the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 01 Jul 2026
Vertical Structure and Seasonal Evolution of Atmospheric Oxidizing Capacity across Urban and Rural Regions: Observational Constraints from OH Radical Production Pathways
Abstract. Atmospheric oxidizing capacity (AOC) drives the formation of secondary pollutants, yet conventional surface observations fail to resolve its pronounced vertical heterogeneity, often leading to incomplete interpretations of regional pollution chemistry. Using ground-based hyperspectral vertical remote sensing observations collected between March and August 2023 at representative urban (AHU) and rural (CF) sites in the Yangtze-Huai River Basin, we quantified the vertical contributions of HONO, HCHO, and O3 photolysis to OH production. AOC showed a strong positive correlation with aerosol loading (R = 0.88–0.93), indicating that enhanced atmospheric oxidation promotes secondary aerosol formation. In urban air masses, the AOC regime exhibited distinct vertical stratification. Rapid oxidation below 1 km was primarily driven by HCHO and HONO, whereas O3 photolysis became the dominant OH source above 2.8 km, accounting for more than 74 % of total OH production. Urban OH production transitioned from near surface HONO dominance in spring (P(OH)HONO=4.43×10-4 ppb·s-1) to HCHO dominance in summer (P(OH)HCHO=5.22×10-4 ppb·s-1). A pronounced elevated HONO enhancement layer emerged near 2.4 km during summer, driven by intensified heterogeneous conversion, with a peak contribution of 30.6 % and a conversion rate C(HONO) of 0.053 h-1. By contrast, near surface OH production at the rural site remained consistently dominated by biogenic HCHO in both spring and summer (P(OH)HCHO=1.82×10-3 ppb·s-1). These findings challenge the conventional assumption that heterogeneous chemistry is confined to the near surface atmosphere. They further provide critical vertical constraints for three-dimensional atmospheric chemistry models and offer a mechanistic explanation for the limited effectiveness of surface-based NOx mitigation strategies under vertically decoupled upper-atmospheric photochemistry.
- Preprint
(2543 KB) - Metadata XML
-
Supplement
(1121 KB) - BibTeX
- EndNote
Status: open (until 12 Aug 2026)
- RC1: 'Comment on egusphere-2026-3286', Anonymous Referee #1, 16 Jul 2026 reply
-
RC2: 'Comment on egusphere-2026-3286', Anonymous Referee #2, 16 Jul 2026
reply
This study utilizes ground-based hyperspectral stereoscopic remote sensing technology to conduct an in-depth investigation of the vertical structure and seasonal evolution of AOC over typical urban (AHU) and rural (CF) sites in the Yangtze-Huai River Basin during the spring and summer of 2023. By quantifying the vertical contributions of HONO, HCHO, and O3 photolysis to OH radical production, the authors innovatively reveal a distinct vertical stratification of oxidation mechanisms within the boundary layer. The study also identifies an anomalous HONO heterogeneous conversion layer at approximately 2.4 km altitude during summer in the urban area and comprehensively elucidates the spatial decoupling of oxidation driving mechanisms between urban and rural environments. These findings challenge the conventional assumption that heterogeneous reactions are confined to the near-surface atmosphere, provide critical vertical observational constraints for three-dimensional atmospheric chemistry models, and offer a mechanistic explanation for the limited effectiveness of current surface-based NOx mitigation strategies. The research is rigorous, innovative, and of significant scientific value, making it highly suitable for publication in ACP.
- Lines 117-120: The AHU and CF sites were selected with azimuth angles of 353° and 209°, respectively. Please clarify whether there are any physical obstructions at low elevation angles for these two azimuth directions, and briefly discuss the spatial representativeness of these two sites.
- Lines 118-119: For the descriptions “(AHU, 31.78°N, 117.20°E, a commercial and traffic intensive area)” and “(CF, 32.21°N, 117.18°E, a farm and agricultural area)”, it is recommended to use semicolons to separate the parallel descriptive phrases within the parentheses for better readability.
- Lines 123-131: Differences in instrument models or configurations between the two sites could introduce systematic observational biases. Please briefly state in the methodology section whether the spectral resolutions (FWHM) of the instruments deployed at AHU and CF are identical.
- Line 126: In the phrase “spectral range: UV: 296–408 nm, VIS: 420–565 nm”, the repeated use of colons is somewhat redundant. It is recommended to revise this to “spectral range: UV (296–408 nm) and VIS (420–565 nm)”.
- Line 146: The DOAS retrieval exclusively utilizes data with SZA < 75°. Please clarify the rationale behind selecting this specific solar zenith angle threshold.
- Lines 147-150: The spectral fitting strictly filters out data with RMS > 1×10-3. What is the justification for choosing this specific threshold? Please compare this criterion with similar MAX-DOAS studies to demonstrate the rationality of this data filtering standard.
- Lines 136-150: Regarding the DOAS fitting process, please elaborate on how the reference spectrum was selected and detail the procedures for dark current and noise subtraction.
