Opposite effects of aerosols and meteorological parameters on warm clouds in two contrasting regions over eastern China

Liu, Yuqin; Lin, Tao; Zhang, Jiahua; Wang, Fu; Huang, Yiyi; Wu, Xian; Ye, Hong; Zhang, Guoqin; Cao, Xin; de Leeuw, Gerrit

doi:https://doi.org/10.5194/egusphere-2023-1615

Preprints

https://doi.org/10.5194/egusphere-2023-1615

Preprints

15 Aug 2023

| 15 Aug 2023

Opposite effects of aerosols and meteorological parameters on warm clouds in two contrasting regions over eastern China

Yuqin Liu, Tao Lin, Jiahua Zhang, Fu Wang, Yiyi Huang, Xian Wu, Hong Ye, Guoqin Zhang, Xin Cao, and Gerrit de Leeuw

Abstract. Aerosol and cloud properties retrieved from the Moderate Resolution Imaging Spectroradiometer (MODIS) on-board the Aqua satellite were used to investigate aerosol-cloud interaction (ACI) over eastern China, using AOD as a proxy for the aerosol concentration, during a period of 15 years (2008–2022). Two contrasting areas were selected: the heavily polluted Yangtze River Delta (YRD) and a clean area over the East China Sea (ECS). Linear regression analysis shows the opposite behaviour of the CER-AOD relationship in the two different aerosol regimes. CER decreases with the increase of AOD in the moderately polluted atmosphere (0.1<AOD<0.3) over the ECS, in agreement with the Twomey effect. However, in the polluted atmosphere (AOD>0.3) over the YRD, CER increases with increasing AOD. Evaluation of the ACI index (here defined as the change in CER as a function of AOD) as function of the cloud liquid water path (LWP) shows that in the moderately polluted atmosphere over the ECS the ACI index is significant and positive in the LWP interval [40 g m^-2, 200 g m^-2], and increases substantially with increasing LWP. In contrast, in the polluted atmosphere over the YRD the ACI index is significant and negative in the LWP interval [0 g m^-2, 120 g m^-2] and does not change notably as function of LWP in this interval. To further analyse the influence of AOD and meteorological conditions on cloud parameters, the geographical detector method (GDM) has been used. The results show that all factors have a significant influence on the cloud parameters over the ECS, except for CTP, but the influence of AOD is larger than that of any of the meteorological factors. Among the meteorological factors, LTS has the largest influence on the cloud parameters and RH the smallest. Over the YRD, the explanatory power of the effect of AOD and meteorological parameters on cloud parameters is much smaller than over the ECS, except for RH which has a statistically significant influence on CTP and can explain 65 % of the variation of CTP. The combined effect of meteorological factors and AOD on cloud parameters enhances the explanatory power over the effect of individual parameters. The study further shows that over the ECS the effect of RH and LTS on the CER/AOD relationship is opposite to that of PVV. Over the YRD, The CER is larger in unstable atmospheric conditions (low LTS) than in stable conditions, irrespective of the AOD and the CER is much larger in high relative humidity conditions than in low relative humidity conditions.

Received: 14 Jul 2023 – Discussion started: 15 Aug 2023

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 1791 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (1791 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

18 Apr 2024

Opposite effects of aerosols and meteorological parameters on warm clouds in two contrasting regions over eastern China

Yuqin Liu, Tao Lin, Jiahua Zhang, Fu Wang, Yiyi Huang, Xian Wu, Hong Ye, Guoqin Zhang, Xin Cao, and Gerrit de Leeuw

Atmos. Chem. Phys., 24, 4651–4673, https://doi.org/10.5194/acp-24-4651-2024,https://doi.org/10.5194/acp-24-4651-2024, 2024

Short summary

Yuqin Liu, Tao Lin, Jiahua Zhang, Fu Wang, Yiyi Huang, Xian Wu, Hong Ye, Guoqin Zhang, Xin Cao, and Gerrit de Leeuw

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-1615', Anonymous Referee #3, 02 Sep 2023
This study investigates the aerosol and meteorological parameters on warm clouds using satellite measurements. The authors focus on the period of 2008-2022 over the two contrasting regions over eastern China, i.e. Yangtze River Delta, a heavily polluted region in eastern China, and the East China Sea with a relatively clean atmosphere. The interaction between AOD and CER has been investigated by considering different AOD and LWP regimes in the both two different aerosol regimes. A new method (geographical detector method) was applied to explore the relative importance of AOD and meteorological parameters on cloud properties. The content of this manuscript is highly relevant to ACP readers. In general, the manuscript is well organized, and the analysis conducted is quite comprehensive. Based on the overall quality, it is recommended that the manuscript be considered for publication if the specific comments provided are addressed.

Specific comments
Abstract: it would be beneficial if the authors emphasized the overall significance or implications of their study at the end of the abstract.

In order to provide a more comprehensive analysis, it would be beneficial for the authors to compare the results obtained in this study with findings from other regions around the world. By doing so, they can examine the unique aerosol effects on clouds in the specific target region.

Page 5，line 146：add “.” in the end.

Page 6，line 150：change “Eastern China Sea(ECS) area (20°N-28°N, 126°E-134°E)” to “Eastern China Sea area (ECS, 20°N-28°N, 126°E-134°E)”

Page 7，line 191：change “(cloud optical thickness, cloud droplet effective radius, etc.)” to “(COT, CER, etc.)”.

Page 8, line 202: change “Where r_e represents the cloud droplet effective radius (CER)” to “Where r_e represents the CER”.

Page 11, line 276: it suggests to define the acronyms about “NW” and “SW”.

Page 14, line 344: remove right parenthesis.

Page 15, line 365: replace “for the YRD” with “over the YRD”.

Page 21, line 501: add right parenthesis after “by Liu et al., (2017”.

Page 22, line 516: change “for three different LWP intervals” to “for five different LWP intervals”.

