the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 30 Aug 2021
Impact of the COVID-19 pandemic on the observed vertical distributions of PM2.5, NOx, and O3 from a tower in the Pearl River Delta
Abstract. The outbreak of the 2019 novel coronavirus (COVID-19) has brought tremendous impact and influence on human health and social economy around the world. The lockdown implemented in China, starting on 23 January 2020, led to large reductions in human activities and the associated emissions. Sharp declines in primary pollution provided a unique chance to examine the relationships between anthropogenic emissions and air quality. Here, we report measurements of air pollutants and meteorological parameters at different heights on a tall tower in the Pearl River Delta, China, to investigate the response of the vertical scales of pollutants to reductions in human activities. Compared to the pre-lockdown period (starting from 16 December 2019), the observations showed that surface layer NOx, PM2.5 and mean values of the daily maximum 8 h average O3 (MDA8O3) had significant reductions of 76.8 %, 49.4 %, and 18.6 % respectively, but the average O3 increased (9.7 %) during lockdown period. The vertical profiles of NOx and O3 changed during the lockdown period, but not those of PM2.5. The correlation between PM2.5 and O3 was statistically significant, but not that between PM2.5 and NOx for data collected at four different heights during the lockdown period. The significance of these correlations was the opposite during the pre-lockdown period, indicating that the main composition of PM2.5 has changed dramatically since the lockdown, which is transited from primary aerosol dominating or nitrate dominating (affected by NOx) before lockdown to secondary organic aerosol dominant dominating (affected by O3) during the lockdown. We find weaker diurnal variation of O3 during the lockdown period is similar to the case at background regions. O3 concentrations were not sensitive to NOx concentrations during lockdown, which implies that O3 levels during the lockdown are more representative of the regional background, for which anthropogenic emissions are low and photochemical formation is not a significant ozone source. This evidence suggests that significant reductions of anthropogenic emissions are effective in simultaneous mitigation of PM2.5 and O3 levels.
- Preprint
(1403 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on acp-2021-579', Anonymous Referee #1, 23 Sep 2021
The outbreak of the 2019 novel coronavirus (COVID-19) has brought tremendous impact on human health and social economy. Sharp declines in primary pollution provided a unique chance to examine the relationships between anthropogenic emissions and air quality. The author investigated the vertical structure of pollutants by the highest meteorological tower in ShenZhen City. They found that O3 concentrations were not sensitive to NOx concentrations during lockdown, which implies that O3 levels during the lockdown are more representative of the regional background. They deduced that reductions of anthropogenic emissions are effective to decline PM2.5 and O3 pollutant levels in the Pearl River Delta. Minor revisions are required before acceptance. Comments:
1. How are the instruments on the tower calibrated and maintained on the meteorological tower to ensure data qualityï¼The methods need to be explained in the second section?
2. In Figure 5, why are the concentrations of PM2.5, O3 and NOx higher up than at the surface?
3. In Figure 9 and 10, how about the local photochemistry in different periods?
4. It is suggested to add motor vehicle data in the article, to explain the change of emission from pre-lockdown to lockdown..Citation: https://doi.org/10.5194/acp-2021-579-RC1 - AC1: 'Reply on RC1', Lei Li, 17 Nov 2021
-
RC2: 'Comment on acp-2021-579', Anonymous Referee #2, 23 Sep 2021
General Comments
The Covid-19 lockdown provides a unique opportunity for assessing the effects of substantial emission reductions on atmospheric chemistry. This paper used ambient measurements from a tower situated in the Pearl River Delta of China to explore the response of air pollutants to the lockdown. While this paper is within the scope of ACP, the present paper is limited to a cursory data analysis, without significant contribution to our existing knowledge. The absence of sound analysis accompanied by a lack of in-depth discussion of the observed phenomenon make this paper unpublishable in the present form. Besides the lack of novel insights, I found the manuscript overall hard to follow due to lots of typos throughout the manuscript. While addressing the specific comments below may improve the paper, I don’t think these improvements could justify publication in ACP. Therefore, I would recommend this paper to be rejected.
Major comments:
1) Line 33-34: Why photochemical reactions are not considered as a significant ozone source? While anthropogenic emissions are low during the lockdown period, the oxidation of biogenic VOC still contributes to the ozone formation given the elevated BVOC emissions over the PRD.2) Line 150-158: Please consider simplifying these statements since they are not key contents regarding scientific publication.
3) Line 196: The authors mention that “observed at different heights”. Does the data presented in Figure 1(a)-(c) represent the average value of vertical observations? Please clarify.
