The First Results of Cloud Retrieval from Geostationary Environmental Monitoring Spectrometer

Kim, Bo-Ram; Kim, Gyuyeon; Cho, Minjeong; Choi, Yong-Sang; Kim, Jhoon

doi:https://doi.org/10.5194/amt-2023-91

Preprints

https://doi.org/10.5194/amt-2023-91

Preprints

10 May 2023

| 10 May 2023

Status: a revised version of this preprint was accepted for the journal AMT and is expected to appear here in due course.

The First Results of Cloud Retrieval from Geostationary Environmental Monitoring Spectrometer

Bo-Ram Kim, Gyuyeon Kim, Minjeong Cho, Yong-Sang Choi, and Jhoon Kim

Abstract. This research introduces the cloud retrieval algorithm for Geostationary Environmental Monitoring Spectrometer (GEMS), the first geostationary orbit satellite, and shows the validation of its cloud products through comparison with other satellites: OMI, TROPOMI, AMI, and CALIOP. The purpose of GEMS cloud products is to correct the impact of clouds on atmospheric components retrieval, which use the O2-O2 absorption band to retrieve the effective cloud fraction (ECF) and cloud centroid pressure (CCP). The GEMS cloud retrieval algorithm showed similar cloud retrieval performance to OMI. We analyzed the cloud retrieval characteristics for cases of air pollution, typhoons, and fog in the East Asia region to evaluate whether GEMS cloud products can represent various cloud features. The present cloud validation results would initiate to improve the GEMS cloud retrieval algorithm in the future.

Received: 25 Apr 2023 – Discussion started: 10 May 2023

Download & links

Bo-Ram Kim et al.

Status: closed

RC1:
'Comment on amt-2023-91', Anonymous Referee #1, 14 Jun 2023

This manuscript explained and summarized the developed algorithm of the cloud for GEMS. It is a noble paper for the report on the GEMS operational algorithm. Because of this reason, it is suitable to publish this special issue.
However, several improvements are required for the confidence of retrieval algorithm results during the revision process.
1) To improve the structure of this paper, it is necessary to revise the order of sections.
Especially, Sections 2.2 and 3.1 are too similar. Please merge the two sections into one section.
2) To explain the GEMS radiance characteristics, sampling and resolution information are wrong. The sampling is 0.2 nm and the resolution is 0.6 nm. Please correct it.
3) Section 2.2 has to be moved before Section 2.1. Algorithm outputs are more appropriate after the algorithm description.
4) L139: For the radiance simulation, this study used 460-485 nm. However, the O2-O2 band at 477 nm significantly affect the spectral range longer than 485 nm. Why does this study use the spectral range with not wide enough?
5) L144: This spectral range is significantly affected the NO2 absorption. Why does this study not consider the NO2 absorption?
6) L156: From Equation (3), this equation is not a full DOAS method. It is linearized absorption signal separation. The full DOAS method is additionally considered the non-linear correction. How about correcting the non-linear effect for the estimation of SCD?
7) L161: For readability, please add the flowchart of the whole cloud retrieval algorithm.
8) L162-L166: It is too simple to explain. Please add the details of the explanation what is the main difference between the two different platforms.
9) Section 3.1 (before revision): This section needs to separate the sections according to the platforms.
10) L248: This manuscript is only focused on the Korean Peninsula, not the GEMS domain. Do you have some reasons? Please clarify this issue.
11) Section 4: This manuscript is focused on the cloud algorithm for GEMS. However, Section 4 shows the results using OMI and TROPOMI. Please clarify the purpose of these results.
12) Figure 3: This figure is one of the key results in this manuscript. However, as shown in this figure, an arbitral stripe pattern exists. I think that this is a problem during the inversion process for the best solution estimation from LUT. This problem is a critical issue for the operational algorithm. Please check and find the reason and need to fix this critical issue.
13) Section 5: Although the manuscript shows several cases, the long-term comparison result is also required to see the stability of the algorithm's accuracy. In addition, this study only showed the 0430 UTC result. It is only allowed to the intercomparison. To see the diurnal variability of cloud retrieval results, this study also has to show the continuous diurnal results. In addition, some performance results (statistical validation results) are essential.
Minor Comment
1) L160: angle geometry --> observation geometry
2) L168: Please add the reference for the VLIDORT NGST version.
3) L199: Please add the reference, and recheck the spatial resolution of TROPOMI in L204.
4) L241: To the clarify the difference, please list-up the difference of definition for cloud parameters.
5) L285-286: This sentence is not clear. Please rephrase it.