Page 16, line 377-380：The statistically significance is used through the manuscript, so it suggests to describe at the first place in the manuscript.
Citation: https://doi.org/10.5194/egusphere-2023-1615-RC1
- AC2: 'Response to Referee #3', Yuqin Liu, 07 Dec 2023
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1615/egusphere-2023-1615-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-1615-AC2
RC2:
'Comment on egusphere-2023-1615', Anonymous Referee #1, 06 Sep 2023

The authors look at the primarily at the controls on the relationship between AOD and cloud effective radius (CER), both meteorological parameters and other cloud properties. Concentrating on two regions (over land and ocean near China), this study also introduces the geographical detector method (GDM) to the study of aerosol-cloud interactions. They also look at the impact of meteorological properties on the relationship between AOD and cloud properties more generally.
The introduction of the GDM to this study is novel and of interest to the readers of ACP. With some extra explanation, I think this paper would make a really useful introduction to the method for others in atmospheric science. However, I have a number of concerns about the paper as a whole that would have to be addressed before I would recommend publication.

Main points
The introduction of the GDM method is a really nice aspect of this study. However, I feel it could be explained and examined in more detail, as there are a number of factors that are unclear to someone meeting this method for the first time. For example, the impact of the explanatory variables in Tables 4 and 5 sum to over 100%. This is not what I would have expected. Similarly, I am not familiar with the 'interaction detector' or the 'interactive q-values'. What do these mean? How should they be interpreted? Likewise, the term 'nonlinear enhancement of the influence of the independent parameters' on L456 is not straightforward to someone new to this method.
The AOD-CER relationship is difficult to interpret as the Twomey effect, especially where LWP is not controlled for (McComiskey and Feingold, 2012). Several previous studies have also investigated the potential controls on the AOD-CER relationship (e.g. Tan et al, 2017; Yuan et al, 2008; Myhre et al, 2007; Tang et al, 2014; Andersen and Cermak, 2015). There should be a clearer distinction around what is added by this work (which can be the GDM).
The majority of more recent studies have used Nd for calculation of the ACI, rather than CER (uaas et al, 2008; Gryspeerdt et al, 2017; McCoy et al, 2018; Hasekamp et al, 2019). There are also useful studies that investigate the susceptibility (AOD-Nd relationship) and the impacts on this value (Jia et al, 2022). This avoids the LWP-CER issue prevalent in previous work and presents a cleaner separation of the Twomey and adjustments. It could be worth including a section on why CER is used and might be something to consider for future work in this area.
Previous studies have shown that it is difficult to interpret correlations over large regions as an aerosol effect due to the impact of meteorological confounders (Grandey and Stier, 2010). Correlations between AOD and cloud properties are also fraught with potential confounding effects (Quaas et al, 2010; Boucher and Quaas, 2012; Gryspeerdt et al, 2014). Does the GDM method address these issues? If so how? If not, this study should be much clearer about the claims of causality it puts forward.
Another important factor is the calculation of LWP that is used for binning in the ACI calculations. As the LWP depends on the CER, does this not lead to an implicit filtering by CER, which would affect the calculation of ACI?
Is there a reason for using AOD, rather than a product such as the aerosol index (Nakajima et al, 2001), which has a stornger link to the CCN concentration?

Specific points
The abstract mostly list results, rather than providing an overview of the paper and the conclusions. Is there an overall picture or aim of the study that could help to structure this?
L39 - Is this opposite effect just because the sign of the pressure vertical velocity is defined differently? I am not sure what opposite means in this context.
L68 - The terms first and second indirect effect are less commonly used in more recent studies. I would suggest referring to adjustments instead (see IPCC AR5), as this more closely links in with the radiative forcing/effective radiative forcing distinction and aligns more closely with other recent work.
L81 - I would have said that satellites typically have a fairly poor temporal resolution (unless the authors are referring to geostationary satellites?)
L93 - Is there any reason for choosing these studies? They seem to be rather disjointed, with some looking at the Twomey effect directly and some considering adjustments. Some notable studies looking at the impact of meteorological parameters on potential adjustments (e.g. Koren et al, 2010) and the particularly Twomey effect (Jones et al, 2009; Jia et al. 2022) are left out.
L96 - PVV is redefined here
L116 - There needs to be some discussion of how the GDM is affected by the results of Grandey and Stier (2010), who suggest that spatial correlations are unreliable. It may be that the results of GS10 are not applicable here, as the GDM method is capable of accounting for the co-variations that drive the results in GS10. If so, it would be good to have some evidence of this, as it would provide more significance to the results presented in this work.
L154 - Why only 2008 to 2022? The MODIS record runs back to 2002/3
L159 - The aerosol and cloud retrievals are necessarily conducted in different regions of the 1x1 degree gridbox (aerosol retrievals are only conducted in clear sky), meaning that they are not coincident. This may not affect the results if the regions are non-precipitating. Jia et al (2022) showed that wet scavenging can have a considerable impact on the susceptibility.
L153 - I would suggest referencing Platnick et al (2016), given the authors are using MODIS collection 6.1.
L169 - Brennan et al (2005) suggests that cloud contamination becomes an issue when the AOD is larger than 0.6. Why is a larger threshold used here?
L172 - Why is 200gm^-2 used as a threshold for the LWP?
L184 - I would suggest putting the URL links in the references or acknowledgements
L187 - ERA5 and ERA Interim both seem to be mentioned at different points in this work. I suggest using only one (preferably ERA5)
L190 - I am not sure this is the definition of the first indirect effect as all of these properties also vary with cloud adjustments.
L192 - Ice nuclei are usually referred to as "ice nucleating particles" (INP) - Vali et al (2015)
L204 - Do the authors mean CCN here (as in Andreae, 2009)
L216 - An explicit list of these parameters, perhaps in the diagram, could be useful for others trying to replicate this study.
L221 - I have not used the Jenks method before, but from what I understand, you have to specify the number of regions/regimes? How is this done and does the number of regions chosen affect the results?
Eq2 - I am not familiar with this method, so might need a bit more explanation. Is sigma here the variance of y within the specified region/regime?
Eq2 - How does this method compare to a more common correlation measure for non-linear relationships, such as Spearman's Rank?
L379 - The p-value for testing here is quoted as 0.01, but elsewhere it appears that 0.1 (a fairly lax criteria) is used.
L422 - The high explanatory power of AOD for CF variations suggests that this method is not actually identifying causal relationships. While strong correlations between AOD and CF have been previously observed, they are likely due to aerosol humidification (Quaas et al, 2010), rather than an aerosol influence. It seems likely the same effect is being observed here, so care should be taken in the presentation of the results not to mis-attribute causality (unless applicable).
L469 - After the introduction of the GDM, sections 4.5 and 4.6 appear to go back to more 'traditional' methods as used by previous paper. I am not sure I really see how these section support the paper in determining the cause of the different ACI values in these regions. It would be good to have a clearer link to the other work performed and how it supports the overall aim and conclusions of the paper.
L601 - Could the authors be more specific on how this study will help improve model parametrisations?

