Estimating Early Summer Snow Depth on Sea Ice Using a Radiative Transfer Model and Optical Satellite Data

Round 1

Reviewer 1 Report (New Reviewer)

General comments:

Snow exhibits a higher albedo and significantly lower thermal conductivity, insulating the sea ice beneath it, reducing heat flux heat transfer between the ice-ocean and atmosphere and lowering both summer sea ice melt and winter sea ice growth. Therefore, correct understanding of snow depth is key to the understanding of regional and global processes and their changes. However, due to the potential complexity of surface melting in early summer, the method of calculating snow depth has a large error during the early summer snowmelt season. This paper proposes a novel method to estimate snow depth using a radiative transfer model and optical satellite data. This is really a meaningful study. However, several issues still need to be resolved before a publication to Remote Sensing.

 

Major Comments:

 

1. L10-12:  “However, Methods to derive snow depth only be applied through the annual winter and early spring period due to the potential complexity of surface melting during early summer.” The expression is not rigorous, such as the day-by-day passive microwave snow deep product. It is proposed to be modified as “However, Methods to derive snow depth has a large error during early summer due to the potential complexity of surface melting.”

 

2. L86-90:  Here should introduce the main problems to be solved and the highlights in the paper, not the structure of the paper.

 

3. L87:  “In Section 3, we provide an overview of the data sets used and the Data usage protocols.” A summary of the data sources and data processing workflow is provided in Section 3, with the “data usage protocol” is not the appropriately outlined for this section

 

4. L89：“intoduce ” should be “introduce”.

 

5. L194: “They span extensive regions including the central Arctic, Beaufort Seas, Chukchi Seas and the Greenland Sea.” It is necessary to explain why these areas were selected and whether they are representative

 

6. L205: “cloud less satellite images around the buoy position were selected.” Please describe how you select cloud-free images and what criteria you use for the selection.

 

7. There should be introduce about the published snow depth data for Arctic early summer. In addition to the comparison with the in-situ observed snow depth data, it is also necessary to contrast the results with existing data sets, i.e., passive microwave snow deep product, and highlighting any improvements made by the algorithm presented in this paper.

 

8. Why this method only be used on early melting seasons? How does it perform in other seasons? Can this method extended to all melting seasons?

 

9. Section 4, the observation area of the IMBs data used for verification needs to be introduced and discussed. Can it represent the situation in other areas?

 

10. L371: “a mean difference between satellite retrieval and buoy measurements of 4.1 cm was found, with 93% of the differences being smaller than 5 cm”  the total depth is very important when you talk about the different of 4.1m, you give the total depth in Figure7, but you should clearly statement the percentage in the discussion section as well.

 

11. L382: “Besides the improved performance, there are several disadvantages to using snow surface albedo for snow depth  retrievals from optical satellite measurements.” You mentioned that there are several disadvantages, but in the end you only give one, which needs to be supplemented, such as the method being easily limited by cloud coverage.

 

12. L 384-385: “The additional surface melting and leads information provided by the optical satellite retrieval can be very helpful for the possible snow depth measurements based on altimetry.” This conclusion is not the work of the paper and has not been supported in the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report (New Reviewer)

This work provides a new inspiration for the snow depth estimate, especially for resolution the differences between IMB data and satellite data in both spatial and temporal. I have the following questions,

1. In the Data section, IMB data and satellite data exist differences. Therefore, how to integrate these data for research is a key technology, and also the highlight of this paper, which can provide a new idea for similar research. I suggest to add more information about the preconditions of this treatment and the scope that can be applied. As mentioned in 3.3, “cloud less satellite images around the buoy position were selected”, so what is the less cloud in satellite images? Is there a quantitative indicator? What is the resolution of 500 * 500 cells? I suggest adding more explanations.

2. In the Results and Discussion section, author can increase the comparison with other similar work, highlighting the novelty of this work.

3. In the References section, author can add new research findings in this field in recent years.

4. There are some errors on the format. For example, in line 65, () is blank, line 84 has two repeated comma, line289,358 of 0.5m without space, and so on. Please carefully modify these errors before publication.

5. There are also some errors in the spelling. For example, is "intoduce"  "intoduce" in line 89?

The above issues should be addressed before the manuscript is publication to increase the readability of the articles.

Minor editing

Author Response

Please see the attachment.

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

 

I am sorry I cannot recommend this paper for publication.  It was difficult to read, with errors in grammar, syntax, and punctuation, and evident lack of proofreading. There are missing verbs and unidentified antecedents. The reference list was sloppily prepared, with missing punctuation and spurious capitalization. Reference 14 duplicates Reference 3.  Reference 29 duplicates Reference 28, but with the wrong first author. Reference 35 is missing the journal name. Lines 113-114 identify “AD” as “Adding-Doubling”, but neither the abbreviation AD nor the phrase Adding-Doubling occurs anywhere else in the text or in the figures.

 

In frustration, I quit reading after page 7.  But I have a more fundamental concern,  for the following two reasons:

The authors have designed a method to infer snow depth on Arctic sea ice in summer, under cloud-free conditions. The first problem is that there is essentially no snow during the 2.5 months from mid-June to the end of August (Table 1 of Webster and Warren, 2022), so the subject matter for this method is absent. In summer the surface is a mixture of dark melt-ponds and bright drained white ice. Each satellite pixel will contain a mixed field of those two types, in variable fractions.

Secondly, the requirement for clear sky prohibits use of the authors’ methods most of the time, because in summer the cloud cover over the Arctic Ocean is 80-90 percent (Figure 11 of Vowinckel and Orvig, 1970).

 

References:

Vowinckel, E., and S. Orvig, 1970: The climate of the North Polar Basin. Climates of the Polar Regions, S. Orvig, Ed., Vol. 14, World Survey of Climatology, Elsevier, 129–252.

Webster, M.A., and S.G. Warren, 2022: Regional geoengineering using tiny glass bubbles would accelerate the loss of Arctic sea ice. Earth’s Future,10, doi:10.1029/2022EF002815

comments included above

Reviewer 2 Report

This paper proposes a new method to estimate snow depth from optical imagery. By carrying out sufficient experiments and accurate result analysis, reliable conclusions are reached. In summary, this paper is well organized and presented. For these reasons, I suggest to accept this paper.

1. Please give the full name of VIS/IR and CICE in the abstract.

2. In Figure 7, the X-axis should be the differences between satellite derived snow depth minus buoy-derived snow depth.

3. The caption of Figure 8 is not consistent with the figure. Please check it.

4. This work focuses on retrieving snow depth during the melting season. Please highlight it in the title.

5. How about its performance in other seasons? Can this method be extended to all seasons?

Reviewer 3 Report

This manuscript present a method which could estimate snow depth by using a radiative transfer model. Optical satellite images are used to classify different surface  types, but author did not present the details as they said in the last paragraph in introduction section, so I do not sure whether this method is work. 

2. Optical satellite images have a significant limitation that it’s easily impact by cloud. However, in Arctic summer, the cloudy is commonly. So, how do you address this problem?

3. How depth of snow could this method detect by using albedo?

4. Section 2.2, you should give more information of the simplification of the Delta-Eddinton model.

5. L141, what is mean of “and (Fig. 6)”?

6. Equation (1), which one is Beta1 or 2?

7. Section 2.2, what is ice surface temperature exactly used for?

8. Section 3, I guess you give a wrong composing, since there is no introduction of methodology.

9. L248 “radii”?

10. You should discuss the accuracy of classification.

11. For the accuracy of snow depth, 55% of the differences are below 5cm. My question is what is the total depth? You should give percentage as well