Intercomparison between the aerosol optical properties retrieved by different inversion methods from SKYNET sky radiometer observations over Qionghai and Yucheng in China

Jiang, Zhe; Duan, Minzheng; Che, Huizheng; Zhang, Wenxing; Nakajima, Teruyuki; Hashimoto, Makiko; Chen, Bin; Yamazaki, Akihiro

doi:https://doi.org/10.5194/amt-13-1195-2020

Articles | Volume 13, issue 3

https://doi.org/10.5194/amt-13-1195-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Special issue:

SKYNET – the international network for aerosol, clouds,...

https://doi.org/10.5194/amt-13-1195-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 13, issue 3

Research article

|

11 Mar 2020

Research article |

| 11 Mar 2020

Intercomparison between the aerosol optical properties retrieved by different inversion methods from SKYNET sky radiometer observations over Qionghai and Yucheng in China

Zhe Jiang, Minzheng Duan, Huizheng Che, Wenxing Zhang, Teruyuki Nakajima, Makiko Hashimoto, Bin Chen, and Akihiro Yamazaki

Download

Final revised paper (published on 11 Mar 2020)
Supplement to the final revised paper
Preprint (discussion started on 06 Mar 2019)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Referee Comment', Anonymous Referee #2, 01 Apr 2019
- AC1: 'Response to RC1', Huizheng Che, 07 Sep 2019
RC2: 'Review of article', Anonymous Referee #1, 02 Apr 2019
- AC2: 'Response to RC2', Huizheng Che, 07 Sep 2019

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Huizheng Che on behalf of the Authors (13 Sep 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (26 Sep 2019) by Stelios Kazadzis

RR by Anonymous Referee #2 (05 Oct 2019)

Suggestions for revision or reasons for rejection

Authors have a made a huge effort by rerunning the algorithms in order to improve the study, also a lot of details about the two measuring sites and their atmospheric conditions has been added. However, my major concerns have not been answered and although a lot of information has been added, I can quote from my earlier review :

“However, the manuscript lacks of explanations on the causes of differences between the v4.2 and v5.0 retrievals and sufficient evidence on the actual seasonal variability of aerosols in the two regions. Algorithms of both versions are clearly described, but it is crucial to pinpoint and discuss the way the differences between the versions affect the retrievals. Since the two algorithms are not treated as a “black box”, it should be more clear which physical processes affects the retrievals and which atmospheric conditions could lead to highest uncertainty. At least some discussion on the uncertainty of each variable in each approach should be provided. Also, the part about the seasonal variability of aerosols properties, results are presented but not
investigated and discussed in the level expected for a scientific study.”

I think nothing has been done in this direction. Also, information added are not used in order to interpret the results and explain the findings of the present study. Also, some discussion is needed to conclude the validity or higher quality of v5.0 retrievals before using them for aerosol characterisation in section 3.2

I suggest a major revision of the manuscript before considering for publication in AMT.

P3l25 Is the same calibration used also for sky radiance measurements?

P6 l23-25 Other differences should at least be mentioned, before referring to Hashimoto et al work. At the end the present manuscript is about these differences.

Section 3.
AOD is derived from Direct Irradiance measurements. In earlier text only the algorithms for inversion products is described. Are there differences in direct sun algorithms between 5.0 and 4.2? Since AOD is presented these details should be explained. Also, even this slight differences of AOD could propagate considerable uncertainties, thus it should be more clear if and how AOD is used in inversion calculations in both versions.

Section 3.1.1

As it is stated now, it seems that the differences are caused by just the different approaches in the versions and the functions that the data are fitted to. Is that the conclusion? Are these differences connected to atmospheric conditions? Additionally, these approach of plotting just mean monthly values hides the real picture. At least some basic statistics (average and standard deviation of the differences at each bin) should be presented and discussed. Currently there is no information on the scattering of differences and if these differences are permanent biases.

Section 3.1.2
At this section there is no discussion on the causes of the differences. g. Are they explained strictly algorithmically or is there some natural process driving them? In most cases v5.0 retrieves lower SSA values suggesting the presence of more absorbing aerosols. Is there any evidence on that?

P8l12-16 Why at Qionghai are only relative differences presented, while for Yucheng both relative and absolute are mentioned? Absolute differences are more important generally and it should be mentioned for both datasets. Also, as mentioned above, at least a standard deviation of these differences should be presented. From the scatter plots it is clear that the deviations are very diverse in each dataset.

Figure 4. At Qionghai at 1020nm, it seems there two groups of measurements. One with values close to 1 and one with values lower than 0.8. Is there a physical explanation for this? It clearly needs more investigation this behavior. The same behavior at 1020 nm is also presented at figure 5 for refractive index.

Section 3.1.3
Same as above. At least some discussion on the scattering of the data, since real part appears to have almost random differences. . Are the differences explained strictly algorithmically or is there some natural process driving them?

Section 3.1.4
This sensitive test Is an appropriate way to understand the algorithmical differences. But still no conclusion is drawn from this test and nothing is discussed in respect to the findings of previous sections.

Figure 8. This approach also helps to increase the understanding of the algorithm. I suggest to plot real differences instead of absolute, because the sign is important to understand weather the version over or underestimates compared to the previous one. Also the lack of discussion on the uncertainty of the retrievals in the manuscript, makes it harder to interpret which range of difference is in the expected uncertainty.

P13 l1-2. Also, why selecting v5.0 for studying the seasonal variability should be explained here.

Section 3.2.1

I honestly have a difficulty understanding why these section for PM2.5 have been added Surely it makes more clear the local emissions types, but it seems unlinked with the rest of the study. Findings mentioned here are nowhere used to explain anything about SKYNET retrievals and their behavior. Also, it is an unexplained decision to study PM 2.5 while from all SKYNET retrievals there is a picture of constant dominance of larger particles in both regions. Unless you could integrate the findings to the discussion in the rest of the study, linked differences found between the two sites and preferably even connect the deviations between the two versions with the types of aerosols, I suggest removing this section.

Section 3.2.2
Which months are considered in each season should be defined. Also, the discussion about the humidity is not clear. It seems that aerosol loads are generally in the same order throughout the year and humidity causes the variations of AOD. This needs more evidence to support it and a lot of discussion and data are needed to prove it. If this is not the case, please restate to make clear the finding of this analysis.

Section 3.2.3
The discussion in this paragraph is not consistent with the next section. Since lower SSA values in winter are explained by the presence of carbonaceous particles, why winter SDF are dominated by coarse mode in both regions?

Referee Report: PDF

Hide

RR by Anonymous Referee #1 (10 Oct 2019)

ED: Reconsider after major revisions (17 Oct 2019) by Stelios Kazadzis

AR by Huizheng Che on behalf of the Authors (10 Jan 2020) Author's response Manuscript

ED: Publish subject to technical corrections (30 Jan 2020) by Stelios Kazadzis

AR by Huizheng Che on behalf of the Authors (06 Feb 2020) Manuscript

Short summary

This study analyzed the aerosol optical properties derived by SKYRAD.pack versions 5.0 and 4.2 using the radiometer measurements over Qionghai and Yucheng in China, which are two new sites of SKYNET. The seasonal variability of the aerosol properties over the two sites were investigated based on SKYRAD.pack V5.0. The validation results provide valuable references for continued improvement of the retrieval algorithms of SKYNET and other aerosol observational networks.