4) Line 205-209: I suggest removing these contents since the MDA8 ozone is a well-known indicator that is used to infer the magnitude of ozone pollution.
5) Line 210: Please clarify the potential reasons for this phenomenon. Possibly attributed to less titration effects of NO because primary NOx emission substantially decreased.
6) Line 216-219: Please clarify the reason for the comparison of air pollutants and meteorological parameters between 2017 and 2020. Does the meteorological condition quite similar? Otherwise, it is not comparable.
7) Line 257-259: While the observations from the tower depict insignificant variations in meteorological parameters, the mesoscale process and large-scale synoptic pattern could still alter the air pollutants levels. I don’t think the evidence is sufficient to make the conclusion.
8) Line 270-272: The authors indicate that nearby forests constrain the dispersion of PM while this explanation appears contradictory to the phenomenon that PM levels at 110-120m are higher than ground-level. The authors should comment on this interesting behavior.
9) Line 331-332: actually, ozone levels did not increase at night (as seen in Figure 5). The elevated ratio clearly demonstrates less effective NOx titration at night which leads to relatively higher ozone concentrations at night compared to pre-covid.
10) Line 383: The PRD is well-known for the substantial biogenic VOCs emitted from vegetation and it is highly possible that the enhanced HCHO column depicted by TROPOMI is attributed to the oxidation of biogenic hydrocarbons. I suggest the authors discuss the type of trees nearby the tower (possibly broadleaf trees that have strong BVOC emission potential).
11) Line 413: I am confused by the statement that PM and ozone don’t have related source. As widely acknowledged in-field measurements and laboratory work, both NOx and VOCs from anthropogenic and biogenic are important precursors for secondary pollutants (PM, ozone).
Minor comments:
1) Please clarify the aim of this study in the last paragraph of the Introduction.2) The grammar is in need of much attention. I suggest the authors carefully read through the manuscript and correct typos.
Technique typos:
Line 30: remove “dominant”.
Line 81: scholars->studies
Line 92: air quality factors->air pollutants levels
Line 106: population->residents
Line 113: Do you mean “favorable to the accumulation and formation of air pollutants”?
Line 145: Elements->parameters
Line 155: her->the
Line 174: substantial ->significant
Line 301: prevention->mitigation
Line 319: furtherly->further
Line 435-437: remove “to be”.
Line 480: curves->patternCitation: https://doi.org/10.5194/acp-2021-579-RC2 - AC2: 'Reply on RC2', Lei Li, 17 Nov 2021
-
RC3: 'Comment on acp-2021-579', Anonymous Referee #3, 02 Oct 2021
This manuscript reports the vertical distributions of PM2.5, NOx and O3 at a tower in Pearl River Delta, China, before and during the COVID-19 lockdown, and analyzed the variations of these pollutants from different aspects and tried to give the responsible reasons. The authors finally concluded that the reductions of anthropogenic emissions by the lockdown were effective in mitigating both PM2.5 and O3. Although the datasets provided in this manuscript are unique and interesting, the organization and analysis is not sound and scientific. Many explanations for the observed data trends were too arbitrary, with little or even no evidence. Importantly, the role of meteorology on the observed trends was underestimated or even neglected. In addition, despite no PM2.5 composition data in this manuscript, a large amount of papers on PM2.5 in the PRD were not cited to support the data analysis. In my point of view, this is a very primary manuscript and no solid science has been drawn. It still has a long distance to publication at a high-level journal like ACP or similar journals, and thus I recommend rejecting it.
Major comments:
1. There seem to be more pollutants which are routinely monitored in China’s air quality network, such as SO2, PM10 and CO. Why these data are not included? They can provide more useful information on primary emissions. In addition, I don't understand why the authors included the 2017 data. As the authors said, both the emissions and weather in 2017 were different from those in 2020.
2. Since meteorology largely influences or even dominates the variations of air pollutants, analysis of major meteorological parameters must always accompany the explorations of the variations of air pollutants. The comparison of averages of meteorological elements in Table 1 is far less than enough. Line 251 “it can be concluded that the meteorological conditions of the Shenzhen region were largely identical before and during the lockdown”, this is too arbitrary. Since the vertical profiles/diurnal variations of the air pollutants in this study are not large (in Fig. 7), the simultaneous variations of temperature, RH and wind should be seriously analyzed. Unfortunately, meteorological analysis seemed to be totally forgotten starting from Section 3.2.
