the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Aerosol radiative effects with dual view AOD retrievals
Stefan Kinne
Peter North
Kevin Pearson
Thomas Popp
Abstract. Seasonal maps of dual view retrieved mid-visible AOD and AODf for four selected years (1998, 2008, 2019, 2020) are introduced and assessed in comparisons to MODIS retrievals and general data of an aerosol climatology. Due to different sensor capabilities (ATSR-2, AATSR and SLSTR) there are still unresolved inconsistencies so that decadal regional trends are not as detectable as with MODIS retrievals. SLSTR retrieval, however, agree with MODIS retrievals that 2020 Covid impacts on AOD values (via comparisons to the pre-COVID 2019 reference) are at best minor and secondary to natural anomalies by wildfires and dust. In radiative transfer applications the dual view AOD data for the four years are processed in the MAC climatology environment to determine aerosol associated radiative effects for total aerosol and for anthropogenic aerosol. Even though the calculated radiative effects are affected by retrieval AOD retrieval tendencies, climate relevant TOA net-flux changes are consistent to result with AOD data from other satellite sensors and a general climatology: −0.9 W/m2 for total aerosol with a significant greenhouse effect and −0.8 and −0.2 W/m2 for anthropogenic aerosol with and without indirect effects, respectively. Aside from global averages, seasonal maps highlight the diversity of regional and seasonal radiative effects.
Stefan Kinne et al.
Status: final response (author comments only)
-
RC1: 'Comment on acp-2021-954', Anonymous Referee #1, 14 Dec 2021
Review
Aerosol radiative effects with dual view AOD retrievals
by Stefan Kinne, Peter North, Kevin Pearson, Thomas Popp
The paper is devoted to the assessment of the quality of aerosol optical depth retrievals from ATSR-2, AATSR and SLSTR satellite instruments and evaluation of the radiative effects obtained using these data. The comparison with MODIS AOD retrievals have revealed significant biases and inconsistency so the authors claimed that decadal regional trends were not possible to retrieve from these data. However, the authors were able to solve two tasks: firstly, according to these data they showed the minor effect of lockdown due to COVID-19 and, secondly, they obtained reasonable climate relevant TOA net-flux changes, which were in agreement with other available satellite datasets and MAC climatology. I guess that the publication of the analysis concerning the quality of these satellite data is important, and the manuscript can be published after editing and some structural changes.
General comments:
- Since the paper in the scientific journal should be clear not only for specialists in satellite aerosol retrievals but for the broad scientific community I would recommend to extend Introduction (adding there the information concerning the current state of the quality of aerosol satellite retrievals, the assessment of radiative effect of aerosols, etc) and the potential advantages of dual view AOD retrievals.
- The Section “ Dual view (DV) radiometer data” is of most concern. I would recommend to include there the description of the method, the characteristics of the satellite instruments. It would be useful to add the Table with different characteristics of the instruments. It is not enough only to mention that the statistics is poor for the ATSR (row 67), but it is necessary to show the statistics for different sensors.
- It is not clear why the authors used MODIS data for the trend analysis if the paper is devoted to the analysis of other sensors. This is possible to do only if these data are somewhat useful for the aims of DV method retrievals or RT assessments. I guess that the analysis of trends according to the MODIS data here is inconsistent with the title of the paper. Or the title of the paper should be changed.
- The study of COVID impact is now a very popular topic. However, to my understanding, it should be analyzed more carefully to evaluate the local effects of aerosol over urban areas. I understand that it is easier to say that it is no effect of COVID-19. I agree that the effect of lockdown on AOD is small, however, it might be seen, if the accurate analysis is done. However, the quality of these satellite retrievals is not good, and the effects are comparable with the uncertainty of satellite retrievals. I would recommend to discuss the uncertainty of the retrieval method and after that make a statement that the evaluation of the COVID-19 lockdown AOD changes lies within this uncertainty.
- I would recommend to re-arrange the material into the standard Sections: Introduction, Methods, AOD quality assessment, Radiative effects and Discussion with Conclusions. Small subsections can be incorporated in this structure.
Specific comments:
- The quality (resolution) of Figures is not very good. It is not clear what means “sea” in the left up corner of them. Does this mean the specific retrieval over sea or seasonal? Please, show the meaning of all abbreviations.
- Section 6 has a poor title. Please, clarify.
- In Section 7 the comparisons of the results with other studies should be demonstrated.
- Discussion is too short. I would recommend to combine this part with the conclusions.
Technical corrections:
- Row 88. SLSTR: forward, ATSR-2 and AATSR: rearward – should be clarified. I understand this, but I am not sure that any reader would be happy with such kind of explanation.
- Row 90. If it is any publication concerning these biases? The reference would be useful here.
- Row 96. 2008, 2019 and 2020 – they are not the same. Please, correct the text.
- Row 168. ref? – Reference should be given.
- Row 172. What is the uncertainty of the applied RT method?
- Row 177. What about SSA and ASYM factors? Were the changes made only in AOD values?
- Row 179. Please, add the reference and short description of the applied indirect method.
- Row 179. It is unclear. what the letter “a” means here.
- Row 188. It is not clear, how the anthropogenic aerosol presence is evaluated from satellite data.
- Row 194. Please, give the reference here.
- Row 195-198 . It is not obvious from the text that satellite data can be used to solve these tasks. Should be clarified.
- Row 221. I do not understand from the text, how the satellite data are used for inferring anthropogenic pattern ( Table 2 , right side).
