https://doi.org/10.5194/egusphere-2025-4519-AC4

Shijie Yu, Hongyu Liu, Hui Wang, Fangcheng Su, Minghao Yuan, and Ruiqin Zhang

Supplement

https://doi.org/10.5194/egusphere-2025-4519-supplement

Data sets

VOCs Dataset S. Yu et al. https://doi.org/10.5281/zenodo.17214861

Shijie Yu, Hongyu Liu, Hui Wang, Fangcheng Su, Minghao Yuan, and Ruiqin Zhang

Viewed

Total article views: 767 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
475	252	40	767	47	33	41

HTML: 475
PDF: 252
XML: 40
Total: 767
Supplement: 47
BibTeX: 33
EndNote: 41

Views and downloads (calculated since 23 Oct 2025)

Month	HTML	PDF	XML	Total
Oct 2025	116	21	9	146
Nov 2025	71	63	10	144
Dec 2025	66	72	12	150
Jan 2026	63	33	2	98
Feb 2026	36	21	0	57
Mar 2026	123	42	7	172

Cumulative views and downloads (calculated since 23 Oct 2025)

Month	HTML	PDF	XML	Total
Oct 2025	116	21	9	146
Nov 2025	71	63	10	144
Dec 2025	66	72	12	150
Jan 2026	63	33	2	98
Feb 2026	36	21	0	57
Mar 2026	123	42	7	172

Viewed (geographical distribution)

Total article views: 743 (including HTML, PDF, and XML) Thereof 743 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 24 Mar 2026

Short summary

As ozone (O₃) pollution remains a major concern, this study draws on long-term observational data to thoroughly analyze O₃ and its precursor substances. This study examines VOC pollution characteristics, explores local O₃ formation mechanisms, and assesses O₃ sensitivity. Our key finding is that motor vehicles are main sources of VOCs and O₃; local O₃ is more sensitive to VOCs. These guide targeted pollution cuts and protect the public environment.


Total:	0
HTML:	0
PDF:	0
XML:	0