3. Section 3.3. Why the authors did not include the ground data in this section? The current vertical trends were represented by only three altitudes. If you add ground data in Fig. 7, will the vertical trends be the same?
4. There are too many subjective inferences without support or citation in the data analysis, such as:
Line 267: “During night and early morning, the height of the mixing layer top is between 120m and 220m, so the curves of the upper and lower layers are quite different.” Any evidence for the height of the mixing layer top?
Line 270: “This may have been caused by the presence of dense forests near the ground observation point (Shiyan Base), which may have obstructed the dispersal of particulate matter and thus reduced the apparent PM2.5 concentration.” I think wind vertical profile may also influence it.
Line 275: “High-height PM2.5 is formed predominantly by chemical reactions, whereas low-height PM2.5 may be derived from multiple sources (predominantly surface-level primary emissions).” This sentence is not sound. In the literature, secondary aerosols account for the major part of PM2.5 in PRD even at the ground level. The influence of reginal transport of secondary aerosols at different altitudes was not discussed here.
Line 288: “It is likely that the morning peak of the pre-lockdown curve was caused by direct emissions from nearby human activities. These emissions were therefore greatly reduced by the lockdown-mediated decrease in human activity and were more easily blocked by the dense forest around the ground observation point.” Such analysis is too subjective.
Line 312: “Since NOx is a primary pollutant, its significantly lower concentrations at the low altitudes implies that near-ground chemical reactions consume it more rapidly than the high-altitude chemical reactions do.” What reactions? Why do they consume NOx more rapidly?
Line 333: “A possible reason leading to this phenomenon is that in the area where the SZMGT is located, the key height of night chemical reactions may be around 110-120 m.” Such analysis is irresponsible.
Lines 357-370. The discussion here is generally weak. How about PM2.5 at the ground? What is the role of wind profile and/or regional transport? The authors may firstly check whether all vertical differences are statistically significant, especially when considering the measurement accuracy. “At the middle level and above, PM2.5 is formed mainly by photochemical 360 reactions (Li et al. 2020)”, such citation is invalid. The vertical profile in Taiwan could be totally different and cannot support the discussion.
Line 413: “It may be inferred that prior to the lockdown, PM2.5 and O3 did not have related sources. However, during the lockdown, both were likely to have a similar source.” Such discussion is too casual.
Line 423: “At lower heights, a considerable part of PM2.5 is primary source and had nothing to do with photochemical reactions”, such analysis is irresponsible.
Line 431: “the primary aerosol like black carbon is not reduced,” where are the black carbon data?Minor comments:
Lines 272-273: what are the reasons for the peak occurring at 17:00–19:00?
Line 298: a typo for mush.
Line 326: the high value can last to about 18:00, not only for 8:00 to 10:00.
Lines 347-349: how to calculate nitrate radical production rate?
Lines 406-411: such basic description is not necessary.
Figures 9 and 10: the correlations were also affected by the data range and amount.
Lines 462-464: this sentence contradicts itself and should be rewritten.Citation: https://doi.org/10.5194/acp-2021-579-RC3 - AC3: 'Reply on RC3', Lei Li, 17 Nov 2021
Status: closed
-
RC1: 'Comment on acp-2021-579', Anonymous Referee #1, 23 Sep 2021
The outbreak of the 2019 novel coronavirus (COVID-19) has brought tremendous impact on human health and social economy. Sharp declines in primary pollution provided a unique chance to examine the relationships between anthropogenic emissions and air quality. The author investigated the vertical structure of pollutants by the highest meteorological tower in ShenZhen City. They found that O3 concentrations were not sensitive to NOx concentrations during lockdown, which implies that O3 levels during the lockdown are more representative of the regional background. They deduced that reductions of anthropogenic emissions are effective to decline PM2.5 and O3 pollutant levels in the Pearl River Delta. Minor revisions are required before acceptance. Comments:
1. How are the instruments on the tower calibrated and maintained on the meteorological tower to ensure data qualityï¼The methods need to be explained in the second section?