References
Andersen, H., & Cermak, J. (2015). How thermodynamic environments control stratocumulus microphysics and interactions with aerosols. Environmental Research Letters, 10, 024004. https://doi.org/10.1088/1748-9326/10/2/024004
Andreae, M. O. (2009). Correlation between cloud condensation nuclei concentration and aerosol optical thickness in remote and polluted regions. Atmospheric Chemistry and Physics, 9(2), 543–556. https://doi.org/10.5194/acp-9-543-2009
Boucher, O., & Quaas, J. (2012). Water vapour affects both rain and aerosol optical depth. Nature Geoscience, 6(1), 4–5. https://doi.org/10.1038/ngeo1692
Grandey, B. S., & Stier, P. (2010). A critical look at spatial scale choices in satellite-based aerosol indirect effect studies. Atmospheric Chemistry and Physics, 10(23), 11459–11470. https://doi.org/10.5194/acp-10-11459-2010
Gryspeerdt, E., Stier, P., & Grandey, B. S. (2014). Cloud fraction mediates the aerosol optical depth-cloud top height relationship. Geophysical Research Letters, 41, 3622–3627. https://doi.org/10.1002/2014GL059524
Gryspeerdt, Edward, Quaas, J., Ferrachat, S., Gettelman, A., Ghan, S., Lohmann, U., et al. (2017). Constraining the instantaneous aerosol influence on cloud albedo. Proceedings of the National Academy of Sciences of the United States of America, 114(19), 4899–4904. https://doi.org/10.1073/pnas.1617765114
Hasekamp, O. P., Gryspeerdt, E., & Quaas, J. (2019). Analysis of polarimetric satellite measurements suggests stronger cooling due to aerosol-cloud interactions. Nature Communications, 10(1). https://doi.org/10.1038/s41467-019-13372-2
Jia, H., Quaas, J., Gryspeerdt, E., Böhm, C., & Sourdeval, O. (2022). Addressing the difficulties in quantifying droplet number response to aerosol from satellite observations. Atmospheric Chemistry and Physics, 22(11), 7353–7372. https://doi.org/10.5194/acp-22-7353-2022
Jones, T. A., Christopher, S. A., & Quaas, J. (2009). A six year satellite-based assessment of the regional variations in aerosol indirect effects. Atmospheric Chemistry and Physics, 9, 4091.
Koren, I., Feingold, G., & Remer, L. A. (2010). The invigoration of deep convective clouds over the Atlantic: aerosol effect, meteorology or retrieval artifact? Atmospheric Chemistry and Physics, 10(18), 8855–8872. https://doi.org/10.5194/acp-10-8855-2010
McComiskey, A., & Feingold, G. (2012). The scale problem in quantifying aerosol indirect effects. Atmospheric Chemistry and Physics, 12, 1031. https://doi.org/10.5194/acp-12-1031-2012
Myhre, G., Stordal, F., Johnsrud, M., Kaufman, Y. J., Rosenfeld, D., Storelvmo, T., et al. (2007). Aerosol-cloud interaction inferred from MODIS satellite data and global aerosol models. Atmospheric Chemistry and Physics, 7(12), 3081–3101. https://doi.org/10.5194/acp-7-3081-2007
Quaas, J., Stevens, B., Stier, P., & Lohmann, U. (2010). Interpreting the cloud cover – aerosol optical depth relationship found in satellite data using a general circulation model. Atmospheric Chemistry and Physics, 10(13), 6129–6135. https://doi.org/10.5194/acp-10-6129-2010
Tan, S., Han, Z., Wang, B., & Shi, G. (2017). Variability in the correlation between satellite-derived liquid cloud droplet effective radius and aerosol index over the northern Pacific Ocean. Tellus B: Chemical and Physical Meteorology, 69(1), 1391656. https://doi.org/10.1080/16000889.2017.1391656
Tang, J., Wang, P., Mickley, L. J., Xia, X., Liao, H., Yue, X., et al. (2014). Positive relationship between liquid cloud droplet effective radius and aerosol optical depth over Eastern China from satellite data. Atmospheric Environment, 84, 244–253. https://doi.org/10.1016/j.atmosenv.2013.08.024
Vali, G., DeMott, P. J., Möhler, O., & Whale, T. F. (2015). Technical Note: A proposal for ice nucleation terminology. Atmospheric Chemistry and Physics, 15(18), 10263–10270. https://doi.org/10.5194/acp-15-10263-2015
Yuan, T., Li, Z., Zhang, R., & Fan, J. (2008). Increase of cloud droplet size with aerosol optical depth: An observation and modeling study. Journal of Geophysical Research, 113(D4). https://doi.org/10.1029/2007JD008632