- Lines 169-171: The total error analysis indicates retrieval errors of 24%, 33%, and 22% for HONO, HCHO, and O3, respectively. It is recommended to provide a breakdown of the contribution from each error source (e.g., smoothing error, noise error) to clearly identify the primary sources of uncertainty.
- Lines 277-280: The explanation for the high near-surface HONO/NO2 ratio at the rural site in summer attributes the phenomenon to agricultural fertilization and soil emissions. Please provide records of agricultural activities (e.g., fertilization dates and types) around the CF site during the observation period to substantiate this conclusion.
- Lines 343-346: The ratio of secondary HCHO to NO2 (FNR) is used to determine O3 formation sensitivity. Please explain the methodology used to effectively distinguish between primary and secondary HCHO.
- Lines 347-348: In the O3 formation sensitivity analysis, a transition zone is defined based on the intersection of the SHCHO and SNO2 slopes. Please discuss the advantages of using this slope-intersection method from linear fitting when processing highly scattered vertical observation data.
- Lines 397-399: The calculation formulas for the OH production rate are consistent with mainstream methods. However, it is recommended to supplement an estimation of the contribution from the HO2 + NO → OH reaction to evaluate its potential impact on total OH production.
- Lines 451-455: The four columns in Figure 6 represent different sites and seasons. Due to significant magnitude differences, a unified color scale might obscure the diurnal variation characteristics in low-value regions. Please verify the appropriateness of the color scale settings; consider using a logarithmic scale or independent color scales if necessary.
- Lines 457-468: Regarding the quantitative AOC metrics in formulas (6) and (7), only the photolysis pathways of HCHO, HONO, and O3 are calculated. It must be explicitly stated that this does not represent the strictly defined total AOC (as it excludes pathways such as ozone reactions with alkenes and NO3 radicals).
- Lines 462-464: AOC is defined here as the daytime total OH production rate. It is recommended to add a brief discussion on nighttime AOC (e.g., the contribution of NO3 radicals) to clearly define the scope of applicability for this study’s conclusions.
- Lines 495-500: Figure 8 reveals a strong correlation between AOC and aerosol loading (R = 0.88–0.93). Although the authors note that AOC promotes secondary aerosol formation, high aerosol concentrations simultaneously alter the radiation field, thereby inhibiting photolysis rates and AOC. It is suggested to add a qualitative description of this “feedback inhibition” effect of aerosols on AOC to provide a more comprehensive discussion of the causal relationship.
- Lines 551-552: Regarding the limitations in spatial representativeness, it is recommended to elaborate on the emission structural characteristics of the Yangtze-Huai River Basin (e.g., the intertwining of dense urban agglomerations and large-scale agricultural areas) to explain why the findings of this study hold universal value for extrapolation to other similar urban agglomerations.
Citation: https://doi.org/10.5194/egusphere-2026-3286-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 32 | 10 | 6 | 48 | 4 | 4 | 5 |
- HTML: 32
- PDF: 10
- XML: 6
- Total: 48
- Supplement: 4
- BibTeX: 4
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review for the Manuscript "Vertical Structure and Seasonal Evolution of Atmospheric Oxidizing Capacity across Urban and Rural Regions: Observational Constraints from OH Radical Production Pathways" by Zou et al. for EGUSPHERE / ACP .
The manuscript by Zou et al. deals with a determination of the main pathways for OH production over a urban and a remote location in China in the lower troposphere. For that purpose, DOAS measurements for the main components for potential OH production have been obtained and their temporal evolution is analysed. Afterwards, with the help of a photolysis rate model, the respective OH production pathways and efficiencies are determined and the relative importance of the individual branches of OH production are analysed.
Overall, the manuscript deals with an interesting topic, but in conclusion it does not become obvious how and whether these findings can be generalised or a subject to the local conditions. Furthermore, the methodolody does not fully become clear, how the conclusion can be reached (see details below). I do not agree with some of the motivation andd concluding remarks, i.e., that these processes and new findings must be considered in chemistry transport models, as this is exactly done in ALL chemistry transport models, namely the inclusion of multiple reaction pathways and their relative importance is considered by treating the chemistry by a set of chemical differential equations, at least as long as all three photolysis rates are included.
The general view and results presented in 3.4 - which are the main findings of the study according to title and abstract - are fine from my point of view.
Overall, there are still many flaws and weaknesses in this manuscript, which must be revised with major changes before a publication can be re-considered.
Major comments:
1) You use measurements under 11 elevation levels. It is not clear to me, how from this you can get vertical profiles with detailed information on a much higher resolution as shown in Fig.2 where you have almost 50 to 100 data points in the vertical. Of course, you can consider a vertical profile (a-priori) with multiple levels, but your different elevation angles provide you with more limited vertical independent information. You only get 11 different SCDs from which you have to generate a vertical profile.
So you have to clarify, how you obtain this high vertical resolution, which might be crucial for all subsequent analysis ! If this is part of VLIDORT and your optical estimation method, elaborate this a bit better!
