Multi-Temporal Influences of Large-Scale Atmospheric Patterns On Antarctic Coastal Polynyas

Ward, Jason; Raphael, Marilyn N.

doi:10.5194/egusphere-2026-1811

Preprints

https://doi.org/10.5194/egusphere-2026-1811

Preprints

04 Jun 2026

| 04 Jun 2026

Multi-Temporal Influences of Large-Scale Atmospheric Patterns On Antarctic Coastal Polynyas

Jason Ward and Marilyn N. Raphael

Abstract. During the austral ice advance season, sea ice expands across the Southern Ocean, significantly reducing gas, heat, and moisture transfer between the atmosphere and ocean. Situated along the Antarctica coastline are areas of open water, minimal sea ice concentration, or relatively thin ice – called coastal polynyas. Coastal polynyas continuously reconnect the atmosphere and ocean. Antarctic coastal polynyas’ greatest geographic influence occurs through their contribution to deep water formation and the global Thermohaline Circulation. Coastal polynya impacts are partially controlled by their size. Larger polynyas allow greater rates of sea ice production, bottom water formation, and total biological productivity. This study examines the role that large-scale atmospheric patterns play in Antarctic coastal polynya size variability. The Southern Annular Mode (SAM), El Niño-Southern Oscillation (ENSO), and the Amundsen Sea Low (ASL) together significantly influence the sizes of 20 of the 25 Antarctic coastal polynyas in this study. The SAM exerts its strongest influence in East Antarctica, where a +SAM stunts coastal polynya growth. ENSO, a tropical phenomenon with significant West Antarctica teleconnections, promotes polynya growth during La Niña. The ASL, an exclusively West Antarctica phenomenon, enhances polynya growth in the Amundsen and Bellingshausen Seas as it intensifies and as it migrates northeast. In contrast, an eastward migration of the ASL suppresses polynya expansion in the Weddell Sea. Variability in the SAM and in the ASL’s intensity and latitudinal location impact monthly and seasonal polynya size variability. Variability in ENSO and in the ASL’s intensity and longitudinal location impact annual, seasonal, and monthly polynya size variability, respectively.

Received: 31 Mar 2026 – Discussion started: 04 Jun 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

Jason Ward and Marilyn N. Raphael

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-1811', Anonymous Referee #1, 01 Jul 2026
Review of “Multi-Temporal Influences of Large-Scale Atmospheric Patterns On Antarctic Coastal Polynyas” by Ward & Raphael
In this manuscript the authors correlate the size of multiple Antarctic coastal polynyas with indices of the Southern Annular Mode, El Niño-Southern Oscillation, and the Amundsen Sea Low. The main results of the study are multiple correlations between these modes of variability and the polynya size. Because the author’s present such a large number of correlations, it is difficult to determine the key messages that should be taken from this study.
While I don’t believe that the results of the study constitute a particularly large scientific impact, I think they are publishable provided a number of major issues are resolved. The first of the major issues I have already mentioned: a lack of clear, key results that should be taken from the study. Second, I believe the presentation of the results, particularly in the figures, requires some improvement, not least to make them more engaging and presentable. These and other major issues are discussed in more detail below.
The lack of great significance in the study’s results, combined with the major issues detailed below, meant I was unsure whether or not to recommend “rejected” or “reconsidered after major revisions”. I have opted for the latter in my formal recommendation.
Major
Figure 1 is quite rudimentary, and I think needs significant improvements to its presentation. Precisely, what do the sets of arrows represent? What the ENSO arrows represent is most unclear. Perhaps this can additionally show sea ice area from the satellite dataset to make it more engaging and informative.

Figures 3, 4, 5, 6, 7, 9, 10. All of these figures present similar-looking bar charts, which collectively make for a repetitive experience for readers. Figure 8 shows correlations coded by colour and is functionally equivalent. I urge the authors to consider alternative methods to portray their key results.

Figures 4, 5, 6, 7, 9, 10. None of these figures can be easily interpreted without having to refer back to Fig 2 for the locations of each polynya. I suggest adding extra info to the figure that aids readers to more quickly know which sector of Antarctica the numbers on the x-axis correspond to.

The vast majority of the sections 2 and 3 is a sequence of stated correlations without any particularly major conclusions or precise focus. This makes it very difficult to know what the key points readers should take away. The authors must make clearer the handful of key results that readers should take away. These key results should be made clear in the abstract and within the text.

Can you provide a reference or evidence for the claim that a positive ENSO trend will “promote the eastward ASL migration”. It’s not obvious that the ASL’s response to ENSO oscillations would be the same as to ENSO trends.

