Pollution Trends in China from 2000 to 2017: A Multi-Sensor View from Space

Round 1

Reviewer 1 Report

In this resubmission, the authors have addressed all my concerns, and I believe they have also adequately addressed the concerns of the other reviewer who questioned the novelty; I believe this paper does contribute significantly to understanding of how to use a range of remote sensing data to asses trends in air quality, and presents novel results specifically for China. 

I only identified one obvious correction: 

Line 128: "a larger different" should be  "a larger difference"

The English language throughout appears grammatically correct, but could still do with some additional cleaning up throughout by a native speaker if possible, as it does read slightly awkwardly in parts. 

Author Response

Thank you very much for pointing out this typo. I have corrected it in the revised manuscript.

Reviewer 2 Report

The author intends to provide a comprehensive view of pollution trends (namely, AOD, AE, AAI, SO2, NO2, and O3) in China using satellite retrievals (MODIS, OMI, MISR). I have 3 main issues.

Issue 1

The author’s novel proposition in terms of methodology or finding is unclear. China is one of the most studied regions due to its elevated pollutant levels. Researches have already analyzed China from multiple pollutants [1]. Thus it is not clear what is that exact scientific gap which is resolved by this paper. It is therefore highly recommended that the author provide a section that first comprehensively reviews and summarizes previous studies regarding the: 1. Sources of pollutants in China (e.g. winter coal burning, power plants etc.) and their trends with respect to previous emission inventory based studies, 2. Influence of meteorology as well as transboundary impacts, 3. Trends uncovered by other remote sensing based studies and 4. Identify the major gaps in previous study that this study ultimately answers.

Following snapshot list of papers answers the above points to some extent:

[1] Chan, C. K., & Yao, X. (2008). Air pollution in mega cities in China. Atmospheric Environment, 42(1), 1–42. https://doi.org/10.1016/j.atmosenv.2007.09.003

[2] Mi, W., Li, Z., Xia, X., Holben, B., Levy, R., Zhao, F., … Cribb, M. (2007). Evaluation of the Moderate Resolution Imaging Spectroradiometer aerosol products at two Aerosol Robotic Network stations in China. Journal of Geophysical Research, 112, 1–14. https://doi.org/10.1029/2007JD008474

[3] Itahashi, S., Uno, I., Yumimoto, K., Irie, H., Osada, K., Ogata, K., … Ohara, T. (2012). Interannual variation in the fine-mode MODIS aerosol optical depth and its relationship to the changes in sulfur dioxide emissions in China between 2000 and 2010. Atmospheric Chemistry and Physics, 12(5), 2631–2640. https://doi.org/10.5194/acp-12-2631-2012

[4] Rohde RA, Muller RA (2015) Air Pollution in China: Mapping of Concentrations and Sources. PLoS ONE 10(8): e0135749. https://doi.org/10.1371/journal.pone.0135749

[5] Misra, P., Fujikawa, A., & Takeuchi, W. (2017). Novel decomposition scheme for characterizing urban air quality with MODIS. Remote Sensing, 9(8), 1–19. https://doi.org/10.3390/rs9080812

[6] Liang, F., Xiao, Q., Wang, Y., Lyapustin, A., Li, G., Gu, D., … Liu, Y. (2017). MAIAC-based long-term spatiotemporal trends of PM2.5 in Beijing, China. The Science of the Total Environment, 616–617, 1589–1598. https://doi.org/10.1016/j.scitotenv.2017.10.155

[7]Ma, Z., Hu, X., Sayer, A. M., Levy, R., Zhang, Q., Xue, Y., … Liu, Y. (2016). Satellite-based spatiotemporal trends in PM2.5 concentrations: China, 2004-2013. Environmental Health Perspectives, 124(2), 184–192. https://doi.org/10.1289/ehp.1409481

[8]Zhao, B., Jiang, J. H., Gu, Y., Diner, D., Worden, J., Liou, K. N., … Huang, L. (2017). Decadal-scale trends in regional aerosol particle properties and their linkage to emission changes. Environmental Research Letters, 12(5). https://doi.org/10.1088/1748-9326/aa6cb2

[9]Wang, Y., Shen, L., Wu, S., Mickley, L., He, J., & Hao, J. (2013). Sensitivity of surface ozone over China to 2000-2050 global changes of climate and emissions. Atmospheric Environment, 75(x), 374–382. https://doi.org/10.1016/j.atmosenv.2013.04.045

[10]Wang, S., Xing, J., Chatani, S., Hao, J., Klimont, Z., Cofala, J., & Amann, M. (2011). Verification of anthropogenic emissions of China by satellite and ground observations. Atmospheric Environment, 45(35), 6347–6358. https://doi.org/10.1016/j.atmosenv.2011.08.054

Issue 2.

