the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Nov 2025
Measurement report: Altitudinal Shift of Ozone Regimes in a Mountainous Background Region
Abstract. Elevated background ozone (O3) poses significant challenges for regional air quality management. Understanding the vertical distribution of O3 and its precursors is critical yet underexplored in Southwest China. This study presents the first comprehensive altitudinal gradient analysis (550 m, 1,774 m, 2,119 m a.s.l.) in the Fanjingshan National Nature Reserve, a remote high-altitude site on the Yunnan-Guizhou Plateau. Continuous measurements (March–August 2024) revealed a marked positive gradient in O3 (14.8 ± 15.2 ppb at mountain foot to 40.2 ± 14.7 ppb at mountaintop), contrasting with declining precursor concentrations. Random Forest–SHAP analysis identified relative humidity and NOx as dominant controls at the mountain foot, whereas temperature and reactive VOCs governed O3 variability aloft. Chemical box modeling (OBM-MCM v3.3.1) demonstrated net O3 destruction at mountain foot (−1.93 ppb hr-1) due to NO titration, shifting to net production at mountainside (0.35 ppb hr-1) and mountaintop (0.29 ppb hr-1). While O3 formation remained NOx-limited across all sites, sensitivity to anthropogenic hydrocarbons increased with altitude (RIR: -0.12 mountain foot to 0.51 mountaintop). Transport analysis indicated O3 accumulation at mountain foot via regional transport, contrasting with mountainside and mountaintop, which function as net source regions. These findings necessitate altitude-specific O3 control: prioritizing NOx reduction at lower elevations while coordinating NOx and VOC controls at higher altitudes. Expanding high-altitude monitoring, especially in under-monitored areas like Southwest China, is crucial for characterizing regional background pollution. Future studies require vertical monitoring with improved models to assess transboundary impacts and changing emissions.
- Preprint
(2623 KB) - Metadata XML
-
Supplement
(2326 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-4818', Anonymous Referee #1, 20 Nov 2025
-
AC1: 'Reply on RC1', Yonghong Wang, 06 Jan 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-4818/egusphere-2025-4818-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Yonghong Wang, 06 Jan 2026
-
RC2: 'Comment on egusphere-2025-4818', Anonymous Referee #2, 21 Dec 2025
The Measurement report "Altitudinal Shift of Ozone Regimes in a Mountainous Background Region" by Yang et al. describes a dataset of observations of ozone and its precursors along an altitudinal gradient on a mountain in south-east China. The measurements cover a period of several months, from March to August and are quite extensive. The authors do not limit themselves to show the observations but also present some analysis of the results, which is commendable. The dataset is novel, in the sense that observations from this part of the world are still sparse, and the conclusions are generally well supported by the data. The manuscript fits the scope of the journal, and I do not see any major issue with it. I recommend publication after the authors have addressed the comments below.
Section 2.1: the mountain-foot and mountain-top sites are described as being close to touristic areas, with shops and restaurants. I assume these involve some, although perhaps small, anthropogenic emissions which may affect the interpretation of the results. Also a "eco-friendly vehicle route" is mentioned, but it is not clear what it means, and whether it implies traffic emisssions. Can the authors clarify and comment? Were data filtered for local emissions?
Section 2.3: it is not clear to me why the Random Forest modelling was used. The results reported in Section 3.2 are interesting, but they could easily be obtained by a simple covariance matrix. Can the authors explain why they decided to use a machine learning algorithm? What kind of additional information is being obtained compared to a multivariabvle correlation plot?
Section 2.5: it is stated that the OBM box model is constrained to the observations. Was this the case only when calculating the ozone production pathways (Fig. 6), or also when calculating the EKMA curves? I think some caution must be applied when interpreting the results of a constrained box-model for long-lived species such as O3. If O3 is constrained itself that could led to errors expecially when considering the loss rates. Moreover I don't think you can reliably calculate EKMA curves using a model constrained to O3. The authors should clarify these questions and discuss these points. All the necessary caveats about using box models should be explicitly mentioned in the text.
The authors also make the assumption that the difference between calculated and measured ozone is due to transport (line 194 and following); they also mention that dilution is included in the boxmodel, while deposition is not mentioned. I think it is a stretch to assume that Rtrans is all advected ozone. First of all it is unclear how would dilution work within a constrained box model, where the concentrations of key species are fixed. Second, vertical transport and deposition cannot be discounted, especially in a mountain environment where uphill and downhill air movements occur on a daily basis. Finally, the MCM is extensive but far from complete: a perhaps significant fraction of photochemically generated ozone is not accounted for by the current chemistry in the MCM, which would lead to overestimating the role of transport. The authors should explain better and in more detail how they are using the model. clarify their procedures and discuss how the potential errors introduced by their assumptions affect their results.
