the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 29 Apr 2025
Technical note: Adaptably diagnosing O3-NOx-VOC sensitivity evolution with routine pollution and meteorological data
Abstract. Elucidating the evolving O3-NOx-VOC sensitivity in response to varying precursor emission trends is critical for mitigating the elevating ozone. Due to the complexities and resource constraints inherent in conventional methods, we developed an adaptable methodology addressing this issue through empirical parametric regression of routine data (O3/NOx/NO2). The log-Bragg3 model (Equation 3) performed best in globally characterizing the daytime ozone production (DPO3)-NOx (or NO2) relation, including regions with severe PM2.5 contamination where ozone formation is additionally influenced by aerosol-inhibited photochemical regime. Over 95 % of these fits achieved statistical significance (p<0.1). This model provides parametric interpretations of ozone formation intensity (d), the associated chemical processes (b), and the O3-NOx-VOC sensitivity partition threshold (e). More vigorous photochemical reactions are implicated in the studied Chinese regions by higher values of parameters b (0.87–2.42) and d (34.72–54.78) relative to EU/US (b=0.26–0.57, d=9.97–31.45). Divergent temporal trends in parameter b further indicate fundamentally distinct evolutionary pathways in regional ozone chemistry between China and EU/US. Specific to MDA8-daytime hours, the Chinese city agglomerations were all diagnosed as being in the VOC-limited regime in both 2014 and 2019 on the regional scale, exhibiting significantly higher spatial predominance than the previous satellite-derived HCHO/NO2 ratio inferences. The DPO3-NO2 pseudo-diagnosis constituted major uncertainty in spatiotemporal diagnosis, whereas the DPO3-NOx curve showed superior reliability. This methodology helps provide critical insights for formulating spatially differentiated precursor control policies.
- Preprint
(1512 KB) - Metadata XML
-
Supplement
(4590 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-1596', Anonymous Referee #1, 27 May 2025
In “Technical Note: Adaptably diagnosing O3-NOx-VOC sensitivity evolution with routine pollution and meteorological data” Huang and Liao investigated the sensitivity of O3 formation at selected sites in China, the US and Europe by applying different fit equations to the datasets. The authors identify most of the studied regions to be dominated by VOC-limited O3 formation sensitivity.
While this is generally an important topic to investigate, unfortunately this study seems incoherent and is often difficult to follow. It remains largely unclear why and how the suggested fit equations are applied to the data and even more important what the added value of this analysis is. The in-situ observations investigated in this study can be used to directly infer the dominating sensitivity instead of using fit functions. Are any generalized conclusions drawn from the fitting? Could it be applied to other regions where observations are not available and how would that be possible considering that crossover points occur at NOx to VOC ratios that are characteristic to each location?
It is further concerning that the authors have published a paper earlier this month (Huang et al., 2025), which they are now referring to have “critical limitations” which “fail” in respect to two different aspects (Line 54f.). This makes me wonder why the authors have not previously fixed these issues, considering that this previous paper was published one month after the submission of this manuscript.
Some statements are further not backed with the current literature. The authors often use terms that are not commonly known in literature and do not provide sufficient definitions or explanations. The same applies to abbreviations that are not defined when first used. The figures have too many panels, are too small and have a low resolution, which makes them difficult to read and understand the results.
Considering these various drawbacks, unfortunately, I cannot recommend this manuscript for publication in its current state as it does neither meet the scientific nor the methodological standards of an ACP publication. If the authors wish to improve their manuscript in the future, please find more detailed comments and questions in the following, which might be helpful for revising the study.
Line 53 ff.: Could the authors describe the study of Huang et al., 2025? What were the methods applied and the findings of this study?
Line 53: What is “OFR”? Please define abbreviations when first used.
Line 54: The authors have published the study they are referring to here (Huang et al., 2025) earlier this month and are now referring to critical limitations of their work. I find this a bit irritating. Why do the authors have not implemented the improvements in the previous study?
Line 55 f.: How do the authors define the NOx-limited/transition boundary? The transition point is commonly referred to as the crossover from NOx- to VOC-sensitive chemistry, but there is no exact definition of a transition region in textbook literature. If I read the graphical abstract correctly it is related to the 95th percentile of O3 production. Where does this definition come from and what’s the reasoning for it?
Line 56: What do the authors mean by parametric modeling? Is this a reference to parameterizations in atmospheric models or something different?
Line 57: What do the authors mean by environmental stability? Whenever using non-textbook terms, I recommend a full definition and explanation.
Line 58: Please elaborate on the “bend” – what is it and where is it coming from? The cited literature Romer et al., 2018 and Guo et al. 2023 do not seem to mention / explain this bend. How can PO3 have two different values for the same NO2?
Line 77 f.: What exactly are parametric vs non-parametric results?