The plots and trends remains vague (even though there is a brief discussion statistical discussion).  It is not clear what quality flag or temporal resolution of imagery was used to estimate the trend. Whether it was daily, weekly or monthly (Section 2.2.1) is unclear. As demonstrated by [11], the choice of averaging and aggregating methods can alter the inferences. How the choice of such parameters can alter (or not) your inferences must be demonstrated to instill confidence in the robustness of the trends.

An additional criticism is that scientific objective behind performing multiple regression, specially since SO2 and NO2 are likely to multi-collinearity issues. Please test this. This modeling needs deeper discussion backed by references. Furthermore, the residence lifetimes are influence by seasonal conditions, hence seasonal models must be prepared to investigate the multi-pollutant relationships. The explanation in Line 373 is not sufficient.

[11]Levy, R. C., Leptoukh, G. G., Kahn, R., Zubko, V., Gopalan, A., & Remer, L. A. (2009). A Critical Look at Deriving Monthly Aerosol Optical Depth From Satellite Data. IEEE Transactions on Geoscience and Remote Sensing, 47(8), 2942–2956. https://doi.org/10.1109/TGRS.2009.2013842

Issue 3.

There is absolute lack of references in the discussion and many portions have been left incomplete as “need further investigation” or are speculations (e.g. Line 248, 418, 435, 170, 426, 446, 380). Each of these need to be discussed further with references, specially the claim of transboundary transportation from South Asia to China (I have never read this before).

Similarly the discussion parts are unscientific and superficial. There is almost no discussion based on numerical result. The trends are described by qualitative words like “higher”, “lower”, “strongly decreasing”, etc. when in fact they should be objectively described in numerical terms. Consider also showing the time series plots (similar to Fig 3) in addition to the maps that summarise the overall trends (e.g. as shown in Fig 2, 4-10,13,14). Trend map are too general that abstract interesting details to be found in time series plots. I recommend the author to choose two contrasting region (trendwise) and analyze them as a detailed case study section.

Additional comments:

Title: “Recent” is subjective. Please be objective, e.g. “2000 to 2017”

It seems “observation” has been used instead of retrieval at several places. Please make a distinction between satellite based observations and satellite based retrievals.

What is the basis of selecting , AOD, AE, AAI, SO2, NO2, and O3 and not more pollutants?

Line 77: Why is AOD downsampled from the 0.1degree to 0.5degree. Which algorithm was used?

Line 185. What is the reason for insignificant trends from MISR. Discuss with a time series plot.

Line 218. Elaborate “noisy” and “incomplete”. What is the effect of clouds in the trends.

Line 243. Is this related with rising desertification across NE?

Line 90: AAI is not a commonly used indicator due to qualitative interpretation. Discuss the strong motivation of using AAI in detail, specially its limitations as found by other studies.

Line 274. In which region is decreasing albedo most prominent?

Line 269. How does higher albedo enhance aerosol absorption?

Fig10. Shouldn’t the unit be molec/cm2

Line279. What is the lifetime of ozone over several regions of China

Line 426. Please elaborate this point.

Line442. Please document all the trend related discussions based on actual numbers

Sec 4. Separate them into 2 sections.

Overall English is generally fine but needs grammatical checks, spellings (line 437 “from”) and a concise manner of writing.

Author Response

Please see the attached file for detailed responses.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Most of the suggestions are well-implemented. However, I have one strong suggestion.

The result section (Section 3) still does not quantitatively discuss the the trends. These instill ambiguous interpretations. e.g. Line 235: what is "strongest increase" quantitatively? ; Line 255: how strong are the "strongest positive trends" ?; Line 427: how distinct are the "distinct seasonal differences", and at many other locations. Please bear in mind that your and reader's interpretation of such qualifiers may not be similar.

Minor suggestions:

Consider placing a list of abbreviations

Discussion needs to be before Conclusions

Author Response

Again I thank the review for his helpful comments and suggestions. I have made necessary changes including:

(1) Adding numerical description at the suggested places;

(2) Adding a table showing the list of abbreviations;

(3) Moving discussion before conclusion.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Review for manuscript:

Recent pollution trends in China: a multi-sensor view from space by Jing Li

The article investigates multi-year satellite observations from OMI, MODIS (Terra), and MISR instruments over China as well as introduces a trend analysis of various variables related to air pollution, including NO₂, SO₂, AOD, Angström Exponent, and AAI. As the author him/her self states, that this study is not the first study to report pollution trends in China. Many published articles have shown similar results that are presented in this manuscript, e.g. the decrease in AOD, NO₂ and SO₂ during recent years in China. From this point of view unfortunately this manuscript do not offer significantly new findings on this topic. It is interesting to look at the variation of several satellite-based parameters together, but the analysis remain somewhat thin and conclusions are weak. I’m missing more in depth analysis what is behind these observed changes, and how these trends compare to the multiple studies published on the topic.

