the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Feb 2025
Evolution of tropospheric aerosols over central China during 2010–2024 as observed by lidar
Abstract. Air quality in China has improved significantly over the past decade. However, recent studies show that this progress has notably slowed in recent years. To investigate regional patterns and driving factors, we examined the long-term evolution of tropospheric aerosols over Wuhan (30.5° N, 114.4° E) from 2010 to 2024, using ground-based polarization lidar observations. Aerosol optical depth (AOD) trends are divided into two phases: a declining trend (-0.077 yr-1) during 2010–2017 (stage I) and a fluctuating period during 2018–2024 (stage II). Contributions from natural (dust) and anthropogenic (non-dust) aerosols were analyzed separately. Dust optical depth (DOD) consistently declined (-0.011 yr-1) until August 2020 and became larger again afterwards. In stage I, anthropogenic aerosols (-0.068 yr-1) were responsible for 88.3 % of the total AOD reduction, primarily due to decreases in boundary-layer AOD. In stage II, anthropogenic AOD fluctuated, possibly due to atmospheric chemistry factors. Seasonal variations were also observed. Anthropogenic aerosols appeared from surface to 2.5 km in summer, with particle extinction and mass concentration of 0.12 km-1 and 83.0 μg m-3, which were concentrated below 0.7 km in winter, with much higher particle extinction and mass concentration of 0.31 km-1 and 211.8 μg m-3. Two case studies highlighted typical pollution events: summertime transboundary agricultural biomass burning smoke in June 2014 and wintertime local anthropogenic aerosol pollution in January 2019. These findings improve our understanding of how regional aerosols respond to local emission controls and long-range transport of dust and smoke.
- Preprint
(3239 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 13 Apr 2025)
-
RC1: 'Comment on egusphere-2025-56', Anonymous Referee #1, 25 Feb 2025
reply
My ratings are based on a balance between the main part of the MS (very good, minor revision) and the descriptions of the case studies (poor, major revision).
The authors present a study on the evolution of aerosols over Wuhan during 2010-2024. The study is based on the use of a polarization lidar providing information on the vertical structure of aerosol particles and their polarization state as a source of information on aerosol sphericity to discriminate between dust and non-dust aerosol, where dust is considered natural and non-dust is considered anthropogenic . In support, they use a comprehensive set of data including satellite observations, ground-based monitoring data, air mass back trajectories, radiosonde data and meteorological information. Considering the variation of the lidar-derived AOD, the authors split the time series into two stages. During stage I (2010-2017) the AOD decreases, during stage II (2018-2024) the AOD is on average constant. The variations of the AOD are discussed, reasons for the decrease in Stage I are presented and fluctuations in Stage 2 are discussed. A similar analysis is made for dust and non-dust aerosol and for the latter also at two levels: in free troposphere and in the total atmospheric boundary layer. Time series of aerosol mass concentrations PM2.5 and aerosol precursor gases NO2 and SO2 are presented in support of suggestions of changes in aerosol chemical composition. In addition, two case studies are presented.
The manuscript is overall well-written and the analysis is interesting. However, I have some comments, questions and suggestions for further improvement which need to be addressed before it can be accepted for publication in ACP. In particular, while the main part of the MS is quite clearly presented (minor revision), two case studies have been added which are much less clear and need much clarification (major revision).
Comments
The main subject of this MS is the decrease of the AOD in 2 different periods. However, there is no discussion on how these different periods were determined. Stage I is from 2010 to 2017, but for DOD it ends 2020. However, the DOD variation in stage II 2017-2024 is very small (smaller than for total AOD in stage II), so what is the motivation to extend the DOD fit to 2020? Another question is that for both total AOD and DOD a maximum is observed in 2014 and AOD was high and variable before the decline in 2014. The Clean Air Action plan started in 2013. Would it be more logical to start the fit in 2013? Does it make a difference?
The case studies need to be better explained. In particular, the authors start with Fig. 7 with only information on the major event, on 13 June, in Fig. 7j. It would be more logical when Fig 8a would be shown as introduction to the event, followed by clear explanations of all other figures presented. In addition, the one air mass trajectory shown does not arrive on at the time of the peak PM, why are not more trajectories shown? That one trajectory indicates an anticyclonic circulation, does that explain the day to day UVAI pattern? And the highest UVAI NW off the burnt area on 6.11 and south on 6.12 and the intensification and elongated pattern In 6.13?
In more detail: Fig. 7 a-e: Why is the MODIS corrected reflectance shown? Or is the burnt area (red spots) the most important? Figs a-e show the haze but also the clouds in 3 out of 5 scenes (a, d, e) which, in e obscure all relevant info. What does that mean for the UVAI? Is it measured only above the clouds? Fig e does not show the burnt area because of the cloud cover. Figs f-j indeed show the increasing UVAI, but, in view of the clouds, what is the relation with the surface concentrations in Fig 8a on 6.13? The variation of UVAI in Fig 8a shows remarkable good agreement with PM2.5, likely because it was scaled this way, but in view of the clouds this does not seem credible. Why is Fig 7I shown? What does it tell us? Depol is very small.