2. In Figure 5, why are the concentrations of PM2.5, O3 and NOx higher up than at the surface?
3. In Figure 9 and 10, how about the local photochemistry in different periods?
4. It is suggested to add motor vehicle data in the article, to explain the change of emission from pre-lockdown to lockdown..Citation: https://doi.org/10.5194/acp-2021-579-RC1 - AC1: 'Reply on RC1', Lei Li, 17 Nov 2021
-
RC2: 'Comment on acp-2021-579', Anonymous Referee #2, 23 Sep 2021
General Comments
The Covid-19 lockdown provides a unique opportunity for assessing the effects of substantial emission reductions on atmospheric chemistry. This paper used ambient measurements from a tower situated in the Pearl River Delta of China to explore the response of air pollutants to the lockdown. While this paper is within the scope of ACP, the present paper is limited to a cursory data analysis, without significant contribution to our existing knowledge. The absence of sound analysis accompanied by a lack of in-depth discussion of the observed phenomenon make this paper unpublishable in the present form. Besides the lack of novel insights, I found the manuscript overall hard to follow due to lots of typos throughout the manuscript. While addressing the specific comments below may improve the paper, I don’t think these improvements could justify publication in ACP. Therefore, I would recommend this paper to be rejected.
Major comments:
1) Line 33-34: Why photochemical reactions are not considered as a significant ozone source? While anthropogenic emissions are low during the lockdown period, the oxidation of biogenic VOC still contributes to the ozone formation given the elevated BVOC emissions over the PRD.2) Line 150-158: Please consider simplifying these statements since they are not key contents regarding scientific publication.
3) Line 196: The authors mention that “observed at different heights”. Does the data presented in Figure 1(a)-(c) represent the average value of vertical observations? Please clarify.
4) Line 205-209: I suggest removing these contents since the MDA8 ozone is a well-known indicator that is used to infer the magnitude of ozone pollution.
5) Line 210: Please clarify the potential reasons for this phenomenon. Possibly attributed to less titration effects of NO because primary NOx emission substantially decreased.
6) Line 216-219: Please clarify the reason for the comparison of air pollutants and meteorological parameters between 2017 and 2020. Does the meteorological condition quite similar? Otherwise, it is not comparable.
7) Line 257-259: While the observations from the tower depict insignificant variations in meteorological parameters, the mesoscale process and large-scale synoptic pattern could still alter the air pollutants levels. I don’t think the evidence is sufficient to make the conclusion.
8) Line 270-272: The authors indicate that nearby forests constrain the dispersion of PM while this explanation appears contradictory to the phenomenon that PM levels at 110-120m are higher than ground-level. The authors should comment on this interesting behavior.
9) Line 331-332: actually, ozone levels did not increase at night (as seen in Figure 5). The elevated ratio clearly demonstrates less effective NOx titration at night which leads to relatively higher ozone concentrations at night compared to pre-covid.
10) Line 383: The PRD is well-known for the substantial biogenic VOCs emitted from vegetation and it is highly possible that the enhanced HCHO column depicted by TROPOMI is attributed to the oxidation of biogenic hydrocarbons. I suggest the authors discuss the type of trees nearby the tower (possibly broadleaf trees that have strong BVOC emission potential).
11) Line 413: I am confused by the statement that PM and ozone don’t have related source. As widely acknowledged in-field measurements and laboratory work, both NOx and VOCs from anthropogenic and biogenic are important precursors for secondary pollutants (PM, ozone).
Minor comments:
1) Please clarify the aim of this study in the last paragraph of the Introduction.2) The grammar is in need of much attention. I suggest the authors carefully read through the manuscript and correct typos.
Technique typos:
Line 30: remove “dominant”.
Line 81: scholars->studies
Line 92: air quality factors->air pollutants levels
Line 106: population->residents
Line 113: Do you mean “favorable to the accumulation and formation of air pollutants”?
Line 145: Elements->parameters
Line 155: her->the
Line 174: substantial ->significant
Line 301: prevention->mitigation
Line 319: furtherly->further
Line 435-437: remove “to be”.
Line 480: curves->patternCitation: https://doi.org/10.5194/acp-2021-579-RC2 - AC2: 'Reply on RC2', Lei Li, 17 Nov 2021
-
RC3: 'Comment on acp-2021-579', Anonymous Referee #3, 02 Oct 2021
This manuscript reports the vertical distributions of PM2.5, NOx and O3 at a tower in Pearl River Delta, China, before and during the COVID-19 lockdown, and analyzed the variations of these pollutants from different aspects and tried to give the responsible reasons. The authors finally concluded that the reductions of anthropogenic emissions by the lockdown were effective in mitigating both PM2.5 and O3. Although the datasets provided in this manuscript are unique and interesting, the organization and analysis is not sound and scientific. Many explanations for the observed data trends were too arbitrary, with little or even no evidence. Importantly, the role of meteorology on the observed trends was underestimated or even neglected. In addition, despite no PM2.5 composition data in this manuscript, a large amount of papers on PM2.5 in the PRD were not cited to support the data analysis. In my point of view, this is a very primary manuscript and no solid science has been drawn. It still has a long distance to publication at a high-level journal like ACP or similar journals, and thus I recommend rejecting it.