Citation: https://doi.org/10.5194/amt-2023-91-RC1
- AC1: 'Reply on RC1', Bo-Ram Kim, 03 Aug 2023
  
  Please find the attached supplement.
  
  Citation: https://doi.org/10.5194/amt-2023-91-AC1
RC2:
'Comment on amt-2023-91', Anonymous Referee #2, 28 Jun 2023

General comments:
This manuscript presents a generally well written study on the algorithm retrieval results of GEMS cloud products. The author presented comprehensive analysis including comparison with the different satellite products along with the algorithm results. I suggest the publication of the paper after minor revisions.
Specific comments:
Line 11: ‘the first geostationary orbit satellite’ -> I recommend that to be more specific as the GEMS is not the first geostationary orbit satellite instrument.
Line 26: missing periods.
Line 23,29: It may not be a serious problem, but I suggest you distinguish the word between “satellite” and “instruments”.
Line 30: gases(GHG) -> gases (GHG)
Line 31: Please check typo “using use spectrometers GHGs”.
Line 41: ‘characteristics vary greatly depending on the spectral band.’ -> Do you have any reference or evidence for this sentence? Or did you want to say retrieval results greatly depending on the instrument characteristics?
Line 58: I thought the spectral resolution of GEMS is about 0.6nm, while sampling is 0.2nm. Could you check again?
Line 60: ‘keeping the Sun-Earth-satellite angle constant’ -> do you mean constant VZA?
Line 64-65: Could you provide a reference for this? Or I think you can probably explain it with low SNR, etc.
Line 84: ‘with a resolution of 0.2 nm’ -> Could you check this again? I thought the spectral resolution of GEMS is about 0.6nm.
Line 92: Could you provide some references?
Line116: Why? Are there no CRF products from other satellites?
Line 144,147,152: Just curious. Why is there no consideration of NO₂ in this equation? I think the impact may be significant, especially over East Asia. What do you mean that the absorption by nitrogen dioxide is linear?
Line 168: Do you have any reference paper for VLIDORT NGST?
Line 173: ‘which has the most similar algorithm design’ -> Do you mean as a prototype?
Line 174: ‘operates simultaneously with GEMS’ -> I suggest ‘in operational since 2018’ rather than operates simultaneously.
Line 175: ‘the same orbit’ -> Does it mean the same geostationary orbit?
Line 182: 0.6 -> 0.6nm
Line 243: Just curious. Are there any standards to select the cases?
Line 258: ‘the nearest neighbor method was based on’ -> ‘the nearest neighbor method was used based on’?
Figure 1: Reason for the stripe pattern?
Line 280: It would be better if you could add the reason briefly.
Figure 2: Caption ‘March 5^th’ -> ‘March 25^th’.
Figure 1-4: I can see that you are using the term “GEMS ECF” or “GEMS CCP”. This is not GEMS data, but GEMS algorithm applied results. I suggest you distinguish between real GEMS products and GEMS algorithm applied products in the Figure, but I’ll let the author decide it.
Figure 3: Reason for the stripe pattern?
Line 329: Just curious. Which channel does the AMI use for the cloud retrieval?
Line 350: Why does the GEMS tends to estimate lower cloud heights than TROPOMI cloud pressure?
Line 367: Again, why does the GEMS tends to estimate lower cloud heights than TROPOMI cloud pressure?
Line 398: Maybe you can add a brief reason why GEMS cloud height is the lowest.

Citation: https://doi.org/10.5194/amt-2023-91-RC2
- AC2: 'Reply on RC2', Bo-Ram Kim, 03 Aug 2023
  
  Please find the attached supplement.
  
