A New Perspective on Four Decades of Changes in Arctic Sea Ice from Satellite Observations

Round 1

Reviewer 1 Report

Thanks for the invitation for the review. This manuscript is very interesting and it has assessed the sea-ice status and future using time series satellite CDR datasets. The methods are clear and not too complicated. Most of the results are clearly presented. I only have a few questions before recommending to publish.

1. In Section 2.1, both APP and APP-x datasets have been introduced. Is the APP-x only used in the analysis? In other words, are the time-series satellite data from exactly same algorithm through the four decades?
2. The reliability of the quantitative estimation in the manuscript relies on the quality of the APP-x product. The quality of the involved data record could be introduced in the manuscript so that the reader could have an overall understanding of the possible uncertainty in the results.
3. Figure 2 and 3, does the line come from a linear regression of the time-series data?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper statistically analyzed the variation of Arctic sea ice characteristics based on composited satellite observations and sea ice thickness data sets retrieved from OTIM model using the Mann-Kendall trend and the Sen's slope statistical test, and obtained several valuable results. The paper is well organized and easy understood and will be very important to monitor and predict Arctic sea ice changes. The paper could be accepted after minor modifications.

1. Some data format are not shown correctly, such as in Line 60 -3.8x103km^3 should be -3.8x10^3 km^3, similar error also can be found in line 60, 69, 70, Line 273-280 and Line 564-567.
2. Line 189, tp is the number of observations for the pth tied group. Please add some more explanations for estimating the parameter tp.
3. Line 202, please give the range for i and j in equation (5).
4. Line 249, 'Though PICA ...do not change during a year', but according to Figure 2, PICA SICA and AICA varies from 1980 to 2020. Please check this sentence or give more explanations.
5. Line 456 '3.4 Sea Ice Thickness' should be '3.4 Sea Ice Volume'.
6. Line 443-444, please add the necessary references for the previous SIC and SIE trends analysis.
7. I also give some notes in the paper and highlight several sentences in the paper. Please check the attached paper and make necessary modifications.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Reviewer Summary:

In this new study, the authors evaluate monthly/yearly changes in Arctic sea ice metrics (concentration, extent, thickness, volume) using a new measure of ice longevity from 1982 to 2020. To assess different quantities of Arctic sea ice observations, they use the AVHRR Polar Pathfinder (APP-x) for sea ice thickness and the NOAA/NSIDC Passive Microwave Sea Ice Concentration (SIC) CDR for sea ice concentration/extent. Unlike lagrangian pixel tracking for sea ice age, their new method quantifies sea ice persistence at each grid point for seasonal ice cover (SICA; covered less than 95% of days in a year), perennial ice cover (PICA; covered 95% or more days of the year), and open water area (covered 0% of days in a year). Using these definitions, the authors evaluate recent trends in SICA and PICA for both regional variability and mean Arctic Ocean averages. Finally, they extrapolate these trends to suggest an ice-free September could occur within a range of 2044-2048 to 2088-2090 using uncertainty bounds of three standard deviations.

General comments:

Overall, this is an interesting proposal for defining Arctic sea ice cover persistence. Given the wide-ranging impacts from recent changes in Arctic sea ice, it remains important to regularly update climate change indicator measurements. Thus, this work is a useful contribution for monitoring satellite observations of Arctic sea ice.

For the most part, this paper is organized and well-written. However, there are several points that need addressed before this paper should be acceptable for publication in Remote Sensing, including: (1) a number of the figures are somewhat blurry and difficult to interpret, (2) the introduction needs updated with more recent references and Arctic climate statistics, (3) a brief discussion is needed for comparison and validation of sea ice thickness data from APP-x, and (4) a refinement of the conclusions to place these new ‘persistence’ definitions of SICA/PICA in the context of other work which monitor changes in the number of days of ice cover per grid cell (e.g., Steele et al. 2019; Smith et al. 2020). I have listed some additional suggestions and comments below, which I hope the authors find helpful for revising this study.

Recommendation:

Major revisions.