Line 80 ff.: Did the authors use these equations to fit their data? How were these fits chosen?
Line 108: What is the study period?
Line 117: What are “records ≤ 4”? Can the authors provide a reasoning for this “no precipitation” definition? I am not aware of being able to accurately infer rainfall from cloud cover.
Line 130: The European Union does not describe a geographical region and some parts of the map in Figures S1 are not part of the EU. I recommend referring to the region as, e.g. Europe.
Figure 1: The panels are too small and the resolution too low. It is difficult to read the legend. It is further difficult to distinguish between any of the stations because the data points are overlapping.
Line 170 ff.: It is unclear what exactly the authors are trying to show in Figure 1. What is a parametric y-axis and a non-parametric x-axis approach?
Line 174 f.: What are the definitions of the scenarios the authors are referring to?
Line 180 ff.: Is this the bend that the authors were referring to earlier? The reaction of OH + NO2 is a termination reaction of the HOx cycle and its dominance characterizes a VOC-limited O3 formation regime. Unlike the authors state, the cited studies do not show the existence of a pseudo NOx limited under a NOx saturated regime. Further evidence would be required to prove this statement of the authors, which does not agree with our current knowledge of O3 formation sensitivity.
Line 193 ff. / Figure 3: What is the added value of these fits? It is possible to determine the dominating sensitivity of O3 formation based on the observational data of O3 and NO2. Why are the fits needed? The individual fit parameters are likely different for each location, as the crossover does not always occur at the same NO2 mixing ratio (depending on the availability of VOCs).
Line 217 ff.: How was the log-Bragg 3 fit chosen? Does it provide the best result? How was this evaluated?
Technical:
Line 29.: Please check the author of this reference (“Collaborators”).
Line 49 f.: There seems to be a part of the sentence missing “As NOx increases.”
Citation: https://doi.org/10.5194/egusphere-2025-1596-RC1 -
RC2: 'Reply on RC1', Anonymous Referee #2, 05 Jun 2025
Huang et al. established statistical fitting equations for data sets from China, the United States, and Europe, and studied the sensitivity of O3 formation to precursors. This is an important topic at present. The author claimed that this parametric methodology overcomes the complexities and resource constraints inherent in conventional methods, and is expected as a unified tool to facilitate global ozone mitigation under evolving precursor emission patterns and climate change. Unfortunately, I do not agree with this view. The main reasons are as follows:
- The authors established a unified statistical fitting equation to fit the nonlinear relationship between ozone and precursors. After comparing multiple nonparametric regression models, the authors found that the log-Bragg3 model is universal. However, the authors did not explain why the log-Bragg3 model equation is universal. The model lacks the basic connotation of atmospheric chemistry and atmospheric physics, but is only a statistical equation obtained by fitting. In terms of mathematics, one can completely fit a very complex nonlinear relationship with high precision by constructing high-order statistical equations. This is just a process of approximating fitting by the infinite series method in mathematics. However, this operation has not made any essential improvement to the study of ozone sensitivity. It is just a mathematical technique. Even if, as the author says, common results have been obtained in the existing data of different cities, there is no guarantee that universal results will be obtained in the future.
In fact, since 2020, there have been many breakthroughs in the research on the sensitivity of O3 formation to precursors. The city lockdown measures during the COVID-19 period provided natural experiments for the emission reduction of precursors. Some new insights have even overturned the past understanding of atmospheric chemistry's sensitivity to ozone, and thereby improved the simulation effect of atmospheric ozone generation. This paper only focuses on research data before 2019, and it seems to deliberately avoid the adverse effects of emission reduction on simulation during the epidemic. Therefore, it is difficult to believe that the results of this study have achieved universal success. Are the study findings applicable during the COVID-19 pandemic in 2020?
- The paper mentioned their past research results, but did not explain in depth the improvements of this method.
- Many details of the paper are obscure and difficult to understand. Especially in the explanation of professional terms and details of model parameters. At the same time, the visualization of the graph is also very vague, which seriously affects normal reading.
Based on the above reasons, I have to recommend rejection of the manuscript.
Citation: https://doi.org/10.5194/egusphere-2025-1596-RC2 -
AC2: 'Reply on RC2', Minjuan Huang, 20 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1596/egusphere-2025-1596-AC2-supplement.pdf
-
AC1: 'Reply on RC1', Minjuan Huang, 20 Jul 2025
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-1596/egusphere-2025-1596-AC1-supplement.pdf
-
RC2: 'Reply on RC1', Anonymous Referee #2, 05 Jun 2025
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,862 | 53 | 17 | 1,932 | 32 | 33 | 41 |
- HTML: 1,862
- PDF: 53
- XML: 17
- Total: 1,932
- Supplement: 32
- BibTeX: 33
- EndNote: 41
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1