Preprints
https://doi.org/10.5194/egusphere-2026-2569
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/egusphere-2026-2569
© Author(s) 2026. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
05 Jun 2026
| 05 Jun 2026
Status: this preprint is open for discussion and under review for The Cryosphere (TC).
New high resolution sea ice leads and floes classification in the CNES SWOT product
Abstract. Global observation of ocean and sea ice dynamics relies heavily on space altimeters. However, due to their high latitudes, which are less covered by space altimeters, and the ice cover, which hides the sea level, polar oceans are still poorly observed and modeled. The SWOT altimeter significantly increases the density and quality of measurements up to 77° latitude, covering the entire Southern Ocean and a large part of the Arctic Ocean. Still, these observations require a distinction to be made between measurements taken on water and measurements taken on ice. The CNES L3 Unsmoothed 250m v2.0.1 product includes a new flag that identifies the type of surface observed for each pixel, enabling processing tailored to the object under study. Here we present the methodology used to calculate this flag and evaluate the results obtained using space imagery, nadir altimetry and OSI SAF concentration products.
How to cite. Jestin, G., Fleury, S., Raynal, M., Alves, M., Piras, F., Treboutte, A., Boy, F., Benabdillah, I., Yu, L., and Dibarboure, G.: New high resolution sea ice leads and floes classification in the CNES SWOT product, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2026-2569, 2026.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Download & links
Download & links
- Preprint
(9713 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 29 Oct 2026)
Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor
| : Report abuse
- RC1: 'Comment on egusphere-2026-2569', Anonymous Referee #1, 24 Sep 2026 reply
Viewed
Total article views: 506 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 05 Jun 2026)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 310 | 165 | 31 | 506 | 29 | 27 |
- HTML: 310
- PDF: 165
- XML: 31
- Total: 506
- BibTeX: 29
- EndNote: 27
Viewed (geographical distribution)
Total article views: 464 (including HTML, PDF, and XML)
Thereof 464 with geography defined
and 0 with unknown origin.
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
Latest update: 07 Oct 2026
Gwenael Jestin
CORRESPONDING AUTHOR
LEGOS, Université de Toulouse, IRD, CNES, CNRS, UPS, Toulouse, France
Sara Fleury
LEGOS, Université de Toulouse, IRD, CNES, CNRS, UPS, Toulouse, France
Matthias Raynal
CNES, Ramonville Saint Agne, 31520, France
Marta Alves
Collecte Localisation Satellites, 31520, Ramonville-Saint-Agne, France
Fanny Piras
Collecte Localisation Satellites, 31520, Ramonville-Saint-Agne, France
Anaëlle Treboutte
Collecte Localisation Satellites, 31520, Ramonville-Saint-Agne, France
François Boy
CNES, Ramonville Saint Agne, 31520, France
Inès Benabdillah
CNES, Ramonville Saint Agne, 31520, France
Louise Yu
CNES, Ramonville Saint Agne, 31520, France
Gérald Dibarboure
CNES, Ramonville Saint Agne, 31520, France
Short summary
With its wide swath of 120 km, the SWOT altimeter offers unrivalled spatial and temporal coverage of the polar oceans. But frozen oceans are difficult to study because the ice cover only reveals the open ocean through cracks and polynyas. Whether studying the ice or the ocean, leads and floes must be distinguished. This classification is now available in the CNES L3 product at a resolution of 250 m. This paper outlines the methodology and improvements of the derived ice concentration maps.
With its wide swath of 120 km, the SWOT altimeter offers unrivalled spatial and temporal...
General comments:
The paper presents the sea ice floe/lead classification approach used to create a new flag available in the SWOT L3 Unsmoothed 250m v2.0.1 (CNES) product. The methodology is sound and the classification approach is well-motivated. The SWOT results are validated using nadir instruments (ICESat-2 and Sentinel-3) and sea ice concentration maps (OSI SAF), with the validation results supporting the reliability of the approach. The topic is timely and relevant to TC's scope, given the growing interest in SWOT's capabilities for sea ice monitoring, and the paper presents a novel product/method with clear potential value to the community. I also appreciate that the authors are already aware of several limitations in their approach and thoughtfully discuss potential ways to address them, which reflects a careful and transparent treatment of the work.
Overall, I believe this study is a valuable contribution and deserves publication in The Cryosphere. However, I highly recommend that the following major and minor comments be addressed before final publication.
Major comments:
Abstracts
The abstract describes the motivation, problem, and the resulting product, but does not report any quantitative results or validation findings. Consider adding a sentence or two on the key results, so readers get a sense of the outcome without reading the full paper.
1 Introduction
The introduction explains that sea ice presence strongly affects sea level measurements, limiting our knowledge of the polar ocean. It lays out the situation, the established solution of using space altimeters, their limitations, and how SWOT could improve on this. It also includes statistics on the global ocean that freeze seasonally to justify the need, but these figures are not attributed to any source. I would recommend adding citations throughout the introduction where claims or figures are presented, to ground the motivation and help readers understand how the proposed approach relates to, and improves on, existing work.
Line 22-23: This 300 m figure varies by altimeter. Worth specifying which altimeter(s) this applies to or rephrase to reflect that sampling density varies across missions.
2 SWOT and datasets used
Line 51-53: The text does not mention that there is a ~20 km gap in KaRIn's coverage at nadir, and that the nadir altimeter only samples a single line through it, not 2D. Worth making that explicit here.
Line 67, 78-80: It is unclear whether these observations are the authors' own findings or drawn from another study, since no citation is given. Please clarify.
3 Sea ice classification methodology
Section 3 covers a lot of important material, and it is clear that considerable effort went into this analysis. That said, the section is too long and reads somewhat densely in places. The flow between certain ideas, particularly where the discussion transitions between different processing steps, could be clearer. A revision pass to trim some of the redundant content and denser passages would make logic easier to follow.
5 Assessments using other spatial observations
Several in-text citations do not have corresponding entries in the reference list. On page 15 alone, I noted several: Matthews 1975, Longépé et al. (2019), Neumann et al. 2019, Kwok et al. ATBD 2022 and Preußer et al. 2019. I also came across similar cases elsewhere in the paper, so I would suggest the authors do a full cross-check of all in-text citations against the reference list to catch anything missing.
Line 330, 398: Both equations define sea ice concentration using the same variable (SIC), but the right-hand sides differ. It would help to use distinct notation (e.g., subscripts) to differentiate them. Also, define/explain the parameters used in the first SIC equation.
Line 404-405: Since the reader has to interpret 12 separate maps, it would help to briefly note what to look for, similar to the level of detail given in earlier sections when discussing feature-specific patterns. This would make the claim easier to verify and the figure easier to follow.
Figure 11: I understand the reasoning that a narrower colorbar (75–100%) helps show more contrast within ice-covered regions. However, since both maps cover the same region and month, if this reasoning holds, it should apply equally to the SWOT map as well. If the narrower range was intentionally applied to only OSI SAF, it would help to explain why the same adjustment was not made for SWOT.
Line 463-464: The v3 product mentioned as an upcoming release appears to have actually already been released earlier this month. Worth updating the sentence and tense to reflect the current status.
Minor comments:
There are quite a few acronyms throughout the paper that are not spelled out at first mention, e.g., SWOT, OSI SAF, KaRIn, Space Agency names, SARM, CRISTAL, ATLAS. These are just a few examples, so it would help to do a full pass through the paper to catch and define any others.
Line 46: Refers to the Canadian Space Agency as 'CNSA,' but this actually refers to the China National Space Administration. The correct abbreviation is 'CSA.'
Line 54-55: This sentence is a bit tricky to follow as written. Consider rephrasing for clarity.
Please keep the product and version names consistent throughout the paper, as they are referred to differently in a few places, for example, CNES L3 Unsmoothed 250m v2.0.1, version 2 of the CNES L3 250m, L3_LR_SSH, OSI SAF, OSI SAF/AMSR-2
Line 94-96: It would help to explicitly mention 'L3' here, so it is clear that this part of the discussion is specifically about the L3 product. Also, provide citation for the product handbook.
Dates are formatted differently throughout the paper, some appear as 'DD Month' others as 'Month DD' and some as numbers only. Please use a single, consistent format.
Line 119-120: Listing all four values with their labels explicitly would make this clearer.
Figure 3: Does not include units for the distances shown. Please add the units for clarity.
Line 132: It is not clear which cutoff (40 km or 2 km) applies to which track direction in the first filter, unlike the second filter, which specifies this explicitly. Spelling that out would make the methods easier to follow.
Line 158: The phrase 'the two first input sets' would read more naturally as 'the first two input sets.'
Line 259: This sentence is a bit tricky to follow as written. Consider rephrasing for clarity.
Line 271: The text states that the MODIS image is displayed on the right panel, but in Figure 8 it actually appears on the left. Also, please add the year to the date (April 4th) for clarity.
Section 4,5: I believe 'crossing points' is meant to refer to 'crossover points,' as commonly used in altimetry literature. If so, it might help to use the more standard term for consistency, though I defer to the authors' preference here.
Line 311: Did you mean 'higher' resolution here?
Line 330, 398: The equations do not appear to follow the journal's preferred format and are not numbered. Please check all equations throughout the paper and ensure they are formatted and numbered consistently.
Line 353: Explicitly stating that Sentinel-3 is a radar altimeter, the way ICESat-2 is labeled lidar, would be better for clarity.
Line 353: Earlier in the paper, Sentinel-3's resolution is given as ~300 m, but here it is stated as ~330 m. Please use a consistent value throughout to avoid confusion.
A few figures, e.g., Figures 9, 10, and 11, appear in the section following where they are first discussed, rather than within their own section. Please check and correct the placement of these and other figures throughout the paper.
Figure panel alphabets use single brackets, figure captions use double brackets, and in-text references use no brackets at all. Please use a consistent format for these throughout the paper.
Figure 10: In the SWOT colorbar, 'unsure leads' appears twice, likely one should be 'unsure floes.' In the S3 colorbar, there is a separate 'ocean' class alongside 'leads,' but it is unclear how the two differ since leads are open water too, and no 'ocean' pixels appear visible in the figure. The caption notes that the red lines indicate the Sentinel-3 across-track resolution. It would help to also mention that the Sentinel-3 classification is shown within this region.
Line 482-485: Two sentences in a row starting with 'However.' Might be worth varying the transition or combining the points for smoother flow.