Please show more clearly what the modes of variability look like. How do atmospheric patterns change during each phase of the modes. Such figures could present composites of pressure/wind fields. Similar figures showing composites of sea ice area could also be presented. This would present an opportunity to make significantly more engaging figures.

L391 states: “When the influence of other large-scale atmospheric phenomena, such as the SAM and ENSO, are removed from the ACP”. I don’t think it makes sense to discuss removing the influence of these modes from indices of the ASL. The ASL is not an independent climate mode, and its variability comes from broader influences. I also don’t think it is accurate to claim that the RCP omits the influences of SAM/ENSO when compared to ACP. Eg, in the case of ENSO, the prototypical Rossby wave train emanating from the tropical Pacific sets up a localised pressure anomaly over the Amundsen Sea which will project onto RCP. L402 also says that RCP “represents the regional effect of the ASL, with the large-scale background influences removed”, which I think is untrue, so I suggest deleting this. Other parts of the text, such as the paragraph at L473, continue to discuss the RCP as being independent of SAM/ENSO, and these also need editing. It is certainly worthwhile to relate variability in the RCP, ACP etc to variability in SAM/ENSO. I suggest shifting any discussion of “removing the influence of ENSO/SAM” more towards a discussion of how ENSO/SAM are behind the variability in the ASL that is important for the polynyas.

Minor
L19-22. “Variability in…” x 2. The final two sentences of the abstract are vague. What are the “impacts” that are referred to in these sentences?
L38. “Static, or slow-moving, features form the windward boundary of these latent heat-induced polynyas”. This is repetition of details in L32.
L44. “surface waters become dense enough to sink to the continental shelf”. The waters are already on the continental shelf. Is the meaning “towards the ocean floor”?
L45. I don’t think that “spill into Antarctic Bottom Water” is an accurate description of the process. The waters can transform into AABW as they cascade down the continental slope.
L47. “gives Antarctic coastal polynyas global influence”. What is this influence? Can this be written more precisely?
L74. “They found that trends in RISP area and timing of the RISP’s annual occurrence are influenced by trends in ENSO and the SAM”. What is this influence that is being described?
L130. “While 70 polynyas…” I found this sentence difficult to follow. Perhaps it can be split into two to make it clearer.
L137-141. I found this paragraph very difficult to follow throughout. What does “aggregated” mean precisely? What are the “initial annual means”? Does “standardized deviations” mean “standard deviations” or something else? Can the meaning of the “over-allocation” of correlation also be expanded on please?
Figure 2. (a) Polynyas 1,2 and 25 are all in the Ross Sea, but appear at different ends of the later bar charts. Perhaps there is a more natural location to start counting the polynyas from, so they are more neatly separated. Somewhere around Haakon VII Sea looks most obvious. (b) The sea ice drift vectors in panel b use odd units. How about m/s?
L148. There is a table immediately after Figure 2, but this is neither a formal, captioned table nor is it properly incorporated into the text.
L232. “With the positive long term trend previously presented in this paper (Fig. 3)…” As far as I can tell, Fig 3 doesn’t show the sign of the trend.
L244. Please comment on the suitability of not using any lags between modes and sea ice response.
L253. This equation is not part of any sentence, but should flow grammatically. What does the “~” symbol represent precisely?
L256. “As stated above…” This sentence is full of repetition, and I suggest cutting it down substantially.
L272. “Thus, polynyas that grow with easterly winds…” The conclusion in this sentence relies on wind-polynya interactions being a symmetric process. This may be true, roughly speaking, but you should state here that you are working under this assumption.
L396. It’s not clear how the polynyas in the Amundsen/Bellingshausen sectors are related to the Ross Gyre. In fact, I don’t think they are related in the way that the text suggests: the Ross Gyre is further to west and is in the deep ocean. Therefore, I suggest deleting this sentence.
L483. This equation is not part of any sentence. It should flow within the paragraph. I also disagree with the equation (see major comment above), so I think it can be removed.
L484. This paragraph discusses more ACP-polynya correlations, but we are now in the RCP section. These should be moved to the ACP section.
L629. The Amundsen and Bellingshausen polynyas do not contribute to AABW formation at all, so this sentence needs editing.
Other
Abstract & elsewhere. In the phrase “polynyas’ greatest geographic influence”, the word “geographic” is a bit vague and I think is misplaced. Perhaps “oceanographic” is more accurate.
L15. While the meaning of syntax like “+SAM” can be inferred, I think it should be explicitly stated.
L16,17. Change “West Antarctica” to “West Antarctic”.
L33, 269. “ice tongue” --> “ice shelf”
L33. “iceberg” --> “grounded iceberg”
L69. “It is important to note…” This sentence feels overly tangential, and I suggest cutting it.
L80. “upward transport of oceanic heat, which is associated with the basal sea ice melt”. How are they associated? Can the text be made more precise?
L345 + other locations. “significant positive relationship”. Does “relationship” here mean “correlation”? There were other instances where “relationship” is used and it is not clear if “correlation” is the intended meaning. I suggest finding and changing of all these to make the meaning clear.
L395. “negative westerly” --> “easterly”
L570, 576. “circumAntarctic” --> “circum-Antarctic”
Citation: https://doi.org/10.5194/egusphere-2026-1811-RC1
RC2: 'Comment on egusphere-2026-1811', Anonymous Referee #2, 29 Jul 2026