  Citation: https://doi.org/10.5194/amt-2023-91-AC2

Status: closed

RC1:
'Comment on amt-2023-91', Anonymous Referee #1, 14 Jun 2023

This manuscript explained and summarized the developed algorithm of the cloud for GEMS. It is a noble paper for the report on the GEMS operational algorithm. Because of this reason, it is suitable to publish this special issue.
However, several improvements are required for the confidence of retrieval algorithm results during the revision process.
1) To improve the structure of this paper, it is necessary to revise the order of sections.
Especially, Sections 2.2 and 3.1 are too similar. Please merge the two sections into one section.
2) To explain the GEMS radiance characteristics, sampling and resolution information are wrong. The sampling is 0.2 nm and the resolution is 0.6 nm. Please correct it.
3) Section 2.2 has to be moved before Section 2.1. Algorithm outputs are more appropriate after the algorithm description.
4) L139: For the radiance simulation, this study used 460-485 nm. However, the O2-O2 band at 477 nm significantly affect the spectral range longer than 485 nm. Why does this study use the spectral range with not wide enough?
5) L144: This spectral range is significantly affected the NO2 absorption. Why does this study not consider the NO2 absorption?
6) L156: From Equation (3), this equation is not a full DOAS method. It is linearized absorption signal separation. The full DOAS method is additionally considered the non-linear correction. How about correcting the non-linear effect for the estimation of SCD?
7) L161: For readability, please add the flowchart of the whole cloud retrieval algorithm.
8) L162-L166: It is too simple to explain. Please add the details of the explanation what is the main difference between the two different platforms.
9) Section 3.1 (before revision): This section needs to separate the sections according to the platforms.
10) L248: This manuscript is only focused on the Korean Peninsula, not the GEMS domain. Do you have some reasons? Please clarify this issue.
11) Section 4: This manuscript is focused on the cloud algorithm for GEMS. However, Section 4 shows the results using OMI and TROPOMI. Please clarify the purpose of these results.
12) Figure 3: This figure is one of the key results in this manuscript. However, as shown in this figure, an arbitral stripe pattern exists. I think that this is a problem during the inversion process for the best solution estimation from LUT. This problem is a critical issue for the operational algorithm. Please check and find the reason and need to fix this critical issue.
13) Section 5: Although the manuscript shows several cases, the long-term comparison result is also required to see the stability of the algorithm's accuracy. In addition, this study only showed the 0430 UTC result. It is only allowed to the intercomparison. To see the diurnal variability of cloud retrieval results, this study also has to show the continuous diurnal results. In addition, some performance results (statistical validation results) are essential.
Minor Comment
1) L160: angle geometry --> observation geometry
2) L168: Please add the reference for the VLIDORT NGST version.
3) L199: Please add the reference, and recheck the spatial resolution of TROPOMI in L204.
4) L241: To the clarify the difference, please list-up the difference of definition for cloud parameters.
5) L285-286: This sentence is not clear. Please rephrase it.

Citation: https://doi.org/10.5194/amt-2023-91-RC1
- AC1: 'Reply on RC1', Bo-Ram Kim, 03 Aug 2023
  
  Please find the attached supplement.
  
  Citation: https://doi.org/10.5194/amt-2023-91-AC1
RC2:
'Comment on amt-2023-91', Anonymous Referee #2, 28 Jun 2023

