the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Oct 2022
Long-term detection, mapping, and interpretation of the trend of ozone in China (1978–2020) by constructing long-term consistent ozone datasets
Rongqi Tang
Jingping Wang
Dujuan Ma
Qicheng Zeng
Jianguang Wen
Qing Xiao
Abstract. The ozone distribution characteristics in the stratosphere or troposphere are worth to be clarified due to their positive/negative impact on climate and human health. Nevertheless, the vertical distribution characteristics of ozone in China have not been fully understood either due to the limited time period of individual satellite records or the inconsistency of the accuracy of ozone products between different satellite records. In response to this challenge, this study first identified the vertical sensitivity of AIRS in detecting trends and verified the sensitivity in the near ground using in-situ measurements. Moreover, these different satellite records were cross-validated in order to check their consistency. In order to construct long-term, consistent ozone datasets dating back to the 1970s was constructed by intercalibrating the ozone products of different satellites using the cumulative distribution function with consideration of the vertical sensitivity. The distribution of ozone in the stratosphere and troposphere was then identified at several altitude layers (i.e., 3 km, 5 km, 12 km, 26 km, 31 km, and 34 km) with obvious interannual variation. The results indicate the seasonal variation of ozone is more significant in the troposphere while the interannual variation of ozone is more significant in the stratosphere. The spatiotemporal variation of ozone in the stratosphere shows a strong dependence on altitudes, and opposite results can be found at different altitudes. The ozone in the troposphere does not present significant interannual variations but shows distinct regional distribution characteristics in the Qinghai Tibet Plateau and Inner Mongolia.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(3063 KB)
Rongqi Tang et al.
Interactive discussion
Status: closed
-
RC1: 'Comment on amt-2022-282', Anonymous Referee #1, 14 Nov 2022
General comments:
The paper by Tang et al. performed some analysis of satellite and in situ ozone data. The basic design or idea of the work is valid and could be interesting. The results mainly rely on SBUVs, AIRS, and in situ data. Some basic analyses were done and some comparison and data harmonization work were been performed. However, there are many major issues with the results and conclusions. The editing of the work itself also needs more effort. I only provided some of the issues I found throughout my reading. I feel the work needs more improvement before it can be published. Unfortunately, I would not recommend a publication.
Major comments:
L24-26. This claim is very strange to me. Please provide how this “more significant” is defined.
L26-28. I think the author wants to say this spatiotemporal variation of stratospheric ozone has a strong altitude dependency, which might even have opposite features. Anyway, this is ambiguous to me, and I am not sure I fully understand the meaning of the sentence.
L28. Please clarify what is “the ozone”, i.e., tropospheric column ozone or surface ozone?
L64-69. There are many papers that studied the trends of ozone partial columns and even surface ozone. I would suggest the author include at least several here. E.g., Tarasick et al., 2016; Weber et al., 2022.
L81. AERONET cannot provide any total ozone information, CIMEL sunphotometer is for observations of aerosol optical depth (AOD) and related properties. This is a wrong or at least misleading claim. Please rewrite this.
L104. Well, it’s a strong claim and might not be true. I.e., similar studies have been performed. Please rewrite this.
L123-125. Please clarify which portion of the ozone is described here, total column, stratospheric, tropospheric, or surface ozone (or maybe all?)?
L135. Change “high levels of ozone” to “high surface ozone levels”.
L138-141. Well, please be aware that the satellite has very limited sensitivity to surface ozone. This is determined by the fact that most nadir-viewing instrument samples the total column, in which 90% is from stratospheric ozone. It is very difficult to separate the surface ozone signals. Anyway, the point is, with current techniques, the satellite still cannot replace in situ to provide reliable and/or high-accuracy surface ozone trends.
L163. A proper reference for SBUV/TOMS V8 algorithm is needed.
L186-187. It is logical to me that the author would provide some information on the data resolution here (the one author generated).
L213. Give the coverages of the 28 layers. Also, proper references are needed for this entire paragraph.