Review of Ward and Raphael, “Multi-Temporal Influences of Large-Scale Atmospheric Patterns On Antarctic Coastal Polynyas”
This manuscript examines the relationships between Antarctic coastal polynya area and three major modes of climate variability (Southern Annular Mode, El Niño Southern Oscillation, and the Amundsen Sea Low). The authors use a data set of 25 coastal polynyas from 1992 to 2017 and separate the atmospheric indices into annual, seasonal, and monthly components. The manuscript addresses a useful scientific question and uses a valuable circumpolar data set. However, the current analysis remains largely based on statistical correlations between climate indices and polynya area. The physical processes that connect atmospheric circulation, local winds, sea ice advection, and polynya opening are not examined directly. In its present form, the manuscript does not yet provide the process based understanding expected for publication in The Cryosphere.
I therefore recommend resubmit after major revision. A revised manuscript would need a major redesign of both the statistical analysis, figure design and the physical diagnosis.
Major comments
1. The manuscript analyzes the relationships of polynya area with the SAM, ENSO, and several ASL indices in separate regression models, although their effects on Antarctic climate are closely connected. Much of the influence of SAM and ENSO over West Antarctica is expressed through changes in the strength and position of the ASL. The authors mentioned this point. However, separate regressions cannot show whether an apparent ASL relationship is independent of SAM or ENSO. Partial correlation and controlled composite analyses would provide more direct tests. The ASL analysis could be strengthened through composite analysis based on different combinations of the three indices. The authors can compare ASL anomaly years with neutral SAM and ENSO conditions against years in which the ASL anomaly co-occurs with either SAM or ENSO, as well as years when all three indices show a physically consistent phase combination. Composite differences in polynya area, local winds, and sea ice motion would help determine whether the observed response is mainly associated with the ASL itself or with the broader circulation background linked to SAM and ENSO.
2. The separation of atmospheric variability into annual, seasonal, and monthly components is useful. The authors should nevertheless test whether the main results depend on the selected decomposition method. Ensemble empirical mode decomposition (EEMD) could be considered as an additional sensitivity analysis. EEMD (Wu and Huang, 2009) provides a data adaptive way to separate nonlinear and nonstationary variability into components with different characteristic timescales, without prescribing fixed frequency bands. It has been used to identify nonlinear climate trends and time varying seasonal or interannual variability in climate records Cornes et al., 2017; Wang and Ren, 2020).
Wu, Z., and Huang, N. E. (2009). Ensemble empirical mode decomposition. A noise assisted data analysis method. Advances in Adaptive Data Analysis, 1(1), 1–41.
Cornes, R. C., Jones, P. D., and Qian, C. (2017). Twentieth-century trends in the annual cycle of temperature across the Northern Hemisphere. Journal of Climate, 30(15), 5755–5773.
Wang, R., and Ren, H. L. (2020). Understanding key roles of two ENSO modes in spatiotemporal diversity of ENSO. Journal of Climate, 33(15), 6453–6469
3. The proposed physical mechanism should be tested more directly. The manuscript argues that large scale circulation changes local winds, which then alter sea ice export and polynya area. At present, this pathway is mainly inferred from earlier studies. The mean sea ice motion shown in Figure 2 does not demonstrate how the ice motion responds to SAM, ENSO, or ASL anomalies. The revised analysis should include local surface wind, sea ice drift, and sea ice divergence. Composite maps can show how pressure, wind, ice motion, and polynya area change during the main atmospheric states. These analyses would be especially valuable for East Antarctica, the Ross Sea, and the Amundsen Bellingshausen sector.
4. The distinction between ACP and RCP needs to be revised. The authors appear to use Eq. 2 as a conceptual statement that ACP reflects both the local intensity of the ASL and the influence of broader climate variability. This interpretation is understandable, but the equation is not valid. RCP is defined as the difference between the ASL central pressure and the mean pressure over the ASL sector. It therefore removes the regional pressure background, not the effects of SAM and ENSO themselves. SAM and ENSO are also climate indices rather than pressure terms, so they cannot be added directly to ACP and RCP. Therefore, Equation 2 is fundamentally invalid and should be removed. The suggestion that some ACP polynya relationships arise from the influence of SAM or ENSO may be reasonable, but it is not demonstrated by the current analysis and should either be supported by additional evidence or stated more cautiously.
5. The sign of the Southern Oscillation Index also needs to be checked throughout the manuscript. Lines 190-191 state that positive SOI anomalies represent El Niño and negative anomalies represent La Niña. In contrast, lines 314-315 describe positive SOI anomalies as La Niña conditions. This inconsistency affects the interpretation of all ENSO results. The authors should verify the index, the signs in the figures, and the related physical explanations.
General comments
Line 7. “Antarctica coastline” should be revised to “Antarctic coastline.”
Line 8. “areas of open water, minimal sea ice concentration, or relatively thin ice - called coastal polynyas” is awkward. Please revise the “-” and wording, for example, “…., or relatively thin ice, known as coastal polynyas.”
Lines 13–15. “SAM, ENSO, and the ASL together significantly influence” is misleading because the three factors are not analyzed together in one model. It would be clearer to state that 20 polynyas show a significant relationship with at least one of the examined indices.
Lines 20–22. The final two sentences of the abstract are difficult to follow because several indices and timescales are listed in one sentence. Please divide them into shorter statements and make clear which ASL metric corresponds to each timescale.
Lines 46–48. “Thermohaline Circulation” and “Meridional Overturning Circulation” should not be presented as fully interchangeable terms. Please use a more precise description of the role of Antarctic Bottom Water in the global overturning circulation.
Lines 59–61. The percentages cited here refer to sea ice production. Please revise “generate 5–10% of Weddell Sea ice” to “account for 5–10% of sea ice production in the Weddell Sea,” and make the same change for the Ross Sea and Antarctic total.
Line 67. “Appeared concurrent with” should be replaced by “coincided with”.
Lines 77–78. “As discussed below, our study found strong influences from the SAM and ENSO” is stronger than the statistical evidence supports. “are associated with” would be more appropriate.
Line 132. The definition of “30% daily occurrence” needs clarification. Whether this percentage is calculated over all April–October days in the full record or separately for each year?
Lines 143–146 and 174–176. The period for sea ice motion is inconsistent. The Figure 2 caption states 1992–2017, while the text states 1992–2016. Please check.
Line 146. “The color” should be “The colors.”
Lines 191–194. “Four indices, which together completely characterize the ASL” is too strong. These indices describe key aspects of the ASL, but I don’t think they can fully characterize its spatial structure.
Lines 242–249. “partial contribution” should be used with caution because the contribution is later derived from t values rather than from a formal partial variance measure.
Figures 4–7 and 9–10. The polynyas are shown only as numbers from 1 to 25. This makes regional patterns difficult to identify. I suggest the authors group the polynyas by Antarctic sector or add short region labels below the horizontal axis. In addition, the bars show both positive and negative values, but R² cannot be negative, the plotted quantity and the meaning of the sign need to be stated correctly in both the figures and caption.