General comments:
This manuscript presents a generally well written study on the algorithm retrieval results of GEMS cloud products. The author presented comprehensive analysis including comparison with the different satellite products along with the algorithm results. I suggest the publication of the paper after minor revisions.
Specific comments:
Line 11: ‘the first geostationary orbit satellite’ -> I recommend that to be more specific as the GEMS is not the first geostationary orbit satellite instrument.
Line 26: missing periods.
Line 23,29: It may not be a serious problem, but I suggest you distinguish the word between “satellite” and “instruments”.
Line 30: gases(GHG) -> gases (GHG)
Line 31: Please check typo “using use spectrometers GHGs”.
Line 41: ‘characteristics vary greatly depending on the spectral band.’ -> Do you have any reference or evidence for this sentence? Or did you want to say retrieval results greatly depending on the instrument characteristics?
Line 58: I thought the spectral resolution of GEMS is about 0.6nm, while sampling is 0.2nm. Could you check again?
Line 60: ‘keeping the Sun-Earth-satellite angle constant’ -> do you mean constant VZA?
Line 64-65: Could you provide a reference for this? Or I think you can probably explain it with low SNR, etc.
Line 84: ‘with a resolution of 0.2 nm’ -> Could you check this again? I thought the spectral resolution of GEMS is about 0.6nm.
Line 92: Could you provide some references?
Line116: Why? Are there no CRF products from other satellites?
Line 144,147,152: Just curious. Why is there no consideration of NO₂ in this equation? I think the impact may be significant, especially over East Asia. What do you mean that the absorption by nitrogen dioxide is linear?
Line 168: Do you have any reference paper for VLIDORT NGST?
Line 173: ‘which has the most similar algorithm design’ -> Do you mean as a prototype?
Line 174: ‘operates simultaneously with GEMS’ -> I suggest ‘in operational since 2018’ rather than operates simultaneously.
Line 175: ‘the same orbit’ -> Does it mean the same geostationary orbit?
Line 182: 0.6 -> 0.6nm
Line 243: Just curious. Are there any standards to select the cases?
Line 258: ‘the nearest neighbor method was based on’ -> ‘the nearest neighbor method was used based on’?
Figure 1: Reason for the stripe pattern?
Line 280: It would be better if you could add the reason briefly.
Figure 2: Caption ‘March 5^th’ -> ‘March 25^th’.
Figure 1-4: I can see that you are using the term “GEMS ECF” or “GEMS CCP”. This is not GEMS data, but GEMS algorithm applied results. I suggest you distinguish between real GEMS products and GEMS algorithm applied products in the Figure, but I’ll let the author decide it.
Figure 3: Reason for the stripe pattern?
Line 329: Just curious. Which channel does the AMI use for the cloud retrieval?
Line 350: Why does the GEMS tends to estimate lower cloud heights than TROPOMI cloud pressure?
Line 367: Again, why does the GEMS tends to estimate lower cloud heights than TROPOMI cloud pressure?
Line 398: Maybe you can add a brief reason why GEMS cloud height is the lowest.

Citation: https://doi.org/10.5194/amt-2023-91-RC2
- AC2: 'Reply on RC2', Bo-Ram Kim, 03 Aug 2023
  
  Please find the attached supplement.
  
  Citation: https://doi.org/10.5194/amt-2023-91-AC2

Bo-Ram Kim et al.

Viewed

Total article views: 542 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
393	129	20	542	12	13

HTML: 393
PDF: 129
XML: 20
Total: 542
BibTeX: 12
EndNote: 13

Views and downloads (calculated since 10 May 2023)

Month	HTML	PDF	XML	Total
May 2023	155	45	4	204
Jun 2023	78	24	4	106
Jul 2023	33	13	2	48
Aug 2023	49	17	5	71
Sep 2023	33	13	5	51
Oct 2023	28	10	0	38
Nov 2023	17	7	0	24
Dec 2023	0

Cumulative views and downloads (calculated since 10 May 2023)

Month	HTML	PDF	XML	Total
May 2023	155	45	4	204
Jun 2023	78	24	4	106
Jul 2023	33	13	2	48
Aug 2023	49	17	5	71
Sep 2023	33	13	5	51
Oct 2023	28	10	0	38
Nov 2023	17	7	0	24
Dec 2023	0

Viewed (geographical distribution)

Total article views: 538 (including HTML, PDF, and XML) Thereof 538 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 02 Dec 2023

Short summary

This study introduces the GEMS cloud algorithm and validates its results using data from GEMS and other environmental satellites. The GEMS algorithm is able to detect cloud heights the lowest among the four satellites, and its effective cloud fraction and cloud centroid pressure are well reflected in the retrieval results. The study highlights the algorithm's usefulness in correcting errors in trace gases caused by cloud in the East Asian region.


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