L266. Well, after Fig. 2, here the figure number jumps to 7. Some basic editing is missing in this work and some proofreading is needed.
Figure. 3. For any year, with 200+ sites, even using monthly averaged data I would expect more co-incident observations. Why there are only 104–140 points per year? Have any filters been applied? This looks strange.
Figure 4. What is the unit? What is the colour bar stand for? These must be included in the figure and its caption. The regression results also need to be provided, not just RMSE and R.
L453-460. I can not see how the authors can see the larger variation trend throughout the troposphere using Fig. 7. To detect a trend, one must do some calculations. Or, at least, plot something better than Figure 7. I only can see jammed lines.
L512-520. The interpretation of the seasonal variation has too many issues. The author must understand the meaning of whisker plots. Just for example, for 31 km results, we see much more dynamics in wintertime (i.e., see the box and whiskers, not just the mean).
Figure 10. Units for the first two columns are missing. For the 3rd column, there is no description of this “slope of ozone” anywhere. In the caption (L609-610), there are two descriptions for the left columns.
L719-723. I do not want to be harsh. But, this is nothing new. Note that all SBUV series are pretty similar instruments, while AIRS are something very different in terms of observation techniques. Also, even if this is a “new” finding, one should at least provide some quantified numbers here to support the claim.
Technical issues:
L17: Define AIRS.
L45. Use proper notation for wavelength.
L58. Define TOMS and OMI.
L60. Change N to °N, same for S.
L81. Define AERONET.
L83. Capitalize “Gosat” to GOSAT, and define it.
L86. Not that, even if you defined AIRS in your abstract, in the body part of the paper, you must redefine it.
L97. SBUV. Also, definition is needed.
L131. Define NOx and VOC.
L152. SBUV.
L188. Change NOAA_16 to NOAA-16.
L208. Change “Initial” to “initial”.
L218. Use proper notation for the unit. It’s not ug, but µg. Same for other places.
L226. Change “hp” to “hPa”.
L251. Change “hp” to “hPa”. Same for other places.
L271. CDF has already been defined.
L337. Figure 3. Y label is wrong. Change “ddetected” to “detected”.
L463. Please do not use * for the multiplication sign. Same for other places.
Reference
Tarasick, D. W., Davies, J., Smit, H. G. J., and Oltmans, S. J.: A re-evaluated Canadian ozonesonde record: measurements of the vertical distribution of ozone over Canada from 1966 to 2013, Atmos. Meas. Tech., 9, 195–214, https://doi.org/10.5194/amt-9-195-2016, 2016.
Weber, M., Arosio, C., Coldewey-Egbers, M., Fioletov, V. E., Frith, S. M., Wild, J. D., Tourpali, K., Burrows, J. P., and Loyola, D.: Global total ozone recovery trends attributed to ozone-depleting substance (ODS) changes derived from five merged ozone datasets, Atmos. Chem. Phys., 22, 6843–6859, https://doi.org/10.5194/acp-22-6843-2022, 2022.
Citation: https://doi.org/10.5194/amt-2022-282-RC1 -
RC2: 'Comment on amt-2022-282', Anonymous Referee #2, 26 Nov 2022
General Comments: The paper written by Tang et al. attempted to analyze the vertical structure of ozone using various satellite data. It may be meaningful in that it analyzed various satellite data and in situ ground measurement data for long-term period. However, it seems that a lot of revision and improvement is needed in the organization of outline, interpretation, and method in this paper. In particular, the results were also not sufficiently explained even though there was a difference from previous studies. In my opinion, it will take a lot of time to improve and modify, so it will be difficult to recommend a publication in its current status.
Major comments:
1) L22-24: Inconsistent with the statement in the results section. In the result section, the interannual variation was not evident in the lower troposphere.
2) L26-28: This sentence is very vague and difficult to understand. What are the opposite characteristics?
3) Introduction Section: Overall, it seems necessary to mention more recent research on trends in tropospheric ozone over East Asia. Recently, many studies on the ozone trend in East Asia have been published, but are not mentioned in this study. I recommend following studies.
