the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Feb 2026
Measurement report: Airborne Observation of CO2 and CH4 in the Urban Atmospheric Boundary Layer in Eastern China
Abstract. To characterize the concentrations of CO2 and CH4 in the urban atmospheric boundary layer (ABL), this study conducted airborne measurements over four cities in Eastern China, obtaining full vertical profiles (ground to 2 km) over Beijing and Nanjing, partial profiles over Hengshiu and Shangqiu. Results showed that the CO2 and CH4 concentrations in the ABL were consistently higher than those in the free atmosphere, with the highest values observed near the surface (Beijing and Nanjing). In Beijing, the daytime and nighttime inversion jumps in the CO2 concentration were −25.2 ppm and −18.0 ppm, respectively. In Nanjing, the corresponding values were −9.5 ppm and −10.0 ppm. For CH4, the inversion jumps were −171 ppb during the day and −203 ppb at night (Beijing); in Nanjing, they were −141 ppb and −108 ppb, respectively. Change in the airmass trajectory altered the free-atmospheric CO2 concentration over Nanjing by 3 ppm in a matter of a few hours. The CH4:CO2 emissions ratio inferred from the nighttime ABL concentration data was lower than that obtained from the EDGAR inventory by 17 % in Beijing and 82 % in Nanjing, indicating that the inventory may have missed the recent energy transition from gasoline and natural gas to electric in the transport sector. The experimental data is available at https://doi.org/10.7910/DVN/ZPVSVU.
- Preprint
(2159 KB) - Metadata XML
-
Supplement
(736 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-6197', Anonymous Referee #1, 05 Mar 2026
-
RC2: 'Comment on egusphere-2025-6197', Joseph Pitt, 23 Mar 2026
This paper describes airborne CH4 and CO2 measurements made over four Chinese cities during a single flight day in May 2023. It is well written and easy to follow. This is a nice dataset and it could form the basis for an AMT measurement report, but there are several general points that need addressing first.
I think the interpretation of the data above the ABL needs rethinking. As is stated in L360, the concentrations here mainly reflect wider regional emissions and long-range transport. But in plenty of other places this data is used to make statements about local emissions from the cities underneath the profiles. In particular for Hengshui there is no data in the ABL, so I don't see how any inferences can be drawn regarding local fluxes. The gradients in the free troposphere may include some contribution from local fluxes that have been transported out of the boundary layer, but they could also reflect longer-range transport. For Shangqiu the qualitative interpretation might be OK, as the flight does appear to dip into the top of the ABL briefly. But the best fit lines in Figure 8 are not representative of emissions form these cities. The current discussion surrounding this figure needs to be removed. The figure itself could either stay or go - if it stays then the discussion in section 4.2 seems far more convincing (although only in a qualitative sense - the best fit lines are not really useful here). Then I think the subsequent discussion in section 4 should only include Beijing and Nanjing.
When considering the data within the ABL for Beijing and Nanjing, until we get to section 4.2 there appears to be an explicit assumption that these are behaving as "dome" cities rather than "plume" cities (i.e. it is assumed that the storage flux dominates the advective flux). With wind data from the aircraft and/or ground observation sites and/or a weather model this assumption could easily be explored. Even if these are dome cities, it can still be difficult to make representative measurements using vertical profiles with limited horizontal extent within the city limits. Fig. 10 goes some way to addressing this, but really we need to see trajectories from the more enhanced ABL samples and the less enhanced ABL samples. Then we can see what the enhancement ratios represent (i.e. the difference between these two air histories). The wind data from the aircraft seems like a really important dataset here but there is virtually no discussion of it in the paper.
For the comparison against EDGAR - I am not convinced that using the profiles later in the day negates the impact of the biosphere on CO2. During the day there will be CO2 uptake, which will switch to net positive emission at some point in the evening due to biospheric respiration. If these cancel out to produce net-zero biospheric impact on the aircraft profiles, it would only be by pure chance. Some sort of biospheric modelling would be needed to assess the likely impact here. That may be beyond the scope of this measurement report, but at the very least there needs to be a strong caveat placed on the EDGAR comparison stating that the impact of the biosphere is unknown and potentially significant.