Citation: https://doi.org/10.5194/egusphere-2023-1615-RC2
- AC1: 'Response to Referee #1', Yuqin Liu, 07 Dec 2023
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1615/egusphere-2023-1615-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-1615-AC1
RC3:
'Comment on egusphere-2023-1615', Anonymous Referee #2, 12 Sep 2023
This study uses correlation and the geographical detector method (GDM) to study relationships between aerosol optical depth (AOD), meteorological indicators, and cloud properties, contrasting a heavily polluted region in mainland China to a cleaner region of the Pacific Ocean influenced by transported pollution. The authors find different signs of the AOD-cloud effective radius relationships and find that AOD explains a very large fraction of variability in cloud properties, especially in the cleaner region.
The manuscript is well written, and the Figures and Tables illustrate the results and discussion well. But I have a mixed opinion about the study. On the one hand, there are innovative aspects, such as the application of the GDM to the aerosol-cloud problem. On the other hand, the study uses large-scale, time-averaged correlations of aerosols and clouds that have been shown to say little about aerosol-cloud interactions. And the impact of AOD on cloud variability that results from the GDM is so large that it would require a strong case to bring confidence in the method and results. On balance, I suggest major revisions to give the authors the chance to justify their results.
Main comments:
Recent papers, and especially Gryspeerdt et al. (2023, 10.5194/acp-23-4115-2023) and Arola et al. (2022, 10.1038/s41467-022-34948-5), have seriously challenged the usefulness of correlating aerosol and cloud parameters as done in the present study. Cloud variability and retrieval errors are such that correlations between aerosol optical depth and cloud properties (CNDC, CER, LWP) can in fact be spurious. That means that a large fraction of past literature on aerosol-cloud interactions (including the studies cited lines 86-107) needs to be look at again critically. Attempts to minimise retrievals uncertainties (lines 172-175) will not address parts of the issues. I think the present study remains interesting (especially the GDM analysis), but the authors need to acknowledge the possibility that the correlations they find do not say much about aerosol-cloud interactions.

The idea of using the GDM is interesting, but it is difficult to make physical sense of the results (Table 4 and 5). First, the q factors do not sum up to 1. What does that mean that AOD explains 87% of CER variability, while RH explains a further 36%? Then, the size of the q factors for AOD in Table 4, and to a lesser extent Table 5, stretches belief. If aerosols were that important in determining cloud properties, then estimating aerosol-cloud radiative forcing would have been very easy. Clearly, something goes wrong here. Is it perhaps that the four variables studied (AOD, RH, LTS, and PVV) are not independent? That the large q-values are simply a symptom of correlations caused by atmospheric circulation in the ECS and YRD? This is essentially what Figure 10 suggests. A strong discussion is needed to support the results.

The GDM assumes that the spatial distributions of independent and dependent variables “should have evident similarities” [line 211]. But at what scale is that assumption true for aerosol-cloud interactions? One could expect the assumption to break down when going down to the scale of a cloud field because clouds evolve after their aerosol-influenced formation phase. Is that a problem?

Other comments:
The name “ACI index” is vague. That quantity is really a sensitivity of cloud effective radius to changes in aerosol optical depth, in a similar way to beta_ln(N)-ln(tau) in Bellouin et al. (2020, doi: 10.1029/2019RG000660)

Lines 26 and 27: “significant” – is that in the statistical sense?

Line 47: “in practice” in the atmospheric sciences. Other disciplines use the term more properly.

Lines 44-54: Those generalities on aerosols are not necessary, so that section could be shortened. In fact, the introduction could start directly from line 54: “Aerosol particles are important for climate…”

Lines 55-78: Note that many papers since Chapter 7 of the IPCC AR5 (Boucher et al. 2013) use the concept of aerosol-radiation and aerosol-cloud interactions and their respective adjustments (e.g., Bellouin et al. 2020, Quaas et al. 2022 10.5194/acp-22-12221-2022). The terms direct/1^st indirect/2nd indirect remain in use in parts of the community, but it would be good to connect to the new terminology.

Lines 85-86: It should be said that using AOD as a proxy for aerosol concentrations when looking at aerosol-cloud interactions raises issues. See Section 6 of Bellouin et al. (2020).

Line 168: Andreae (2009) is often cited as justification for using AOD for looking at aerosol-cloud interactions, but ironically its Figure 1 shows that the correlation only exists across aerosol regimes. For a given regime (as done in the present study) there is essentially no correlation. I could not see why Kourtidis et al. (2015) justifies the use of AOD, but I may have missed it.

Lines 169-170: That assumes that cloud contamination has a lesser impact on smaller AODs. Is that true?

Line 207: “intermittently” Probably not the correct word. Interchangeably?

Line 231: Does q sum up to 1 for all factors considered? Is it also able to quantify an unexplained fraction that could suggest the need for more factors?

Figure 3: What does that Figure tell the reader? It’s impossible to say from its caption or from lines 241-245. The discussion needs to cover each of the panels in turn.

Lines 251-252: “for several different purposes”. Give examples based on the papers cited.

Line 265: “averaged over the years 2008-2022”. Does the study use 14-year averaged distributions, or a less dramatic averaging (e.g., multi-annual monthly means)? How can correlation of distributions averaged over such a long period still inform about the physical correlation between clouds and aerosols?

Lines 285-286: Or it could be due to meteorological factors.

Line 373: Missing word: “in the range of”
Citation: https://doi.org/10.5194/egusphere-2023-1615-RC3
- AC3: 'Response to Referee #2', Yuqin Liu, 07 Dec 2023
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1615/egusphere-2023-1615-AC3-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-1615-AC3

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-1615', Anonymous Referee #3, 02 Sep 2023
This study investigates the aerosol and meteorological parameters on warm clouds using satellite measurements. The authors focus on the period of 2008-2022 over the two contrasting regions over eastern China, i.e. Yangtze River Delta, a heavily polluted region in eastern China, and the East China Sea with a relatively clean atmosphere. The interaction between AOD and CER has been investigated by considering different AOD and LWP regimes in the both two different aerosol regimes. A new method (geographical detector method) was applied to explore the relative importance of AOD and meteorological parameters on cloud properties. The content of this manuscript is highly relevant to ACP readers. In general, the manuscript is well organized, and the analysis conducted is quite comprehensive. Based on the overall quality, it is recommended that the manuscript be considered for publication if the specific comments provided are addressed.

Specific comments
Abstract: it would be beneficial if the authors emphasized the overall significance or implications of their study at the end of the abstract.

In order to provide a more comprehensive analysis, it would be beneficial for the authors to compare the results obtained in this study with findings from other regions around the world. By doing so, they can examine the unique aerosol effects on clouds in the specific target region.

Page 5，line 146：add “.” in the end.

Page 6，line 150：change “Eastern China Sea(ECS) area (20°N-28°N, 126°E-134°E)” to “Eastern China Sea area (ECS, 20°N-28°N, 126°E-134°E)”

Page 7，line 191：change “(cloud optical thickness, cloud droplet effective radius, etc.)” to “(COT, CER, etc.)”.