-Zhang et al., (2021) Long-term ozone variability in the vertical structure and integrated column over the North China Plain: results based on ozone sonde and Dobson measurements during 2001–2019
-Shin et al. (2021) Total Ozone Trends in East Asia from Long-Term Satellite and Ground Observations
-Zhu et al. (2022) Satellite-Based Long-Term Spatiotemporal Patterns of Surface Ozone Concentrations in China: 2005–2019
-Wang et al. (2019) TwentyâFive Years of Lower Tropospheric Ozone Observations in Tropical East Asia: The Influence of Emissions and Weather Patterns
4) L101-104: Does this mean that ozone exchange at different altitudes within the troposphere will also be identified?
5) Section 2.1: If the main description of Section 2.1 is to classify China into six sub-regions (Northest, North, East, Northwest, Southwest, and Central South..), why not present an explanation of the classification of sub-areas fist and then explain the reasons for this classification by climate and regional emission characteristics?
6) L115-122: References related to the monsoon climate in China should be presented.
7) L132-133: Reference Required
8) L159-168, L169-174: Descriptions of each satellite, sensor, and output data require references.
9) L194: Is the spatial resolution of the AIRS data analyzed in this study 13.5 km?
The horizontal resolution of satellite data seems to be important for trend analysis, but the horizontal resolution of satellite data used in this study is not clearly described. If grided satellite data (level 2 or level 3) is used, the horizontal resolution should be presented.
10) L191-200: References related to AIRS data must be added.
11) L238-240: Did you perform statistical analysis other than R and RMSE? Other statistical analysis results were not seen in the result section. If you used only two indexes, you should change “several indexes” to “two indexes”. However, further statistical analysis is recommended.
12) L252-L254: What is the basis for excluding data points with absolute differences lager than 20μg/m3 from the comparative analysis?
This is a very strange analysis method for me. It's like tuning to make the results look good.
13) L276-281: Reference to CDF?
14) L311: What are the references to the "simple empirical formula" and equation (5)? Is this "simple empirical formula" a method applied in some previous studies?
15) Figure 3: Figure 3 shows the comparison analysis for whole Chinese areas? The range of study region should be presented. In addition, the sample sizes (number of data) in Figure 3 are 146~163. But, if daily data is used in this analysis, it should be much more data samples because the number of sites is greater than about…200 (as shown in Figure 1).
16) L345-347: "in situ measurement" means herein a "sonde measurement"? If so, it should be stated that the “in situ measurement” in this sentence means sonde observation data for the purpose of distinguishing it from the ground measurement data described in the above paragraph.
17) Section 4.1.2: If Section 4.1.2 only presents R and RMSE values, the average value of each data must be presented for the interpretation of RMSE values.
18) L449-490: Detailed description of this method is required. This calculation seems to be important to the results, but it is presented too briefly. Also, what area is for “spatial mean” here?
19) Figure 7: Figure 7 should be modified. Additional graphs showing zoom-in for 1) the lower troposphere (3-6 km) and 2) peak altitude (26-31 km) should be added for meaningful analysis. In the current graph, the difference between the highest and lowest values is so huge that it seems to be no interannual changes of O3 in the upper and lower layers, so it seems meaningless.
20) Figure 8 and Section 4.3.1: Interannual variation was analyzed through Figure 8 in “4.3.1. Monthly variation” section. In the box-and-whisker plot in Figure 8, it is necessary to describe what each box and whisker mean. In particular, it is necessary to explain whether the box range means interannual variation. In addition, in order to accurately analyze whether there is a large increase or decrease in the annual O3 variation, it is necessary to analyze the interannual variation for the six regions like Figure 9.
21) L572-L577: Need to mention in more detail why the two groups differ.
22) L582-584: Need a detailed explanation of what “atmospheric circulation” means.
23) L599-601: Why does O3 trends at only 34 km not increase with latitude?
24) In Figure 10-(f), the reason why the slope decreases evenly according to the latitude is there any issue due to satellite sensor aging? Why does Slope show this pattern?