A few minor points:
*) lines 257-263. It does not look to me as if temperature, humidty and wind have "statistically insignificant" differences. In fact they all seem quite different by looking at figure 4. Please rephrase or clarify.
*) it would be interesting to connect the O3 diurnal cycles at different altitudes with the different OH reactivities. How do the cycles of VOCs and NOx shape the diurnal O3 cycle?
*) isoprene dominates reactivity at the mountaintop but the model analysis indicates anthropogenic VOCs, namely aromatics, as the main contributors to RIR (line 377 and figure 8). The authors should expand the discussion on this point.
*) the measurements cover the period March-August, yet there is not mention of possible seasonal effects. Are there any differences beween spring and summer data? If so, they should be discussed. Otherwise, it should be said that there are no differences.
*) this is just a suggestion, but it would be interesting to see figure 3 with the stations sorted by altitude, in addition to sorted by concentrations.
Citation: https://doi.org/10.5194/egusphere-2025-4818-RC2 -
AC2: 'Reply on RC2', Yonghong Wang, 06 Jan 2026
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-4818/egusphere-2025-4818-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Yonghong Wang, 06 Jan 2026
Data sets
Measurement report: Altitudinal Shift of Ozone Regimes in a Mountainous Background Region Y. Yang et al. https://doi.org/10.5281/zenodo.17239121
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 175 | 108 | 22 | 305 | 43 | 25 | 24 |
- HTML: 175
- PDF: 108
- XML: 22
- Total: 305
- Supplement: 43
- BibTeX: 25
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Overall Evaluation
This manuscript presents a comprehensive and valuable study on the altitudinal distribution of ozone (O₃) and its precursors on Mount Fanjing, a remote background site on the Yunnan-Guizhou Plateau in Southwest China. The research is timely and addresses a critical knowledge gap, as high-altitude observations, particularly in this understudied region, are sparse. The experimental design is robust, incorporating a multi-platform approach with gradient observations, advanced statistical analysis (Random Forest with SHAP), chemical box modeling (OBM-MCM), and Concentration-Weighted Trajectory (CWT) model. The key findings-a positive O₃ gradient with altitude, a shift from net O₃ destruction at the foot to net production aloft, and an altitude-dependent shift in chemical regimes-are well-supported by the data and clearly presented. The study makes a significant contribution to the field of mountain atmospheric chemistry and provides actionable insights for region-specific air quality management. I recommend publication after minor revisions to address the points outlined below.
Major Strengths
1.This is the first detailed altitudinal gradient study of O₃ and its precursors in the Fanjingshan region. The data provides a crucial benchmark for understanding background pollution in Southwest China.
2.The combination of in-situ measurements, machine learning for driver attribution, and detailed chemical modeling is a powerful and modern approach that strengthens the conclusions significantly.
3.The clear demonstration of shifting O₃ regimes with altitude-from NOₓ-dominated titration at the foot to VOC-sensitive production influenced by temperature and transport aloft-is a key scientific result. The discussion of the decoupling between VOC concentration and OH reactivity (e.g., isoprene) is particularly insightful.
4.The conclusion that O₃ control strategies must be altitude-specific is well-argued and has practical implications for regional air quality planning.
Specific Comments and Suggestions for Revision
1)The manuscript mentions 57 VOCs species were measured. It would be highly beneficial to include a table in the supplement listing these species and their average concentrations at each site. This is critical for reproducibility and for readers to assess the VOCs mix.
2)The 0-D box model (OBM-MCM) is a suitable tool, but its inherent limitation in not accounting for advective transport should be explicitly stated in the methodology or discussion. Acknowledging that the calculated R_trans is a residual helps, but a sentence on the model's limitations would strengthen the manuscript.
1)Global Comparison (Figure 3): The comparison is useful for context. However, to make it more robust, please consider adding the time period (year/season) of the compared data in the figure or its caption, as O₃ levels can have temporal trends.
2)Negative RIR Values: The negative RIR for VOCs at the mountain foot is a critical finding. The explanation is correct (strong NOₓ-limited regime where VOC reduction can increase O₃), but this non-intuitive concept could be elaborated upon slightly for clarity, perhaps with a reference to the classical EKMA diagram concept.
1)While the figures are informative, some captions are very dense (e.g., Figure 2). Consider streamlining the captions and moving detailed descriptions of plot elements (e.g., the "cloud," "raindrop" components in Figure 2) to the main text or supplement.
2)Some sentences, particularly in the abstract and introduction, are very long and complex. A thorough proofread to break down overly long sentences would improve readability.
3)Check for consistency in reference formatting (e.g., journal name abbreviations, use of "et al.").
Typographical and Minor Errors
Conclusion
This is an excellent piece of work that provides a valuable dataset and insightful analysis of ozone photochemistry in a complex, high-altitude terrain. The minor revisions suggested above will further polish the manuscript and solidify its arguments. I look forward to seeing the publication.