Figure 7 shows an air mass trajectory but why for only 1 day? Why not for every day? Or several times on a day? And what does it show? It went back from Wuhan, at 2000LT, and as indicated by dates (m.d), it arrived on the 11th? It probably started low, on the 9th, suddenly was lifted to 3 km and stayed there all day on the 10th and dropped again very fast back to 1 km on the 11th. Is that how we should read it? And then we should look at the other maps when the trajectory overpasses a burnt area and may have picked up smoke. The maps indicate it was on the 10th , between about 33 and 34 N, and the CALIOP track (what was the overpass time? Ascending or descending? What are the red lines at the bottom of the CALIOP figures? Please mention) shows high AOD between 32 and 34 N but at the east of the air mass trajectory, and UVAI (when was OMPS overpass LT?) was not very high under the CALIOP overpass (Fig. g). CALIOP has crossed the air mass trajectory but a little later. The air mass passed over the burnt area on 6.10 and may have picked up some smoke, but UVAI shows a relatively small signal. The green lines in the CALIOP trajectory indicate smoke presence but UVAI is not enhanced. Fig. g to f suggests that smoke had been transported south and arrived over Wuhan. Why are Figures shown for 12 and 13 June? In the text:
L319-320 burnt area over northern Anhui, so why is smoke formed in Henan? How does that relate to the air mass trajectory and the CALIOP and OMPS observations?
In other words: guide the reader and explain what is important!
Likewise, guide us through Fig. 8. The first thing I wonder about is that PM2.5 and CO peak in the evening of 12 June. So why was the air mass trajectory not calculated for the peak PM2.5 and CO time? That would show the sources, right?
Another is that CALIOP overpass is early in the afternoon and your lidar observations started at 16:00 LT, both on the 10th. Why is there no comparison? There is a description between L324 en L339, but I miss guidance. For instance, L333-334; “severe air pollution was observed”: when (date, time, height), etc. Fig b shows me that BLH was about 1.5 km at 16:00 on 11.6, but the temperature profiles was at 20:00, when BLH had dropped to about zero. And depol at that time was close to 0.1 (red) in the BL. (in contrast to L336) and how do these contrasting statements and observation reflect the next sentence? (L336-339). Further questions are how the extinction, and RH profiles are used? These profiles show clear gradients and transitions which may indicate different layers and may be connected with backscatter profiles in Fig 8b. The authors mention high RH. Which however is only shown in Fig 8e and 8g in a thin layer just above the surface. Further above the RH is so low that little or no hygroscopic growth can be expected. And also PDR does not show correlation with he RH variation.
In summary: there is a lot to be said about Figs 7 and 8, please explain.
And this also applies to the haze case in Sect. 4.2. Fig.s d-k are not explained or used, so why are they presented. Figs a-c are used but attributing the increase in PM only to NO3 formation seems a big step. Why did that not happed on the previous days? Rather, the decrease in NO2 in the evening of the 25th, as opposed to increase of nocturnal NO2 due to chemical reactions, could also be due to a change in transport from areas with smaller NO2 emissions.Detailed suggestions
L42: In most of the para the authors discuss AOD, and the slow-down in the decrease of AOD , which is also discussed further below (L165). Therefore It seems a bit strange that on L42 the slow down in PM2.5 is mentioned, which has been reported to respond differently than AOD.
L90 “Note that 𝛼୮ values below 0.3 km were assumed equal to that at 0.3 km, possibly causing an uncertainty of <0.05 in AOD”. Fig. 6 shows seasonally averaged profiles with strong gradients above 300m. Would it be more logical to linearly interpolate to the surface?
L100 suggest to change policy to protocol
L174 attributed
L190 Figure 2a caption “fewer than 15 cloud-free profiles being recorded in a given month” why this restriction to 15 profiles? It is not easy to see in the Figure whether the extinction profiles are monthly averaged. Or is that because you use monthly mean data in Fig b?
L 222 Fig 3 caption: change to “DOD for monthly mean” and “The dark orange line “
L235 “In Wuhan, the non-dust component is primarily attributed to anthropogenic aerosols”. IN the case studies you show the effect of biomass burning aerosol (anthropogenic). Attributed to straw burning. To your knowledge, are there wildfires in the area contributing to the aerosol content, i.e. which are not anthropogenic?
L237 notably attributed
L240-245 The ABL develops throughout the day and aerosols are usually mixed throughout the ABL, with a gradient across the inversion at the top of the BL. Were the BLH be determine from the individual lidar profiles? If not, how was BLH determined?
L261 Fig 4 caption: “as in figure 2”
L 267 & 270 concentrations
L 273 effectiveness of emission control measures?
L281 Profiles are sown in Fig. 6a
L288 combusted or combustion, change to burned or burning (here and all other occurrences)
L307 Fig. 6 caption: seasonally averaged; the same as in Figure ;
L325 UVAI is zero and indicates the presence of UV-absorbing aerosols: is that a typo?
L406 fluctuated with a rate of 0.002 /yr. That would mean that AOD increased with that rate and varied around that line by +/- 0.2
L416 “plenty of aerosols”, do you mean “much aerosol” or “high aerosol concetrations?”
L418&374 AOD increase 6.1 times, do you mean “increased by a factor of 6.1?”
L421 variationsCitation: https://doi.org/10.5194/egusphere-2025-56-RC1
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 126 | 27 | 7 | 160 | 5 | 6 |
- HTML: 126
- PDF: 27
- XML: 7
- Total: 160
- BibTeX: 5
- EndNote: 6
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| China | 1 | 52 | 33 |
| United States of America | 2 | 47 | 29 |
| Germany | 3 | 8 | 5 |
| France | 4 | 7 | 4 |
| India | 5 | 5 | 3 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 52