25) L615: Reference is required to refer to “general variation laws”
Minor comments:
1) Define AIRS, AERONET, GOSAT, TOMS, OMI, etc. without using abbreviations when words first appear.
2) All "hp" in the paper should be replaced with "hPa".
3) The ug/m3 that appears occasionally in the paper should be corrected to μg/m3.
4) L387: put a period at the end of sentence.
5) L457: TROOMI --> TROPOMI
6) L490: from 1978 --> from 1978 to 2020?
7) Figure 12: “Spetember” --> “September”
8) L661-662: Provide the approximate range of latitude and longitude for QTP and North China Plain.
Citation: https://doi.org/10.5194/amt-2022-282-RC2
Interactive discussion
Status: closed
-
RC1: 'Comment on amt-2022-282', Anonymous Referee #1, 14 Nov 2022
General comments:
The paper by Tang et al. performed some analysis of satellite and in situ ozone data. The basic design or idea of the work is valid and could be interesting. The results mainly rely on SBUVs, AIRS, and in situ data. Some basic analyses were done and some comparison and data harmonization work were been performed. However, there are many major issues with the results and conclusions. The editing of the work itself also needs more effort. I only provided some of the issues I found throughout my reading. I feel the work needs more improvement before it can be published. Unfortunately, I would not recommend a publication.
Major comments:
L24-26. This claim is very strange to me. Please provide how this “more significant” is defined.
L26-28. I think the author wants to say this spatiotemporal variation of stratospheric ozone has a strong altitude dependency, which might even have opposite features. Anyway, this is ambiguous to me, and I am not sure I fully understand the meaning of the sentence.
L28. Please clarify what is “the ozone”, i.e., tropospheric column ozone or surface ozone?
L64-69. There are many papers that studied the trends of ozone partial columns and even surface ozone. I would suggest the author include at least several here. E.g., Tarasick et al., 2016; Weber et al., 2022.
L81. AERONET cannot provide any total ozone information, CIMEL sunphotometer is for observations of aerosol optical depth (AOD) and related properties. This is a wrong or at least misleading claim. Please rewrite this.
L104. Well, it’s a strong claim and might not be true. I.e., similar studies have been performed. Please rewrite this.
L123-125. Please clarify which portion of the ozone is described here, total column, stratospheric, tropospheric, or surface ozone (or maybe all?)?
L135. Change “high levels of ozone” to “high surface ozone levels”.
L138-141. Well, please be aware that the satellite has very limited sensitivity to surface ozone. This is determined by the fact that most nadir-viewing instrument samples the total column, in which 90% is from stratospheric ozone. It is very difficult to separate the surface ozone signals. Anyway, the point is, with current techniques, the satellite still cannot replace in situ to provide reliable and/or high-accuracy surface ozone trends.
L163. A proper reference for SBUV/TOMS V8 algorithm is needed.
L186-187. It is logical to me that the author would provide some information on the data resolution here (the one author generated).
L213. Give the coverages of the 28 layers. Also, proper references are needed for this entire paragraph.
L266. Well, after Fig. 2, here the figure number jumps to 7. Some basic editing is missing in this work and some proofreading is needed.
Figure. 3. For any year, with 200+ sites, even using monthly averaged data I would expect more co-incident observations. Why there are only 104–140 points per year? Have any filters been applied? This looks strange.
Figure 4. What is the unit? What is the colour bar stand for? These must be included in the figure and its caption. The regression results also need to be provided, not just RMSE and R.
L453-460. I can not see how the authors can see the larger variation trend throughout the troposphere using Fig. 7. To detect a trend, one must do some calculations. Or, at least, plot something better than Figure 7. I only can see jammed lines.
L512-520. The interpretation of the seasonal variation has too many issues. The author must understand the meaning of whisker plots. Just for example, for 31 km results, we see much more dynamics in wintertime (i.e., see the box and whiskers, not just the mean).
Figure 10. Units for the first two columns are missing. For the 3rd column, there is no description of this “slope of ozone” anywhere. In the caption (L609-610), there are two descriptions for the left columns.