Citation: https://doi.org/10.5194/egusphere-2026-1811-RC2

Jason Ward and Marilyn N. Raphael

Viewed

Total article views: 95 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
64	22	9	95	7	8

HTML: 64
PDF: 22
XML: 9
Total: 95
BibTeX: 7
EndNote: 8

Views and downloads (calculated since 04 Jun 2026)

Month	HTML	PDF	XML	Total
Jun 2026	21	7	3	31
Jul 2026	36	12	6	54
Aug 2026	7	3	0	10

Cumulative views and downloads (calculated since 04 Jun 2026)

Month	HTML	PDF	XML	Total
Jun 2026	21	7	3	31
Jul 2026	36	12	6	54
Aug 2026	7	3	0	10

Viewed (geographical distribution)

Total article views: 83 (including HTML, PDF, and XML) Thereof 83 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 09 Aug 2026

Short summary

Large-scale atmospheric systems significantly impact Antarctic coastal polynya size at various timescales. East Antarctica polynyas are influenced by the Southern Annular Mode monthly and seasonally. West Antarctica polynyas are influenced by the El Niño-Southern Oscillation interannually and the Amundsen Sea Low monthly and seasonally. This study uses a multi-scale perspective to identify and contextualize strong, large atmosphere-polynya relationships hidden in previous studies.


Total:	0
HTML:	0
PDF:	0
XML:	0