2) For the TUV calculations, you use AOD from your own retrievals. How is the AOD vertically distributed? If this is the result shown in Fig.S5, where does the high vertical resolution originate from? How does the data look like at the other station? How much of this information depends on the initial profile assumption? Can elevated aerosol layers above the PBL be measured or retrieved? Why is there such a strong decline in aerosol load after 3pm LT?
And do the O3 values for TUV include the DOAS measurements or is it assumed that all O3 is located above, i.e., mainly the stratospheric column with the TROPOMI satellite data?
3) What is the detection limit and accuracy level for the DOAS measurements? The HONO values are often < 100 pptv which could already be close to detection? Into which uncertainty for mixing ratios do the retrieval or fit uncertainties propagate, especially including the vertical distribution (point 1)?
This becomes even more important, when the direct emissions of HONO are excluded by your selected correction mechanism. The correlation for this correction with a value of R=0.75 is not very strong. Nevertheless, the remaining (secondary) HONO is < 50 pptv. You should discuss whether this quantity can reliably be determined from your measurements ! Furthermore, as OH production from HONO originates from the photolysis of HONO (independent whether HONO is primarily emitted or secondarily formed) you should justify the restriction to secondary HONO only, even though all HONO can photolyse and produce OH.
4) The heterogeneous formation of HONO from NO2 uptake on wet (aerosol) surfaces can be a source of HONO. However, this HONO will only degas under certain conditions (enhanced acidity, drying / evaporation of water from the aerosol up to crystallisatsion thresholds). Otherwise the HONO will not be a gas phase reservoir and consequently a potential gaseous OH source. If aerosol HONO will photolyse also the produced OH will react in the aerosol phase and not degas to contribute to "classical" (= gas phase) oxidation capacity.
5) The boundary layer height is not discussed at all throughout the manuscript. Even though the trace gas profiles show distinct characteristics of the diurnal boundary layer, especially in suzmmer for HCHO, this effect is not considered or discussed at all, e.g., whether the O3 related OH production is only happening above the PBL and whether the vertical gradients in the considered compounds are related to the PBL dynamics. A distinction between a morning boundary layer and potential effects of the residual layer from the previous day(s) is not done, even though Fig 2. indicates this, e.g., double peaks for HCHO at AHU and complex structure of O3 in the lowest 1.5 kilometers, which can originate from the chemical effects as well as from boundary layer dynamics.
6) How is cloudiness taken into account in the measurements? DOAS retrievals will only be possible to a certain degree under cloudy or broken-cloudy conditions, substantially influencing the retrieval. Does this cause a sampling bias, e.g., in the afternoon hours with more active continental convective activity?
7) What is the data source for the NO, NOx or NO2 data? What are the corresponding uncertainty limits? Is NO2 also retreived from the DOAS? Or is it just a surface measurement, only available at one location? How is the secondary HONO determined at the other location if there is no NOx data? And how are the C_HONO tvalues containing the HONO to NOX ratios in Fig.4 calculated? If NOx is determined from an assumed constant(?) ratio of HONO/NOx, then this has a very high uncertainty. On the other hand, using this relation afterwards to determine C_HONO woud not allow for any variability in NOx independent of HONO.
8) The Chapter about daytime O3 sources should in my opinion be removed, as it does not really contribute to the question of OH abundance and the sources of OH. If it is supposed to remain in the manuscript it must be substantially modified as at the moment the conclusions are in my opinion erroneously formulated. The analysis shows that near the surface there is a lot of NOx and HONO (being a primary OH source), therefore the conclusion that this region is NOx limited is simply wrong (Fig.5). If there is more NOx than HCHO in the determination of the FNR, then the system is VOC limited and not NOx limited. At maximum, the lowermost layers with enhanced HONO (and consequently NOx) are transition of even VOC limited. In the layers above the surface in summer, the authors state that this is a NOx limited regime, but do not provide any information where the NOx data have been derived from! In spring, the analysis states that in the lower layers, there is a VOC limitation which is not that obvious from Fig. 2m which shows also in spring and most dominant substantial HCHO mixing ratios in altitudes up to 1.5 km altitude. All of this appears in either a transition regime. If this kind of analysis is conducted you have to have proper NOx vertical profiles and show the FNR instead ! If the results are intended to show different things than those which I interpreted from the graphic, then this should be substantially clarified.
9) The link to the aerosol particles in 3.5 are over-simplifying the aerosol system in my point of view. If all the aerosol would originate from oxidised organics, this relation could hold. However, no chemical analysis of the aerosol is undertaken. Consequently, especially the urban aerosol will contain many additional compounds (sulfate, nitrate, ammonium) which is only indirectly affected by the oxidation capacity and therefore those connections are only weak and can be dominated by the effects of e.g., boundary layer height and boundary layer mixing. The correlation between AOC and aerosols can therefore originate from co-correlated variables and not necessarily from the chemical relations.