Another major comment is related to the concept of Absorbing Aerosol Index trend. From the text I got the impression that the author has not really understood the complexity of interpreting the changes in AAI. The author suggests that “trends of carbonaceous and dust aerosols can be inferred from AAI”. While this is partially true, the paper lacks discussion and explanation on which parameters AAI is dependent on and from which values the AAI trend has been defined (all or accounting only positive values?). Positive AAI values indicate the presence of absorbing aerosol (it has been shown also that high clouds with effective cloud fraction close to 1 can cause slightly positive values of AAI). AAI is a function of aerosol content (AOD), aerosol type (SSA), and aerosol layer height. Hence, for example, an increase in positive AAI can mean simply increasing aerosol layer height. So, my main question is, without some additional information on each case, how you can actually interpret the changes in AAI, case by case, and as a trend? There is some discussion on MODIS AOD trend with respect to OMI AAI trend, but it remains somewhat weak. Besides, the author has chosen to use only MODIS Terra AOD, while OMI observations would be temporally closer to MODIS Aqua.

There are also several minor issues/statements that are not accurate/ are wrong. For example (line 192) “MISR shows much fewer significant trends than MODIS, because its noisy and incomplete time series prevents the detection of a significant trend. “ on what basis is this statement made? The statement on O₃ formation (line 316) is inaccurate: “Moreover, although NO₂ is the major precursor of tropospheric ozone in cities, at rural places, ozone can still form from the oxidation of biogenic VOCs, resulting in less spatial contrast in ozone concentration.” For this I suggest the author to read e.g. a paper by Jin and Holloway, “Spatial and temporal variability of ozone sensitivity over China observed from the Ozone Monitoring Instrument”, JGR 2015. Also, for example statements such as (line 75 on MODIS Terra and Aqua AOD) “But since the results of the two are quite similar, we only report Terra MODIS results to save space” are not good scientific practice, especially without any references.    

Unfortunately it seems that this study is incomplete. Therefore I do not recommend the publication of this manuscript in its current form.

Author Response

Please find the responses to the reviewer in the attached file.

Author Response File: Author Response.docx

Reviewer 2 Report

My comments are described in a separate file.

Comments for author File: Comments.pdf

Author Response

Please find the responses to the reviewer in the attached file.

Author Response File: Author Response.docx

Reviewer 3 Report

This is a very useful and comprehensive study which demonstrates the utility of long-term pollutant monitoring using multiple satellite sensors; from an applied perspective, it demonstrates some success of policy change in China regarding aerosol precursors, but also highlights an increasing and spatially uniform increase in ozone, which is of interest. Methodologically, the use of multiple regression to infer contributing factors to AOD trends is useful and could be applied elsewhere.  The statistical trend and regression analyses are simple but appear rigorous and appropriate. I recommend this paper be accepted subject to a few minor corrections, primarily in formatting and English language.

Specific comments:

There are some minor English language corrections that need to be made, eg.

Line 13 - "produce" should be "product"

Line 17 - "continuous upward trend" should be "a continuous upward trend" - there are occasionally missing articles "a" and "the" throughout the paper that the author should check carefully

Line 36 - "damages" should be "damage"

Line 46 - "still lack" should be "still a lack"

Line 72 - should read "near global coverage every 2 days" 

Line 121 - "profiles" should be "profile"

Other comments:

Line 78 - you state you grid the pixel data to 0.5 x 0.5 degree grids - what summary method was used, eg. the mean or median AOD within the grid cells?

Line 107 - Formatting error: 1 DU = 2.69x1016 - in the submitted manuscript the "16" is not shown as a superscript, rather "1016" is all written as a single number, this should be corrected.

Equation 2 - in the submitted manuscript, this equation is displayed incorrectly - there are two blank boxes after the S= which I assume should contain some symbol.  Also, could you please define in the text the "sgn" function, I'm not sure if this is well known.

Paragraph 3.1 - There appears to be a formatting error here, this entire paragraph is centre-justified and in a larger font. 

Line 211 - You say here that a positive AAI and a negative AE correspond with a "decrease in mean particle size...due to increased dust fraction".  I would assume an increased dust fraction would mean an INCREASE in mean particle size, and your abstract confirms this, where you say "increased fraction of large particles".  I think line 211 needs to be changed to say "increase in mean particle size"?

Author Response

Please find the responses to the reviewer in the attached file.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Thank you for your polite response. I think that the items I pointed out have been improved well.