Major comments:
1. There seem to be more pollutants which are routinely monitored in China’s air quality network, such as SO2, PM10 and CO. Why these data are not included? They can provide more useful information on primary emissions. In addition, I don't understand why the authors included the 2017 data. As the authors said, both the emissions and weather in 2017 were different from those in 2020.
2. Since meteorology largely influences or even dominates the variations of air pollutants, analysis of major meteorological parameters must always accompany the explorations of the variations of air pollutants. The comparison of averages of meteorological elements in Table 1 is far less than enough. Line 251 “it can be concluded that the meteorological conditions of the Shenzhen region were largely identical before and during the lockdown”, this is too arbitrary. Since the vertical profiles/diurnal variations of the air pollutants in this study are not large (in Fig. 7), the simultaneous variations of temperature, RH and wind should be seriously analyzed. Unfortunately, meteorological analysis seemed to be totally forgotten starting from Section 3.2.
3. Section 3.3. Why the authors did not include the ground data in this section? The current vertical trends were represented by only three altitudes. If you add ground data in Fig. 7, will the vertical trends be the same?
4. There are too many subjective inferences without support or citation in the data analysis, such as:
Line 267: “During night and early morning, the height of the mixing layer top is between 120m and 220m, so the curves of the upper and lower layers are quite different.” Any evidence for the height of the mixing layer top?
Line 270: “This may have been caused by the presence of dense forests near the ground observation point (Shiyan Base), which may have obstructed the dispersal of particulate matter and thus reduced the apparent PM2.5 concentration.” I think wind vertical profile may also influence it.
Line 275: “High-height PM2.5 is formed predominantly by chemical reactions, whereas low-height PM2.5 may be derived from multiple sources (predominantly surface-level primary emissions).” This sentence is not sound. In the literature, secondary aerosols account for the major part of PM2.5 in PRD even at the ground level. The influence of reginal transport of secondary aerosols at different altitudes was not discussed here.
Line 288: “It is likely that the morning peak of the pre-lockdown curve was caused by direct emissions from nearby human activities. These emissions were therefore greatly reduced by the lockdown-mediated decrease in human activity and were more easily blocked by the dense forest around the ground observation point.” Such analysis is too subjective.
Line 312: “Since NOx is a primary pollutant, its significantly lower concentrations at the low altitudes implies that near-ground chemical reactions consume it more rapidly than the high-altitude chemical reactions do.” What reactions? Why do they consume NOx more rapidly?
Line 333: “A possible reason leading to this phenomenon is that in the area where the SZMGT is located, the key height of night chemical reactions may be around 110-120 m.” Such analysis is irresponsible.
Lines 357-370. The discussion here is generally weak. How about PM2.5 at the ground? What is the role of wind profile and/or regional transport? The authors may firstly check whether all vertical differences are statistically significant, especially when considering the measurement accuracy. “At the middle level and above, PM2.5 is formed mainly by photochemical 360 reactions (Li et al. 2020)”, such citation is invalid. The vertical profile in Taiwan could be totally different and cannot support the discussion.
Line 413: “It may be inferred that prior to the lockdown, PM2.5 and O3 did not have related sources. However, during the lockdown, both were likely to have a similar source.” Such discussion is too casual.
Line 423: “At lower heights, a considerable part of PM2.5 is primary source and had nothing to do with photochemical reactions”, such analysis is irresponsible.
Line 431: “the primary aerosol like black carbon is not reduced,” where are the black carbon data?Minor comments:
Lines 272-273: what are the reasons for the peak occurring at 17:00–19:00?
Line 298: a typo for mush.
Line 326: the high value can last to about 18:00, not only for 8:00 to 10:00.
Lines 347-349: how to calculate nitrate radical production rate?
Lines 406-411: such basic description is not necessary.
Figures 9 and 10: the correlations were also affected by the data range and amount.
Lines 462-464: this sentence contradicts itself and should be rewritten.Citation: https://doi.org/10.5194/acp-2021-579-RC3 - AC3: 'Reply on RC3', Lei Li, 17 Nov 2021
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,365 | 420 | 66 | 1,851 | 39 | 45 |
- HTML: 1,365
- PDF: 420
- XML: 66
- Total: 1,851
- BibTeX: 39
- EndNote: 45
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1