Page 8, line 202: change “Where r_e represents the cloud droplet effective radius (CER)” to “Where r_e represents the CER”.

Page 11, line 276: it suggests to define the acronyms about “NW” and “SW”.

Page 14, line 344: remove right parenthesis.

Page 15, line 365: replace “for the YRD” with “over the YRD”.

Page 21, line 501: add right parenthesis after “by Liu et al., (2017”.

Page 22, line 516: change “for three different LWP intervals” to “for five different LWP intervals”.

Page 16, line 377-380：The statistically significance is used through the manuscript, so it suggests to describe at the first place in the manuscript.
Citation: https://doi.org/10.5194/egusphere-2023-1615-RC1
- AC2: 'Response to Referee #3', Yuqin Liu, 07 Dec 2023
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1615/egusphere-2023-1615-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-1615-AC2
RC2:
'Comment on egusphere-2023-1615', Anonymous Referee #1, 06 Sep 2023

The authors look at the primarily at the controls on the relationship between AOD and cloud effective radius (CER), both meteorological parameters and other cloud properties. Concentrating on two regions (over land and ocean near China), this study also introduces the geographical detector method (GDM) to the study of aerosol-cloud interactions. They also look at the impact of meteorological properties on the relationship between AOD and cloud properties more generally.
The introduction of the GDM to this study is novel and of interest to the readers of ACP. With some extra explanation, I think this paper would make a really useful introduction to the method for others in atmospheric science. However, I have a number of concerns about the paper as a whole that would have to be addressed before I would recommend publication.

Main points
The introduction of the GDM method is a really nice aspect of this study. However, I feel it could be explained and examined in more detail, as there are a number of factors that are unclear to someone meeting this method for the first time. For example, the impact of the explanatory variables in Tables 4 and 5 sum to over 100%. This is not what I would have expected. Similarly, I am not familiar with the 'interaction detector' or the 'interactive q-values'. What do these mean? How should they be interpreted? Likewise, the term 'nonlinear enhancement of the influence of the independent parameters' on L456 is not straightforward to someone new to this method.
The AOD-CER relationship is difficult to interpret as the Twomey effect, especially where LWP is not controlled for (McComiskey and Feingold, 2012). Several previous studies have also investigated the potential controls on the AOD-CER relationship (e.g. Tan et al, 2017; Yuan et al, 2008; Myhre et al, 2007; Tang et al, 2014; Andersen and Cermak, 2015). There should be a clearer distinction around what is added by this work (which can be the GDM).
The majority of more recent studies have used Nd for calculation of the ACI, rather than CER (uaas et al, 2008; Gryspeerdt et al, 2017; McCoy et al, 2018; Hasekamp et al, 2019). There are also useful studies that investigate the susceptibility (AOD-Nd relationship) and the impacts on this value (Jia et al, 2022). This avoids the LWP-CER issue prevalent in previous work and presents a cleaner separation of the Twomey and adjustments. It could be worth including a section on why CER is used and might be something to consider for future work in this area.
Previous studies have shown that it is difficult to interpret correlations over large regions as an aerosol effect due to the impact of meteorological confounders (Grandey and Stier, 2010). Correlations between AOD and cloud properties are also fraught with potential confounding effects (Quaas et al, 2010; Boucher and Quaas, 2012; Gryspeerdt et al, 2014). Does the GDM method address these issues? If so how? If not, this study should be much clearer about the claims of causality it puts forward.
Another important factor is the calculation of LWP that is used for binning in the ACI calculations. As the LWP depends on the CER, does this not lead to an implicit filtering by CER, which would affect the calculation of ACI?
Is there a reason for using AOD, rather than a product such as the aerosol index (Nakajima et al, 2001), which has a stornger link to the CCN concentration?

Specific points
The abstract mostly list results, rather than providing an overview of the paper and the conclusions. Is there an overall picture or aim of the study that could help to structure this?
L39 - Is this opposite effect just because the sign of the pressure vertical velocity is defined differently? I am not sure what opposite means in this context.
L68 - The terms first and second indirect effect are less commonly used in more recent studies. I would suggest referring to adjustments instead (see IPCC AR5), as this more closely links in with the radiative forcing/effective radiative forcing distinction and aligns more closely with other recent work.
L81 - I would have said that satellites typically have a fairly poor temporal resolution (unless the authors are referring to geostationary satellites?)
L93 - Is there any reason for choosing these studies? They seem to be rather disjointed, with some looking at the Twomey effect directly and some considering adjustments. Some notable studies looking at the impact of meteorological parameters on potential adjustments (e.g. Koren et al, 2010) and the particularly Twomey effect (Jones et al, 2009; Jia et al. 2022) are left out.
L96 - PVV is redefined here
L116 - There needs to be some discussion of how the GDM is affected by the results of Grandey and Stier (2010), who suggest that spatial correlations are unreliable. It may be that the results of GS10 are not applicable here, as the GDM method is capable of accounting for the co-variations that drive the results in GS10. If so, it would be good to have some evidence of this, as it would provide more significance to the results presented in this work.
L154 - Why only 2008 to 2022? The MODIS record runs back to 2002/3
L159 - The aerosol and cloud retrievals are necessarily conducted in different regions of the 1x1 degree gridbox (aerosol retrievals are only conducted in clear sky), meaning that they are not coincident. This may not affect the results if the regions are non-precipitating. Jia et al (2022) showed that wet scavenging can have a considerable impact on the susceptibility.
L153 - I would suggest referencing Platnick et al (2016), given the authors are using MODIS collection 6.1.
L169 - Brennan et al (2005) suggests that cloud contamination becomes an issue when the AOD is larger than 0.6. Why is a larger threshold used here?
L172 - Why is 200gm^-2 used as a threshold for the LWP?
L184 - I would suggest putting the URL links in the references or acknowledgements
L187 - ERA5 and ERA Interim both seem to be mentioned at different points in this work. I suggest using only one (preferably ERA5)
L190 - I am not sure this is the definition of the first indirect effect as all of these properties also vary with cloud adjustments.
L192 - Ice nuclei are usually referred to as "ice nucleating particles" (INP) - Vali et al (2015)
L204 - Do the authors mean CCN here (as in Andreae, 2009)
L216 - An explicit list of these parameters, perhaps in the diagram, could be useful for others trying to replicate this study.
L221 - I have not used the Jenks method before, but from what I understand, you have to specify the number of regions/regimes? How is this done and does the number of regions chosen affect the results?
Eq2 - I am not familiar with this method, so might need a bit more explanation. Is sigma here the variance of y within the specified region/regime?
Eq2 - How does this method compare to a more common correlation measure for non-linear relationships, such as Spearman's Rank?
L379 - The p-value for testing here is quoted as 0.01, but elsewhere it appears that 0.1 (a fairly lax criteria) is used.
L422 - The high explanatory power of AOD for CF variations suggests that this method is not actually identifying causal relationships. While strong correlations between AOD and CF have been previously observed, they are likely due to aerosol humidification (Quaas et al, 2010), rather than an aerosol influence. It seems likely the same effect is being observed here, so care should be taken in the presentation of the results not to mis-attribute causality (unless applicable).
L469 - After the introduction of the GDM, sections 4.5 and 4.6 appear to go back to more 'traditional' methods as used by previous paper. I am not sure I really see how these section support the paper in determining the cause of the different ACI values in these regions. It would be good to have a clearer link to the other work performed and how it supports the overall aim and conclusions of the paper.
L601 - Could the authors be more specific on how this study will help improve model parametrisations?