L719-723. I do not want to be harsh. But, this is nothing new. Note that all SBUV series are pretty similar instruments, while AIRS are something very different in terms of observation techniques. Also, even if this is a “new” finding, one should at least provide some quantified numbers here to support the claim.
Technical issues:
L17: Define AIRS.
L45. Use proper notation for wavelength.
L58. Define TOMS and OMI.
L60. Change N to °N, same for S.
L81. Define AERONET.
L83. Capitalize “Gosat” to GOSAT, and define it.
L86. Not that, even if you defined AIRS in your abstract, in the body part of the paper, you must redefine it.
L97. SBUV. Also, definition is needed.
L131. Define NOx and VOC.
L152. SBUV.
L188. Change NOAA_16 to NOAA-16.
L208. Change “Initial” to “initial”.
L218. Use proper notation for the unit. It’s not ug, but µg. Same for other places.
L226. Change “hp” to “hPa”.
L251. Change “hp” to “hPa”. Same for other places.
L271. CDF has already been defined.
L337. Figure 3. Y label is wrong. Change “ddetected” to “detected”.
L463. Please do not use * for the multiplication sign. Same for other places.
Reference
Tarasick, D. W., Davies, J., Smit, H. G. J., and Oltmans, S. J.: A re-evaluated Canadian ozonesonde record: measurements of the vertical distribution of ozone over Canada from 1966 to 2013, Atmos. Meas. Tech., 9, 195–214, https://doi.org/10.5194/amt-9-195-2016, 2016.
Weber, M., Arosio, C., Coldewey-Egbers, M., Fioletov, V. E., Frith, S. M., Wild, J. D., Tourpali, K., Burrows, J. P., and Loyola, D.: Global total ozone recovery trends attributed to ozone-depleting substance (ODS) changes derived from five merged ozone datasets, Atmos. Chem. Phys., 22, 6843–6859, https://doi.org/10.5194/acp-22-6843-2022, 2022.
Citation: https://doi.org/10.5194/amt-2022-282-RC1 -
RC2: 'Comment on amt-2022-282', Anonymous Referee #2, 26 Nov 2022
General Comments: The paper written by Tang et al. attempted to analyze the vertical structure of ozone using various satellite data. It may be meaningful in that it analyzed various satellite data and in situ ground measurement data for long-term period. However, it seems that a lot of revision and improvement is needed in the organization of outline, interpretation, and method in this paper. In particular, the results were also not sufficiently explained even though there was a difference from previous studies. In my opinion, it will take a lot of time to improve and modify, so it will be difficult to recommend a publication in its current status.
Major comments:
1) L22-24: Inconsistent with the statement in the results section. In the result section, the interannual variation was not evident in the lower troposphere.
2) L26-28: This sentence is very vague and difficult to understand. What are the opposite characteristics?
3) Introduction Section: Overall, it seems necessary to mention more recent research on trends in tropospheric ozone over East Asia. Recently, many studies on the ozone trend in East Asia have been published, but are not mentioned in this study. I recommend following studies.
-Zhang et al., (2021) Long-term ozone variability in the vertical structure and integrated column over the North China Plain: results based on ozone sonde and Dobson measurements during 2001–2019
-Shin et al. (2021) Total Ozone Trends in East Asia from Long-Term Satellite and Ground Observations
-Zhu et al. (2022) Satellite-Based Long-Term Spatiotemporal Patterns of Surface Ozone Concentrations in China: 2005–2019
-Wang et al. (2019) TwentyâFive Years of Lower Tropospheric Ozone Observations in Tropical East Asia: The Influence of Emissions and Weather Patterns
4) L101-104: Does this mean that ozone exchange at different altitudes within the troposphere will also be identified?
5) Section 2.1: If the main description of Section 2.1 is to classify China into six sub-regions (Northest, North, East, Northwest, Southwest, and Central South..), why not present an explanation of the classification of sub-areas fist and then explain the reasons for this classification by climate and regional emission characteristics?