- Row 227-228. There is relatively large difference in radiative effects between MAC climatology and satellite retrievals. This should be underlined. Otherwise it is strange, since there is large difference in AOD.
- Row 229-230. I do not understand the last sentence. What is SR – single retrieval? In this case the short description of this method should be given.
- Row 231. I would recommend to add the quantitative estimates in Section 7. Otherwise, the statements are too lightweight. It would be useful to show the limits in the range of different parameters (1- SSA, size distribution, etc), when their uncertainties do not play vital role.
- Row 261 and 266. What does word “presence” mean here?
- Row 264. What is “dir: at the upper part of Figure 6? Does this mean direct effect? All abbreviations should be clarified.
- Row 277. It is not clear again, what does mean “aerosol presence”? May be better to use “direct effect” or it is not the same thing?
- Row 300. I do not understand “…also informs on…”? Informs about, may be? Or “separated on”?
- Row 321. From Fig.9 I eventually understand that the anthropogenic AOD is evaluated using AODf retrievals from these satellite instruments. Please, clarify this in a better way in the text and show the equation, how the anthropogenic AOD is estimated.
- Row 335. I do not clearly understand the following sentence: “Compared to the direct impact ( at TOA?? should be added!), net flux reductions at the surface are mainly increased over (dark) ocean regions affected by pollution outflow." Does this mean that over bright surface net flux does not have reduction due to pollution? Or there is another meaning?
- Row 347. There is a difference in numbers here and in Table 2. They should be the same or not? It should be clarified.
- Row 377. “ay” – misprint.
Citation: https://doi.org/10.5194/acp-2021-954-RC1 -
AC1: 'Reply on RC1', Stefan Kinne, 15 Dec 2021
The comment was uploaded in the form of a supplement: https://acp.copernicus.org/preprints/acp-2021-954/acp-2021-954-AC1-supplement.pdf
-
RC2: 'Review of Kinne et al.', Anonymous Referee #2, 05 Jan 2022
The study is essentially four papers in one, often loosely linked to the stated aim of being a user case study of an Aerosol_cci product. Unfortunately none of the four items are analysed with enough depth and novelty to recommend publication.
The first item, in section 2, focuses on the performance of the AOD and AODf dual-view retrievals based on “SR heritage” sensors. That analysis is clearly premature. As the authors write on line 89, there is a clear need for “DV sensor retrieval biases [to be] better understood and corrected”. The SR retrieval teams should first publish a paper documenting time series of AOD and AODf over 1998-2020 with the different SR sensors, identifying possible global biases and coming up with ways to splice the different datasets consistently.
For the second item, in section 3, the authors give up on dual-view retrievals and use MODIS to very briefly look at decadal trends in aerosol optical depth. That analysis is out of the stated scope of the paper, and does not really add value to previous work, for example https://doi.org/10.5194/acp-20-139-2020 .
The third item, in section 4, looks at possible changes in AOD from decreases in activity due to the Covid pandemic. That is an interesting question, but which requires a more careful analysis than done here. First, many aerosol sources remained active during lockdowns (https://doi.org/10.1029/2020GL088913) so the impact of AOD is smaller than could be anticipated. In addition, one cannot simply compare 2020 to 2019, because that ignores interannual variability (https://doi.org/10.1029/2020GL091805, https://doi.org/10.1029/2021GL093841, https://doi.org/10.1029/2020gl091699), the complexity of aerosol chemistry (https://doi.org/10.1029/2020GL088533, https://doi.org/10.1093/nsr/nwaa137), and a potential masking of the Covid signal by anomalous meteorology in 2020 (https://doi.org/10.1029/2020JD034090, https://doi.org/10.1126/science.abb7431).
The fourth item, in sections 6-8, looks at the impact of using DV-retrieved AOD fields on aerosol radiative effects in the MAC climatology framework. Differences can be explained by differences in AOD, and the authors find that assumptions on the single-scattering albedo have a large impact. Here, the use of DV retrievals feels unnecessary to draw these conclusions. The same analysis could have been done within MAC itself, and indeed most conclusions can already be drawn from the analysis by Kinne et al. 2019 (https://doi.org/10.5194/acp-19-10919-2019).
Other comments:
Section 2: The presentation of the three SR sensors in that section is minimal, yet the importance of their different capabilities between them and with MODIS is invoked very early in the analysis. What are the different swaths? What are the different fractional coverages?
Lines 16-17: “aerosol with a significant greenhouse effect” Need to clarify that this expression refers to longwave radiative effects.
Lines 31-32: “not associated with any specific year”. That cannot be correct, since the reader is told a few lines later that MAC is suitable for looking at decadal trends.
Lines 53-54: Does “size” refer to diameter or radius? Is AODf defined in the same way in the DV dataset, in MODIS, and in MAC?
Lines 87-88: Why would the type of sideway viewing influence seasonal statistics?
Technical comments:
- Many acronyms are never spelled out.
- There is no section 5 but there are two sections 7.
Citation: https://doi.org/10.5194/acp-2021-954-RC2 -
AC2: 'Reply on RC2', Stefan Kinne, 05 Jan 2022
The comment was uploaded in the form of a supplement: https://acp.copernicus.org/preprints/acp-2021-954/acp-2021-954-AC2-supplement.pdf
Stefan Kinne et al.
Stefan Kinne et al.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 498 | 193 | 33 | 724 | 8 | 51 |
- HTML: 498
- PDF: 193
- XML: 33
- Total: 724
- BibTeX: 8
- EndNote: 51
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1