References
Andersen, H., & Cermak, J. (2015). How thermodynamic environments control stratocumulus microphysics and interactions with aerosols. Environmental Research Letters, 10, 024004. https://doi.org/10.1088/1748-9326/10/2/024004
Andreae, M. O. (2009). Correlation between cloud condensation nuclei concentration and aerosol optical thickness in remote and polluted regions. Atmospheric Chemistry and Physics, 9(2), 543–556. https://doi.org/10.5194/acp-9-543-2009
Boucher, O., & Quaas, J. (2012). Water vapour affects both rain and aerosol optical depth. Nature Geoscience, 6(1), 4–5. https://doi.org/10.1038/ngeo1692
Grandey, B. S., & Stier, P. (2010). A critical look at spatial scale choices in satellite-based aerosol indirect effect studies. Atmospheric Chemistry and Physics, 10(23), 11459–11470. https://doi.org/10.5194/acp-10-11459-2010
Gryspeerdt, E., Stier, P., & Grandey, B. S. (2014). Cloud fraction mediates the aerosol optical depth-cloud top height relationship. Geophysical Research Letters, 41, 3622–3627. https://doi.org/10.1002/2014GL059524
Gryspeerdt, Edward, Quaas, J., Ferrachat, S., Gettelman, A., Ghan, S., Lohmann, U., et al. (2017). Constraining the instantaneous aerosol influence on cloud albedo. Proceedings of the National Academy of Sciences of the United States of America, 114(19), 4899–4904. https://doi.org/10.1073/pnas.1617765114
Hasekamp, O. P., Gryspeerdt, E., & Quaas, J. (2019). Analysis of polarimetric satellite measurements suggests stronger cooling due to aerosol-cloud interactions. Nature Communications, 10(1). https://doi.org/10.1038/s41467-019-13372-2
Jia, H., Quaas, J., Gryspeerdt, E., Böhm, C., & Sourdeval, O. (2022). Addressing the difficulties in quantifying droplet number response to aerosol from satellite observations. Atmospheric Chemistry and Physics, 22(11), 7353–7372. https://doi.org/10.5194/acp-22-7353-2022
Jones, T. A., Christopher, S. A., & Quaas, J. (2009). A six year satellite-based assessment of the regional variations in aerosol indirect effects. Atmospheric Chemistry and Physics, 9, 4091.
Koren, I., Feingold, G., & Remer, L. A. (2010). The invigoration of deep convective clouds over the Atlantic: aerosol effect, meteorology or retrieval artifact? Atmospheric Chemistry and Physics, 10(18), 8855–8872. https://doi.org/10.5194/acp-10-8855-2010
McComiskey, A., & Feingold, G. (2012). The scale problem in quantifying aerosol indirect effects. Atmospheric Chemistry and Physics, 12, 1031. https://doi.org/10.5194/acp-12-1031-2012
Myhre, G., Stordal, F., Johnsrud, M., Kaufman, Y. J., Rosenfeld, D., Storelvmo, T., et al. (2007). Aerosol-cloud interaction inferred from MODIS satellite data and global aerosol models. Atmospheric Chemistry and Physics, 7(12), 3081–3101. https://doi.org/10.5194/acp-7-3081-2007
Quaas, J., Stevens, B., Stier, P., & Lohmann, U. (2010). Interpreting the cloud cover – aerosol optical depth relationship found in satellite data using a general circulation model. Atmospheric Chemistry and Physics, 10(13), 6129–6135. https://doi.org/10.5194/acp-10-6129-2010
Tan, S., Han, Z., Wang, B., & Shi, G. (2017). Variability in the correlation between satellite-derived liquid cloud droplet effective radius and aerosol index over the northern Pacific Ocean. Tellus B: Chemical and Physical Meteorology, 69(1), 1391656. https://doi.org/10.1080/16000889.2017.1391656
Tang, J., Wang, P., Mickley, L. J., Xia, X., Liao, H., Yue, X., et al. (2014). Positive relationship between liquid cloud droplet effective radius and aerosol optical depth over Eastern China from satellite data. Atmospheric Environment, 84, 244–253. https://doi.org/10.1016/j.atmosenv.2013.08.024
Vali, G., DeMott, P. J., Möhler, O., & Whale, T. F. (2015). Technical Note: A proposal for ice nucleation terminology. Atmospheric Chemistry and Physics, 15(18), 10263–10270. https://doi.org/10.5194/acp-15-10263-2015
Yuan, T., Li, Z., Zhang, R., & Fan, J. (2008). Increase of cloud droplet size with aerosol optical depth: An observation and modeling study. Journal of Geophysical Research, 113(D4). https://doi.org/10.1029/2007JD008632