It would also be good to dig further into where EDGAR thinks the CH4 emissions are coming from. Is it really dominated by the mobile (vehicle) sector in the inventory? Perhaps it is for Chinese cities, I've never investigated, but I would have guessed most of the urban CH4 emissions in EDGAR would be associated with heating and landfills. The fact that we are looking at data from a single flight in May also needs to be better highlighted, especially if heating emissions could be a significant source. Many emission sources are known to exhibit variability on diurnal, weekly and seasonal timescales so it is not really possible to draw inferences about EDGAR based on a single day of flying (even ignoring the issue with the biosphere)
Another broad issue is the general lack of uncertainty values associed with the quantitative results. For example, for the jump values the standard deviations within and above the ABL could be combined to give some measure of uncertainty due to measurement variability. Considering these might then change some elements of the discussion, because in some cases the difference between values being compared is much smaller than the variability associated with them. Putting uncertainties on the enhancement ratios when comparing to EDGAR would surely result in agreement between the inventory and aircraft data for Beijing within uncertainties. The uncertainties may even be so large that there is no statistically significant difference for Nanjing either.
Specific points:
I am no expert on boundary layer dynamics, but I think the first paragraph in the introduction is quite a simplification. Maybe this is OK, because this is not the focus of the paper, but I suggest that at least the language should be adapted here. For example, L33 could be "These two scalars are typically conserved if the ABL...", and L35 should make it clear that CO2 and CH4 are conserved in the absence of emissions (which is obvious from the discussion in the next paragraph, but also needs stating here).
L60 - there have that observe a change in CH4 mole fraction between the boundary layer and free troposphere within an urban plume, although in most cases this is not the focus of the study and so the "inversion jump" values are not explicitly stated in the text. See next comment...
L74 - this presumably should say "not aware". Maybe this partly answers my previous comment in explicitly stating this is considering only studies directly over urban land. Most other urban studies have been performed downwind of the urban area, sampling the urban plume. It might be worth making this even more explicit, also at L427.
L118 and following lines - "millions" should be "million"
Fig. 1 - the left hand panel in particular is hard to read for colour blind people. Could you change the colour scheme and run it through a colour blind simulator please? Many of the other figures are not ideal either, but this is the worst.
L205 - it could imply this, or it could be another temporal change. With the data here it is impossible to tell - I think this caveat needs adding here
L237 - this could also be due to horizontal gradients - i.e. higher concentrations (on average) along the roads than at the airport. This possibility should be discussed here.
Fig. 7 - the fit to the NJ up profile needs more discussion. There are multiple plausible causes for the structure in the data here. Looking at the profiles, one possibility is that the stable stratification leads to the local plume becoming compressed in altitude relative to the down profile, with a more methane-heavy regional plume sitting on top of it. The enahancement ratio for the lower section looks by eye to be very similar to the NJ down plume, which would support this interpretation, but there are many other possibilities too.
L345 - also differences in the advective flux could explain the difference in concentration. This should definitely be mentioned here.
Fig. 10 - it would be helpful to label the cities that are discussed on the text on the map.
L469 - It would be helpful to expand here on what this study has told us about future study design.
Citation: https://doi.org/10.5194/egusphere-2025-6197-RC2
Data sets
Replication Data for: Airborne Observation of CO2 and CH4 in the Urban Atmospheric Boundary Layer in Eastern China J. Wang et al. https://doi.org/10.7910/DVN/ZPVSVU
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 168 | 105 | 18 | 291 | 39 | 13 | 36 |
- HTML: 168
- PDF: 105
- XML: 18
- Total: 291
- Supplement: 39
- BibTeX: 13
- EndNote: 36
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of “Measurement report: Airborne Observation of CO2 and CH4 in the Urban Atmospheric Boundary Layer in Eastern China” by Wang et al.
This paper presents a study on airborne measurements of CO2 and CH4 in the urban atmospheric boundary layer (ABL) over four cities in China – Beijing, Hengshui, Shangqiu, and Nanjing. The aircraft also conducted vertical profile measurements over each city. In addition, mobile measurements were performed in Nanjing. Using the results from the observations, CH4:CO2 emission ratios were derived and compared with estimates from the EDGAR inventory data.
Comments:
Minor comments: