Weddell Sea polynya analysis using SMOS–SMAP apparent sea ice thickness retrieval

Mchedlishvili, Alexander; Spreen, Gunnar; Melsheimer, Christian; Huntemann, Marcus

doi:https://doi.org/10.5194/tc-16-471-2022

Articles | Volume 16, issue 2

https://doi.org/10.5194/tc-16-471-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/tc-16-471-2022

© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 16, issue 2

Research article

|

09 Feb 2022

Research article |

| 09 Feb 2022

Weddell Sea polynya analysis using SMOS–SMAP apparent sea ice thickness retrieval

Alexander Mchedlishvili, Gunnar Spreen, Christian Melsheimer, and Marcus Huntemann

Download

Final revised paper (published on 09 Feb 2022)
Preprint (discussion started on 17 Jun 2021)

Interactive discussion

Status: closed

RC1:
'Comment on tc-2021-138', Anonymous Referee #1, 23 Jul 2021

The comment was uploaded in the form of a supplement: https://tc.copernicus.org/preprints/tc-2021-138/tc-2021-138-RC1-supplement.pdf

Citation: https://doi.org/10.5194/tc-2021-138-RC1
- AC1: 'Reply on RC1', Alexander Mchedlishvili, 22 Sep 2021
  
  The comment was uploaded in the form of a supplement: https://tc.copernicus.org/preprints/tc-2021-138/tc-2021-138-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/tc-2021-138-AC1
RC2:
'Comment on tc-2021-138', Anonymous Referee #2, 03 Aug 2021

Summary,

Weddell Sea polynya is a large anomalous opening that does not always appear and had a 40 years absence before the 2017 occurrence again. Differing from many costal polynyas that are mostly due to the katabatic wind events, the Weddell polynya was mostly due to upward of warm deep water that melts the overlying sea ice and stops the surface water from refreezing to become ice. If this warm water continues, the polynya continues; otherwise, refreezing happens and thin ice forms, until the next upward of warm water to melt the ice and open the polynya again. This paper develops a SMOS and SMAP thin ice thickness product to map the thin sea ice (indicating the polynya), as well as the ERA5 surface wind reanalysis data to develop/support a theory that direct atmosphere forcing is a major factor for the polynya occurrence, beside the warm deep water forcing. And they conclude various different factors must occur simultaneously for the polynya to occur.

Specific comments

Based on the definition of this type of polynya, open water is obvious indicator of the polynya. Thin ice is an indirect indicator that shows open happened here but now thin ice and will soon become thick ice if the warm water upward does not appear again in a later date. In my opinion, the open water is better mapped by the ASI ice concentration map (6.5km resolution) as indicated in the paper and prior studies. The new thin ice thickness product is 45km and is not a good indicator of the small opening; also there is publication about thin ice thickness retrieval from passive microwave remote sensing such as the AMSE-E/2, that could be up to 6.5km in spatial resolution (see paper Dai et al., 2020, Remote Sens. 2020, 12(9), 1484; https://doi.org/10.3390/rs12091484). However, this paper did not mention this method in their paper at all, indicating some lack in literature review. To confirm and validate their thin ice method for mapping polynya, Sentinel-1 SAR image is a much better approach than the ASI concentration since the Sentinel-1 has much higher spatial resolution (also in Dai et al., 2020 paper and other papers).

In the paper, authors made it clear that their SMOS-SMAP retrieval algorithm assumes ~100% SIC, while there is low SIC with polynya. This causes a concern on their results. For example, in their text line 108-110, “Thin ice thickness is … a combined ice area and thickness anomaly and not be used to calculate... ice volume…”.

This paper claimed that it is the first time to confirm that wind is a major factor for the Weddell Sea polynya, although I am not sure if they have enough data to confirm this finding from these polynya events 4o years ago. Otherwise, are you so confident that the results from these two years of data can apply to other times?

they conclude various different factors must occur simultaneously for the polynya to occur. But this statement is not an approved statement.

the paper writing needs to improve, I have listed a few in the details below, but many can be found throughout the paper. One big comment is the section 4 (conclusions). Most of the content in this section should be in the discussion not in the conclusions

Details:

L2, “fully opened again on …2107”, but figure 1 and text shows 2016 opened.

L3, “lasted until melt” is not clear and confused. Maybe change to “lasted until the summer melting season”? “80 days, 2017,” should be “80 days in 2017.”, right? are you sure it reached early December before all surrounding ice melted?

L4, “actually was not the…”, what is the subject of this sentence? You missed it.

L59-61, I have question for this purpose of the study: “we aim to ….”, why you want to using the thin ice thickness which is not already existing and will be much coarse resolution as compared with the exiting ASI ice concentration data. AMSR-E/2 can also be used to derive thin ice thickness. Is this your thickness compatible with the AMSR-E/2 derived?

L85: a root mean square difference (RMSD)

L92, “growing sea ice”, do you mean “sea ice growing season”?

L110-111, I am not sure how long this thin ice would last once the upward of warm deep water stops or weaken, since most of the time the Weddell Sea polynya is an open water area as indicated (for 80 days in 2017). Once the upward of warm deep water stops, thin ice would form and would thicker, also thicker ice from surrounding would come to fill the open and thin ice area soon as I can imagine.

L141: Can you explain why 2017 and 2018 are chosen?

L143-144: Can you clarify the strength of SMOS-SMAP SIT compared to the SIC datasets?

L149: Can you mention this spatial resolution of ASI SIC in section 2.2?

L169, “….area that is classified as open water”. I have question. this means it was 0% open water before Sept 13? if this is the case, then 100% ice covered, then ice thickness should be higher but why it was actually lower compared with after Sept 13? this is not possible. am i wrong?

L170, 0% open water should come with high SIC and SIT, right?

L175: There is no section 3.2. Please check it. Should it be “4. Discussion”?

L180: Can you add any references here?

L191: Can you present any statistical parameters, such as correlation coefficient (R)?

L193-194: I guess the higher resolution of ASI SIC can affect this result. In Figure 3 and Figure 6, it seems that ASI SIC (6.25 km) has a much finer spatial resolution than SMOS SIT (45 km?). If so, the ASI SIC data should underestimate the thin-ice (or open water) area compared to the SMOS SIT data. Maybe you need to discuss the effect of different spatial resolution.

L201-203, sentence “we see… freeze up”, please break up…

L213, “east and southeast directions”, but the figure 4 does not show these directions.

L216-219, Figure 7: Can you mark the extent of sea ice anomaly area in Figure 7?

L226-227, “therefore also…”, please cite papers here since it is your finding.

L 237-238: I just wonder if there is any possibility of sea ice advection (e.g. drift of thin sea ice from other regions?). By seeing the time-series video of SIT, you may be able to confirm if those thin ice events are all really “polynya-type” events or they are advection of thin ice from surrounding area.

L268-269, the last sentence “Moreover, it is the combination….”. this sentence is suspicious, since your paper did not approve it.

L281, “…low SIC (most likely minor lead openings) is recorded”. I really hope the paper use the Sentinel-1 SAR data to validate or confirm…

Figure 1: Can you briefly explain how to define/distinguish “polynya events” and “ice thinning anomalies”?

Figure 3. what are the resolutions for them? it is really not easy to match the two sets.

Figure 4 and Figure 5: I am just curious why you mention 2017 (Figure 5) first, and then 2018 (Figure 4). Would it better to mention 2017 first prior to 2018? In Discussion, you describe 2017 first and then 2018, so it is somehow confusing to read the text and figure together. And same for Figure 7 (2018) and Figure 8 (2017). Also Figure 4b, if 0% SIC for open water, should be 100% sea ice, but figure 4c shows 0% ice from August 11-Sept 4, please explain?

Figure 6, really the SIC and SIT do not match much, except the 2016 and 2017.

Figure 8: Same to Figure 7, can you mark the extent of sea ice anomaly area in Figure 8?

Citation: https://doi.org/10.5194/tc-2021-138-RC2
- AC2: 'Reply on RC2', Alexander Mchedlishvili, 22 Sep 2021
  
  The comment was uploaded in the form of a supplement: https://tc.copernicus.org/preprints/tc-2021-138/tc-2021-138-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/tc-2021-138-AC2
RC3:
'Comment on tc-2021-138', Anonymous Referee #3, 24 Aug 2021

The comment was uploaded in the form of a supplement: https://tc.copernicus.org/preprints/tc-2021-138/tc-2021-138-RC3-supplement.pdf

Citation: https://doi.org/10.5194/tc-2021-138-RC3
- AC3: 'Reply on RC3', Alexander Mchedlishvili, 22 Sep 2021
  
  The comment was uploaded in the form of a supplement: https://tc.copernicus.org/preprints/tc-2021-138/tc-2021-138-AC3-supplement.pdf
  
  Citation: https://doi.org/10.5194/tc-2021-138-AC3

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (further review by editor and referees) (04 Oct 2021) by Ludovic Brucker

AR by Alexander Mchedlishvili on behalf of the Authors (14 Oct 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (26 Oct 2021) by Ludovic Brucker

RR by Anonymous Referee #1 (09 Nov 2021)

Suggestions for revision or reasons for rejection

I thank the authors for thoughtfully addressing the concerns raised in the first round of reviews. While I am generally satisfied with the revised manuscript, which has substantially improved, I think there are several issues that need to be addressed before publication can be recommended. My main issue is with the revised Introduction section, where the authors attempt to summarize the oceanographic literature on polynya formation. As I detail below, the authors introduced statements that I view as misleading and inaccurate.

That said, I think the issues mentioned above are relatively minor and the scientific core of the paper remains valuable. While this new sea ice thickness retrieval comes with many caveats, the authors are clear about its limitations and demonstrate that it provides information that is distinct and independent of existing sea ice area information. Future work is needed to validate these retrievals and quantify their uncertainties, but this study is a good start.

Detailed comments:

- Title: This title is fine but many people in the ocean community understand "Weddell Sea Polynya" to mean large open-ocean polynyas that span *most* of the Weddell Sea, which this paper does not analyze. I would suggest a more general title, such as "Analysis of open-ocean polynyas in the Weddell Sea using SMOS-SMAP apparent sea ice thickness retrieval"

- Line 2 (abstract): Related to the above comment, this definition of "Weddell Sea open-ocean polynyas" does not quite align with definitions found in the literature. While open-ocean polynyas are often found near the Maud Rise seamount, they are not exclusive to this specific region. Perhaps an opening sentence like the following would suffice, "The Weddell Sea is known to feature large openings in its winter sea ice field, otherwise known as open-ocean polynyas...."

- Lines 3-4 (abstract): "2017, however, is far from the only year that the imprint of a polynya can be identified at this location." This is an odd sentence considering the previous sentence states that these polynyas have intermittently appeared over the past 40 years. Perhaps the authors wish to communicate that their new estimates of sea ice thickness reveal periods of the substantial thinning that are not apparent in existing observations of sea ice area/concentration.

- Lines 23-34: I would delete "surprisingly".

- Lines 26-28: I would suggest referring to these openings as "open-ocean polynyas". I think the use of lower-case "p" in the term "Weddell Sea polynya" only makes things more confusing.

- Line 29: "The Weddell Sea polynya is an anomalous opening in sea ice that is generally classified as an open-ocean polynya" See above comment.

- Line 32: "ocean processes" is rather vague. I prefer the original description that mentions the upwelling of warm water.

- Lines 32-34: "The main mechanism that preconditions the Weddell Sea polynya is described by Martinson et al. (1981) as destratification of water masses leading to a raised pycnocline. The raised pycnocline compresses the winter surface mixed layer that is made dense through heat loss and brine rejection from ice formation, resulting in density overturning." This explanation is rather muddled and arguably incorrect. I would delete this sentence since the one that follows gets right message across.

- Lines 37-45: This section contains several statements that are confusing, misleading, and at times inaccurate. This is perplexing since the section was mostly fine in the original submission. To give a few specific examples: I don't understand what is meant by "...destratification of water masses leading to a raised pycnocline..." or the idea that a raised pycnocline "compress[es]" the mixed layer. Additionally, the authors appear to use "weak stratification" and "lack of stratification" inter-changeably, which are very different.

Martinson et al. (1981), which the authors try to summarize, proposes that open-ocean polynyas in the Weddell Sea may be initiated when the winter mixed layer becomes sufficiently dense that it triggers deep convection, which results in the upward mixing of warm deep water. These events may be preconditioned by processes that weaken the upper ocean stratification, such as the anomalous upwelling or advection of high salinity water into the surface mixed layer. I don't think the authors need to spend more than a couple sentences reviewing these mechanisms. Like I mentioned before, the initial text was mostly fine.

- Lines 44-53: Reading this for a second time, I think this extended discussion about Taylor columns and flow-topography interactions lacks appropriate nuance and does not add much to the paper. With my previous comment in mind, I would also suggest reducing this entire paragraph to a 1-2 sentences. For the purposes of this study, I think it is fine to simply state (with supporting citations) that interactions between the mean flow and topography preconditions the area above Maud Rise for anomalously high vertical heat fluxes, which favors thinner sea ice.

- Lines 58-59: "Direct atmospheric forcing" and "large-scale climate processes" are framed as two separate entities but the former is a subset of the latter.

- Lines 117-118: "Patilea et al. (2019) estimated the uncertainty at 90% SIC from SIT values up to 50%." While I get the gist of this discussion, I'm a bit confused by this line.

- Lines 123-124: "...this compromises the validity the of the sea ice thicknesses retrieved..." Typo. "the" after validity is not necessary.

- Lines 125-126: "While the degree by which the ice thins is difficult to quantify in terms of uncertainty, our analysis has shown that the pattern of thin ice anomalies above Maud Rise is not random nor is it identical to the distribution of low SIC areas..."

Is there a simple, quantifiable way of demonstrating this statement? I wish I suggested this in my first review, but it would be interesting to compare these SIT measurements with available in situ ocean salinity measurements or perhaps a data-assimilative model such as the Southern Ocean State Estimate. Periods of thinning should correspond to periods of mixed layer freshening. In fairness to the authors, I will not demand that they do this validation but I hope they consider doing so in the future.

Lines 294-295: "anomalous episode of sea ice" is a bit vague. How about "...episodes of anomalous wintertime sea ice loss..."?

Line 321-322: "In the end, strength of the wind magnitude present above the region is most directly connected with drops in ASIT and SIC." This is an interesting finding!

Hide

RR by Anonymous Referee #2 (16 Dec 2021)

Suggestions for revision or reasons for rejection

Overall, I see the paper has been improved a lot, although I am not sure why they do not want to comment or discuss the Nihashi et al. 2017 and Dai et al. 2020 papers and cite them. If the method used in these two papers had problems in their opinion, it is a good place/time to say a few words, so future improvements/discussions can be made, right? I notice the English writing has been improved as well, but still need more work. For example, I went through the first part of the paper (up to page 7) and I list quite a few here. I would hope the authors to go through the entire paper and make corrections before final acceptance.

Line 6, to make a consistent tense through the paper, I would think “we found” should be “we find”
L7, change “occurring” to “occurred”
L16-17, change “effect” to “phenomenon”? “is” to “was”, “and similar anomalies are identified” to “but also identified”…
L23, change “are” to “were”
L26, change “will refer to” to “refer”
L42, to cite all previous /published papers, I believe the past tense should be used. In this line and many other places please make the changes. “go further” to “went further”…
L54, the same as comment above, “describe” to “described”
L55, “discuss” to “discussed”. Since this already happened, the past tense should be used. Everything you are doing, discussing and finding, should be present tense in this paper. So readers can separate this paper’s work (present tense) and previous works (past tense).
L63, “go further” to “went further”
L64, “hypothesize” to “hypothesized”
L65, “preventing stabilization” to “preventing that stabilization”
L76, add “that” before “the existence of …”; change “conditions that although” to “conditions although”
L77, change “are” to “is”; change “use” to “used”
L100, change “which” to “This”
L158, change “report” to “reported”
Figure 1, based on the 66S and 3E, I don’t see the thin ice is on top of the Maud Rise? Do you mean north of the Maud Rise?
L168, change “it the area” to “it is the area”
L174, change “shows” to “showed”
L175, change “there is” to “there was”

Hide

ED: Publish subject to minor revisions (review by editor) (17 Dec 2021) by Ludovic Brucker

AR by Alexander Mchedlishvili on behalf of the Authors (24 Dec 2021) Author's response Author's tracked changes Manuscript

ED: Publish as is (03 Jan 2022) by Ludovic Brucker

Short summary

In this paper we show that the activity leading to the open-ocean polynyas near the Maud Rise seamount that have occurred repeatedly from 1974–1976 as well as 2016–2017 does not simply stop for polynya-free years. Using apparent sea ice thickness retrieval, we have identified anomalies where there is thinning of sea ice on a scale that is comparable to that of the polynya events of 2016–2017. These anomalies took place in 2010, 2013, 2014 and 2018.