Citation: https://doi.org/10.5194/egusphere-2023-1615-RC2
- AC1: 'Response to Referee #1', Yuqin Liu, 07 Dec 2023
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1615/egusphere-2023-1615-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-1615-AC1
RC3:
'Comment on egusphere-2023-1615', Anonymous Referee #2, 12 Sep 2023
This study uses correlation and the geographical detector method (GDM) to study relationships between aerosol optical depth (AOD), meteorological indicators, and cloud properties, contrasting a heavily polluted region in mainland China to a cleaner region of the Pacific Ocean influenced by transported pollution. The authors find different signs of the AOD-cloud effective radius relationships and find that AOD explains a very large fraction of variability in cloud properties, especially in the cleaner region.
The manuscript is well written, and the Figures and Tables illustrate the results and discussion well. But I have a mixed opinion about the study. On the one hand, there are innovative aspects, such as the application of the GDM to the aerosol-cloud problem. On the other hand, the study uses large-scale, time-averaged correlations of aerosols and clouds that have been shown to say little about aerosol-cloud interactions. And the impact of AOD on cloud variability that results from the GDM is so large that it would require a strong case to bring confidence in the method and results. On balance, I suggest major revisions to give the authors the chance to justify their results.
Main comments:
Recent papers, and especially Gryspeerdt et al. (2023, 10.5194/acp-23-4115-2023) and Arola et al. (2022, 10.1038/s41467-022-34948-5), have seriously challenged the usefulness of correlating aerosol and cloud parameters as done in the present study. Cloud variability and retrieval errors are such that correlations between aerosol optical depth and cloud properties (CNDC, CER, LWP) can in fact be spurious. That means that a large fraction of past literature on aerosol-cloud interactions (including the studies cited lines 86-107) needs to be look at again critically. Attempts to minimise retrievals uncertainties (lines 172-175) will not address parts of the issues. I think the present study remains interesting (especially the GDM analysis), but the authors need to acknowledge the possibility that the correlations they find do not say much about aerosol-cloud interactions.

The idea of using the GDM is interesting, but it is difficult to make physical sense of the results (Table 4 and 5). First, the q factors do not sum up to 1. What does that mean that AOD explains 87% of CER variability, while RH explains a further 36%? Then, the size of the q factors for AOD in Table 4, and to a lesser extent Table 5, stretches belief. If aerosols were that important in determining cloud properties, then estimating aerosol-cloud radiative forcing would have been very easy. Clearly, something goes wrong here. Is it perhaps that the four variables studied (AOD, RH, LTS, and PVV) are not independent? That the large q-values are simply a symptom of correlations caused by atmospheric circulation in the ECS and YRD? This is essentially what Figure 10 suggests. A strong discussion is needed to support the results.

The GDM assumes that the spatial distributions of independent and dependent variables “should have evident similarities” [line 211]. But at what scale is that assumption true for aerosol-cloud interactions? One could expect the assumption to break down when going down to the scale of a cloud field because clouds evolve after their aerosol-influenced formation phase. Is that a problem?

Other comments:
The name “ACI index” is vague. That quantity is really a sensitivity of cloud effective radius to changes in aerosol optical depth, in a similar way to beta_ln(N)-ln(tau) in Bellouin et al. (2020, doi: 10.1029/2019RG000660)

Lines 26 and 27: “significant” – is that in the statistical sense?

Line 47: “in practice” in the atmospheric sciences. Other disciplines use the term more properly.

Lines 44-54: Those generalities on aerosols are not necessary, so that section could be shortened. In fact, the introduction could start directly from line 54: “Aerosol particles are important for climate…”

Lines 55-78: Note that many papers since Chapter 7 of the IPCC AR5 (Boucher et al. 2013) use the concept of aerosol-radiation and aerosol-cloud interactions and their respective adjustments (e.g., Bellouin et al. 2020, Quaas et al. 2022 10.5194/acp-22-12221-2022). The terms direct/1^st indirect/2nd indirect remain in use in parts of the community, but it would be good to connect to the new terminology.

Lines 85-86: It should be said that using AOD as a proxy for aerosol concentrations when looking at aerosol-cloud interactions raises issues. See Section 6 of Bellouin et al. (2020).

Line 168: Andreae (2009) is often cited as justification for using AOD for looking at aerosol-cloud interactions, but ironically its Figure 1 shows that the correlation only exists across aerosol regimes. For a given regime (as done in the present study) there is essentially no correlation. I could not see why Kourtidis et al. (2015) justifies the use of AOD, but I may have missed it.

Lines 169-170: That assumes that cloud contamination has a lesser impact on smaller AODs. Is that true?

Line 207: “intermittently” Probably not the correct word. Interchangeably?

Line 231: Does q sum up to 1 for all factors considered? Is it also able to quantify an unexplained fraction that could suggest the need for more factors?

Figure 3: What does that Figure tell the reader? It’s impossible to say from its caption or from lines 241-245. The discussion needs to cover each of the panels in turn.

Lines 251-252: “for several different purposes”. Give examples based on the papers cited.

Line 265: “averaged over the years 2008-2022”. Does the study use 14-year averaged distributions, or a less dramatic averaging (e.g., multi-annual monthly means)? How can correlation of distributions averaged over such a long period still inform about the physical correlation between clouds and aerosols?

Lines 285-286: Or it could be due to meteorological factors.

Line 373: Missing word: “in the range of”
Citation: https://doi.org/10.5194/egusphere-2023-1615-RC3
- AC3: 'Response to Referee #2', Yuqin Liu, 07 Dec 2023
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1615/egusphere-2023-1615-AC3-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-1615-AC3

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

AR by Yuqin Liu on behalf of the Authors (07 Dec 2023) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (14 Dec 2023) by Stelios Kazadzis

RR by Anonymous Referee #1 (21 Dec 2023)

Suggestions for revision or reasons for rejection

I thank the authors for their excellent response to the majority of my points and the extra analysis they have done. My previous points have been largely addressed, although a few small points remain that I think would be good to clarify before publication.

Main points

Many thanks for this additional explanation of the q-factors. I have one further question about the interpretation. With the new text 'Explanatory power of combined effects is larger...' (L44), this seems to imply that the impact of aerosol and meteorological factors are combining to generate a larger effect. Could it not be some other factor that causes the meteorology and aerosol to co-vary along with the clouds, such that it appears there is a combined effect, when it is actually a confounder?