Specific points:

1. There are grammar issues throughout the text that should be addressed during the proofing stage.
2. L10; What do you mean by ‘significantly’ here? Statistically significant?
3. L13; While implied, where does this paper show that the Arctic has warmed in the results section?
4. L18; I am still somewhat unclear what you mean by “a new perspective” – there have been many studies that have already evaluated the regional number of ice-free days (including grid cell by grid cell) or duration of the ice-free season in the Arctic Ocean (e.g., Bliss et al. 2019; Lebrun et al. 2019).
5. L19-20; Is this factor larger than that of sea ice age/thickness for regional susceptibility to summer melt?
6. L20-21; Does this result make the assumption that current and future sea ice trends are linear with time?
7. L26-66; Given the magnitude of the rapid change in the Arctic, it is important to refer to the latest climate statistics. Many of these references should be updated to the latest literature.
8. L31-32; Recent changes in Arctic climate variables are documented in Druckenmiller et al. 2021.
9. L32-34; Didn’t 2020 observe the 2^nd lowest September Arctic sea ice extent on record? (see Perovich et al. 2020)?
10. L45-48; Changes in seasonal Arctic temperatures are very sensitive to the future emissions scenario. Thus, revise “will likely increase”
11. L54-58; Why are trends in ice thickness only discussed through 2012, which is nearly a decade ago?
12. L62-65; What about CMIP6?
13. L66; The reference for [34] was already cited in the references at [2]
14. L68-69; “…discovered that Arctic sea ice [volume] experienced net…”
15. L85; Change “1982~2020” to “1982-2020”
16. L86; Sorry, but I am not sure I understand what you mean by “features of the trends”?
17. L100-102; State the version number for the NOAA/NSIDC Passive Microwave Sea Ice Concentration (SIC) CDR.
18. L107-109; What was the method of interpolation? Is this a 25 km EASE grid?
19. L120-122; Given that this paper only uses one dataset for sea ice thickness/volume, can you be more specific for how well APP-x compares with other data products? Are there any known biases? For example, how does it compare with PIOMAS over this time series – both spatially and (updated) overall temporal trends.
20. L133-137; Are your results sensitive to these definitions? Is there a physical reasoning for choosing 95%? Given that there are uncertainties in sea ice concentration data, would using other datasets impact grid cells that fall near the 95% threshold?
21. L145-149; What about gridded ice-free data?
22. L237-283; In my view, this almost makes it sound like there are no trends in SICA, PICA, and AICA. I suggest rewording to improve clarity.
23. L273-283; I think the scientific notation/exponents may not be completely formatted here.
24. L317-318; Why are these time periods selected? In particular, it seems that comparing March 1982 to September 2020 is not a fair timeslice comparison given the seasonal cycle and long-term trend. Perhaps the label should say 03/2020 on Figure 5-left?
25. L374; Start a new paragraph here.
26. L393-396; Does this compare well with datasets like PIOMAS?
27. L466-467; So is SIV calculated for each individual grid cell to create these spatial composites?
28. L537-541; There is very little information presented here for how this dataset compares with other observationally-derived products.
29. L547-548; Do you mean for sea ice age?
30. L551; A reference like Cohen et al. (2020) would be helpful here.
31. L554-555; These acronyms have already been referred to multiple times.
32. L590; Remove “of course”
33. L605-608; To improve open-source science, is there a code availability statement?

Figures/Tables:

1. Figure 1; Could the labels be centered for each colorbar interval? This would help improve reader clarity.
2. Figure 1; How are SSCA and PSCA defined?
3. Figure 2; I suggest writing “Annual” for the title, rather than including another acronym. Is this “sea ice area” or “sea ice extent” for calculating AICA, PICA, and SICA?
4. Figure 3; While it could just be my PDF viewer, this figure is blurry and there is a large space in the month letters – e.g., in months that start with “J”
5. Figures 5-6,8-9, 11-12 Please use different sets of colormaps for sequential and diverging schemes. See Crameri et al. (2020) and Hawkins et al. (2015) on the problems with using jet/rainbow for accessibility and interpreting scientific visualizations.
6. Figures 6,9,12 – It is quite challenging to see the values of the sea ice trends due to the figure colormap, resolution, and colorbar labels. I also cannot see the “minus signs”
7. Figure 10; The labels are fairly small here. I believe the caption should say “(PICA+SICA)”
8. Figure 13; What is the meaning of the gray shading here?

References:

Bliss, A. C., Steele, M., Peng, G., Meier, W. N., & Dickinson, S. (2019). Regional variability of Arctic sea ice seasonal change climate indicators from a passive microwave climate data record. Environmental Research Letters, 14(4), 045003.

Cohen, J., Zhang, X., Francis, J., Jung, T., Kwok, R., Overland, J., ... & Yoon, J. (2020). Divergent consensuses on Arctic amplification influence on midlatitude severe winter weather. Nature Climate Change, 10(1), 20-29.