6) L115-122: References related to the monsoon climate in China should be presented.
7) L132-133: Reference Required
8) L159-168, L169-174: Descriptions of each satellite, sensor, and output data require references.
9) L194: Is the spatial resolution of the AIRS data analyzed in this study 13.5 km?
The horizontal resolution of satellite data seems to be important for trend analysis, but the horizontal resolution of satellite data used in this study is not clearly described. If grided satellite data (level 2 or level 3) is used, the horizontal resolution should be presented.
10) L191-200: References related to AIRS data must be added.
11) L238-240: Did you perform statistical analysis other than R and RMSE? Other statistical analysis results were not seen in the result section. If you used only two indexes, you should change “several indexes” to “two indexes”. However, further statistical analysis is recommended.
12) L252-L254: What is the basis for excluding data points with absolute differences lager than 20μg/m3 from the comparative analysis?
This is a very strange analysis method for me. It's like tuning to make the results look good.
13) L276-281: Reference to CDF?
14) L311: What are the references to the "simple empirical formula" and equation (5)? Is this "simple empirical formula" a method applied in some previous studies?
15) Figure 3: Figure 3 shows the comparison analysis for whole Chinese areas? The range of study region should be presented. In addition, the sample sizes (number of data) in Figure 3 are 146~163. But, if daily data is used in this analysis, it should be much more data samples because the number of sites is greater than about…200 (as shown in Figure 1).
16) L345-347: "in situ measurement" means herein a "sonde measurement"? If so, it should be stated that the “in situ measurement” in this sentence means sonde observation data for the purpose of distinguishing it from the ground measurement data described in the above paragraph.
17) Section 4.1.2: If Section 4.1.2 only presents R and RMSE values, the average value of each data must be presented for the interpretation of RMSE values.
18) L449-490: Detailed description of this method is required. This calculation seems to be important to the results, but it is presented too briefly. Also, what area is for “spatial mean” here?
19) Figure 7: Figure 7 should be modified. Additional graphs showing zoom-in for 1) the lower troposphere (3-6 km) and 2) peak altitude (26-31 km) should be added for meaningful analysis. In the current graph, the difference between the highest and lowest values is so huge that it seems to be no interannual changes of O3 in the upper and lower layers, so it seems meaningless.
20) Figure 8 and Section 4.3.1: Interannual variation was analyzed through Figure 8 in “4.3.1. Monthly variation” section. In the box-and-whisker plot in Figure 8, it is necessary to describe what each box and whisker mean. In particular, it is necessary to explain whether the box range means interannual variation. In addition, in order to accurately analyze whether there is a large increase or decrease in the annual O3 variation, it is necessary to analyze the interannual variation for the six regions like Figure 9.
21) L572-L577: Need to mention in more detail why the two groups differ.
22) L582-584: Need a detailed explanation of what “atmospheric circulation” means.
23) L599-601: Why does O3 trends at only 34 km not increase with latitude?
24) In Figure 10-(f), the reason why the slope decreases evenly according to the latitude is there any issue due to satellite sensor aging? Why does Slope show this pattern?
25) L615: Reference is required to refer to “general variation laws”
Minor comments:
1) Define AIRS, AERONET, GOSAT, TOMS, OMI, etc. without using abbreviations when words first appear.
2) All "hp" in the paper should be replaced with "hPa".
3) The ug/m3 that appears occasionally in the paper should be corrected to μg/m3.
4) L387: put a period at the end of sentence.
5) L457: TROOMI --> TROPOMI
6) L490: from 1978 --> from 1978 to 2020?
7) Figure 12: “Spetember” --> “September”
8) L661-662: Provide the approximate range of latitude and longitude for QTP and North China Plain.
Citation: https://doi.org/10.5194/amt-2022-282-RC2
Rongqi Tang et al.
Rongqi Tang et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 279 | 101 | 15 | 395 | 7 | 11 |
- HTML: 279
- PDF: 101
- XML: 15
- Total: 395
- BibTeX: 7
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1