It is good to have a justification for using the CER instead of the Nd, but some of the points should be referenced or removed. The studies that used CER alone were not really looking at the Twomey effect in isolation, such that they were not really studying the RFaci either (McComiskey and Feingold, 2012). While Nd is affected by biases in the CER retrieval, these are different to the CER biases alone (and in some cases may offset each other; Painemal and Zuidema, JGR, 2011). For marine stratocumulus clouds, the Nd retrieval appears to be surprisingly accurate (Gryspeerdt et al, ACP, 2022). I would also note that while it is clear there is a relationship between CER and LWP, given that LWP is calculated from CER and cloud optical depth (as is Nd), neither of these is a ``retrieval error'' as such. All the properties could be retrieved perfectly and you could still find a relationship between CER and LWP (particularly for adiabatic clouds).

I think it would be fine just to state that you are stratifying by LWP and focussing on CER sensitivity (without a focus on the RFaci). I would still suggest you consider Nd for future studies. There are easily available Nd products that might be useful (such doi:10.5285/864a46cc65054008857ee5bb772a2a2b and https://catalogue.ceda.ac.uk/uuid/cf97ccc802d348ec8a3b6f2995dfbbff ).

Also, it seems like if the GDM doesn't detect significant relationships for regions smaller than 9x9. When compared to the 4x4 region recommended by Grandey and Stier (2010), might this suggest that the results in this work are due to a misleading spatial covariation? Naturally, the GDM depends on spatial variability, so cannot easily operate on small regions (as I understand it) and there are other benefits to using it. I am not suggesting that this invalidates your results or that you have to do a lot of extra work, but I think this is an important aspect that should be discussed in the paper so it is clear to further potential users of this method.

Minor points

L68 - RFari is defined before RF

L89 - The 'cloud lifetime effect' is less used today, as it is not clear a cloud lifetime is really involved (see the IPCC AR5 chapter on clouds and aerosols). I think you could just say 'are sometimes referred to', if you want to keep the terms in.

L254 - Christensen - this paper also doesn't seem to show up in the references

L274 - I don't think Liu, 2002b is the usual ERA5 reference

L323 - I am not clear what 'dot files' are. I assume it is some kind of 2D(?)array representation?

L420 - Aerosol is likely a secondary control on CF, rather than the primary cause. CER is similar, with cloud depth (and hence LWP) being a more important factor in determining CER, rather than aerosol.

L552 - While there is a strong correlation between AOD, CF and CTP, that is not good evidence of an aerosol effect (Quaas et al., ACP, 2010; Gryspeerdt et al., ACP, 2014c).

Results section - I would be a little careful in attributing causality to these results, as they show a correlation, but not evidence that the aerosol variation caused some change in cloud properties. This doesn't require much change, but I would watch out for cases talking about the 'effect' of AOD on cloud properties (e.g. L689) or 'influence' (e.g. L787), as it is not clear there is actually an impact of aerosol on cloud properties from these results.

Hide

RR by Anonymous Referee #3 (22 Jan 2024)

ED: Publish subject to minor revisions (review by editor) (22 Jan 2024) by Stelios Kazadzis

AR by Yuqin Liu on behalf of the Authors (30 Jan 2024) Author's response Author's tracked changes Manuscript

ED: Publish as is (16 Feb 2024) by Stelios Kazadzis

AR by Yuqin Liu on behalf of the Authors (23 Feb 2024)

Journal article(s) based on this preprint

18 Apr 2024

Opposite effects of aerosols and meteorological parameters on warm clouds in two contrasting regions over eastern China

Yuqin Liu, Tao Lin, Jiahua Zhang, Fu Wang, Yiyi Huang, Xian Wu, Hong Ye, Guoqin Zhang, Xin Cao, and Gerrit de Leeuw

Atmos. Chem. Phys., 24, 4651–4673, https://doi.org/10.5194/acp-24-4651-2024,https://doi.org/10.5194/acp-24-4651-2024, 2024

Short summary

Yuqin Liu, Tao Lin, Jiahua Zhang, Fu Wang, Yiyi Huang, Xian Wu, Hong Ye, Guoqin Zhang, Xin Cao, and Gerrit de Leeuw

Viewed

Total article views: 465 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
317	116	32	465	16	11

HTML: 317
PDF: 116
XML: 32
Total: 465
BibTeX: 16
EndNote: 11

Views and downloads (calculated since 15 Aug 2023)

Month	HTML	PDF	XML	Total
Aug 2023	120	38	5	163
Sep 2023	71	21	8	100
Oct 2023	22	8	3	33
Nov 2023	16	5	1	22
Dec 2023	37	17	8	62
Jan 2024	10	2	12
Feb 2024	13	12	0	25
Mar 2024	20	14	3	37
Apr 2024	8	1	2	11
May 2024	0
Jun 2024	0
Jul 2024	0
Aug 2024	0
Sep 2024	0

Cumulative views and downloads (calculated since 15 Aug 2023)

Month	HTML	PDF	XML	Total
Aug 2023	120	38	5	163
Sep 2023	71	21	8	100
Oct 2023	22	8	3	33
Nov 2023	16	5	1	22
Dec 2023	37	17	8	62
Jan 2024	10	2	12
Feb 2024	13	12	0	25
Mar 2024	20	14	3	37
Apr 2024	8	1	2	11
May 2024	0
Jun 2024	0
Jul 2024	0
Aug 2024	0
Sep 2024	0

Viewed (geographical distribution)

Total article views: 451 (including HTML, PDF, and XML) Thereof 451 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 18 Sep 2024

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (1791 KB)
Metadata XML

Short summary

Aerosols and clouds are very important for their effect on climate. Clouds are formed by condensation of water vapor on aerosol particles. We explore the interactions between aerosols, cloud and meteorological parameters in a highly polluted area over Eastern China and a relatively clean area over the East China Sea, using satellite data and different analysis methods. We describe the opposing effects of aerosol and meteorological parameters on cloud properties in the two areas.


Total:	0
HTML:	0
PDF:	0
XML:	0