Cool Skin Effect as Seen from a New Generation Geostationary Satellite Himawari-8
Round 1
Reviewer 1 Report
This paper employed satellite observations (Himawari-8) and in situ observations (iQuam) and explored the features of cook skin effect in the west pacific region. Those data were well quality-controlled and the data matchup criterion between satellite data were strictly set up. There are three main findings in my impression after reading this paper. The first one is that the mean intensity of cool skin as revealed by the geostationary satellite is slightly weaker than previous cool skin models. The second one is that a large portion (27%) of data with significant warm skin phenomena was detected, despite strict exclusion of daytime data. The last one is that the oceanic CO2 uptake computed with existing cool skin models was probably overestimated.
In general, the findings of this paper is significant and the manuscript is well organized and written. The analysis is reasonable and the discussion is adequate. I would like to suggest this paper be considered for possible publication in Remote Sensing after minor revision. Please find my comments below.
Main:
Q1: Why is the current study area chosen? Some necessary reasons are absent. Or the study area can be embedded in the title.
Q2: Section 3.4 seems to be out of coordinate with the contents in Section 3. It might be more suitable for removing it to Section 2. The impact of VZA on the SSTskin measurement belongs to the scope of quality control. Relevant VZA threshold can be formulated to help improve the quality of satellite cool skin measurement further.
Q3: The findings of warm skin signals and their relationship with air-sea temperature differences are quite interesting. Such signals can also be found in the pure in situ observations in reference 23 (see the grey dots below zero in Fig. 10). What I want to ask is that what implication such findings can have on improving the cool skin model in the next?
Specific:
L104: “Tskin.” -> remove the dot here.
Fig. 2: It is hard to understand this figure without reading the main text. Please give a detailed figure caption for it.
Table 2: Should be Table 1, because there is only one table in the whole text. And the “Not available” here should be 628, according to the Fig. 10 in literature 23.
L261: “two points” -> two aspects
Author Response
Please see the attachment. All revisions are marked with tracks in the word file.
Author Response File: 
Author Response.docx
Reviewer 2 Report
ERA5 includes warm-layer and cool-skin effects. Comparison of night-time difference between ERA5 skin and bulk temperatures using hourly ERA5 is needed to establish that the definition of night-time by the authors is not resulting in their warm skin results.
Only drifting buoy data has been used, where surface Argo was not. It is best to use another platform such as Argo to be certain about the deviations from past results.
Author Response
Please see the attachment. All revisions are marked with tracks in the word file.
Author Response File: Author Response.docx
Reviewer 3 Report
The authors examined the cool skin effect using one of the latest geostationary satellite data. The quality of the data is good, and the number of collected match-ups is large enough for reliable analyses. It is valuable to indicate that the cool skin effect is smaller than previous studies in high-wind conditions, although its reason is not discussed.
Overall, their argument is clear and concise, and the manuscript is well organized. There is no serious flaw, except for lack of Table 1. This manuscript is worth publishing with minor revisions.
1) L.73-74: "Secondly, geostationary satellites have ..."
Is it true in general? The orbit altitude of geostationary satellites is much higher than that of polar-orbiting satellites. Doesn't the latter have higher resolution than the former in general?
2) L.88-105:
Describe the algorithms deriving SST more. NLSST? MCSST? or another? How many channels are used to derive SST?
3) L.170:
Where is Table 1?
4) L.176-178: "Different data sources. .. suffer from different data sources."
This part is redundant and unnecessary.
5) L.178:
What is "sampling capability"?
6) L.183-184: "(3) Excessive data amount. .. statistical results."
I can't understand it and I don't agree to it. Increase of samples usually decreases error.
7) L.224-239, Figure 4
The difference between this result and previous studies is the core of this study. Discuss why this difference was caused. Did previous studies lack matchups in high-wind conditions?
8) L.251: "rapidly"
How did the authors estimate the response time?
Author Response
Please see the attachment. All revisions are marked with tracks in the word file.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Comments have been addressed