Crameri, F., Shephard, G. E., & Heron, P. J. (2020). The misuse of colour in science communication. Nature communications, 11(1), 1-10.

Druckenmiller, M. L., Moon, T. A., Thoman, R. L., Ballinger, T. J., Berner, L. T., Bernhard, G. H., ... & Ziel, R. (2021). The Arctic. Bulletin of the American Meteorological Society, 102(8), S263-S316. https://doi.org/10.1175/BAMS-D-21-0086.1

Hawkins, E. (2015). Scrap rainbow colour scales. Nature, 519(7543), 291-291.

Lebrun, M., Vancoppenolle, M., Madec, G., & Massonnet, F. (2019). Arctic sea-ice-free season projected to extend into autumn. The Cryosphere, 13(1), 79-96.

Perovich, D., Meier, W., Tschudi, M., Hendricks, S., Petty, A. A., Divine, D., ... & Wood, K. (2020). Arctic report card 2020: Sea ice. https://doi.org/10.25923/n170-9h57

Smith, A., Jahn, A., & Wang, M. (2020). Seasonal transition dates can reveal biases in Arctic sea ice simulations. The Cryosphere, 14(9), 2977-2997. https://doi.org/10.5194/tc-14-2977-2020

Steele, M., A. C. Bliss, G. Peng, W. N. Meier, and S. Dickinson. 2019. Arctic Sea Ice Seasonal Change and Melt/Freeze Climate Indicators from Satellite Data, Version 1. [Indicate subset used]. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. https://doi.org/10.5067/KINANQKEZI4T. [Date Accessed].

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

This paper presents an interesting and useful analysis of satellite data from 1982 to 2020 regarding sea ice concentration and thickness. I recommend publication and ask the minor issues listed below be addressed in addition to the following. In several parts of the paper the division of the Arctic into SICA and PICA is justified by the statement that this “perspective on sea ice is important because it is the presence and persistence of ice in an area that directly influences local weather and climate, indirectly influences larger-scale climate, and impacts marine transportation and ecosystems.” Please add a paragraph, perhaps more, on how this analysis directly helps those issues. This will strengthen the impact of the paper, especially with suitable references to show how this approach advances the study of the changing Arctic sea ice. The suggested text could include how this analysis, compared to previous others, justify this paper’s approach. This should be part of the discussion and would be largely qualitative, I expect.

Minor but important comments are:

Lines 13-14: This paper does not directly address warming, as spelled out in lines 125-126 that describe the focus of paper. Please revise this sentence in the abstract.

Line 13-17. Thickness and volume changes should be presented in a parallel manner, i.e. include for both the same criteria such as the rate of change, percent reduction, amounts in 1982 and 2020, etc. This information is available in lines 272-281, 521-527, and elsewhere.

Line 173-178: Would it be better to say “preceding” rather than “remaining”? Perhaps I am confused, but Equation 1 compares the i^th value to the ones preceding it, not the remaining ones. In line 178, data values i and j as far as Equation 1 shows, indicate that when the i^th value is, say 10, you are summing up the sign, based on the j values from 1 to 9, so they are not “sequential” in that sense. It would also help the reader to identify here what the “tied group” is, for reference in lines 189-190. Please spend some time clarifying the text.

Lines 298-300 and Table 1. Line 300 notes that “September is not statistically significant” for SICA with a value of 0.514 but seems to imply that October and November are significant. They have values for alpha of 0.933 and 0.250 that are more, and less, than September. It is not clear to me how September is not, and Oct and Nov are, significant.  Also, note that Table 3 has the non-significant values highlighted in red. For consistency, it would be helpful to do the same for the other tables or omit the red entirely.

Line 456 – Section 3.4 is Sea Ice Volume

Figure 1. LAND, SSCA, and PSCA are defined for the first time in this legend. It will help the reader to define them in the paragraph of text in lines 133-149.  Also, my choice would be to have the label names placed at the midpoints of each color, to the right of the colorbar, rather than at the intersections.

Figure 5. The left figure is labeled 1982 at the top, but the legend says 2020.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

General comments:

The revised manuscript has addressed all of my previous concerns. It now places the results in better context with the literature and includes updated Arctic climate statistics. In my view, this study is now acceptable for publication in Remote Sensing.

Recommendation:

Accept.

Author Response

Thank you for your recommendation for publication in Remote Sensing.