the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 09 Jul 2026
Impacts of record Antarctic sea ice losses in 2022–2023 on swell-induced flexure of ice shelves
Abstract. Recent case studies have highlighted regional loss of Antarctic sea ice triggering large-scale ice shelf calving events, by allowing damaging ocean swell to reach shelf fronts. Thus, the dramatic and widespread loss of Antarctic sea ice in recent years has implications for the stability of ice shelves. Here, observations and models are used to conduct a broad assessment of changes in swell-induced flexural stress levels experienced by Antarctic ice shelves, with a focus on the record sea-ice lows of 2022 and 2023. Daily time series are constructed for effective lengths of the sea-ice barriers protecting fourteen Antarctic ice shelves, and for the incoming swell, along with yearly values for shelf front thickness, over a decade leading up to and including 2022–2023. The flexural stress levels are generally found to be far greater in 2022 and 2023 than the preceding eight years, and this is shown to be primarily driven by sea-ice loss, although sometimes mitigated by reduced peak periods of incoming swell. Further, it is shown that flexural stress anomalies are strongly correlated with anomalies in effective sea-ice length and peak period, and that an increase or decrease in flexural stress can be predicted by relative changes in sea-ice length and peak period. The findings indicate that more Antarctic ice shelves will become susceptible to enhanced swell-induced flexural stress if sea-ice losses and ice shelf thinning continue. Under these scenarios, predictions are made of when flexural stress will become a dominant contributor to total ice shelf stress for each of the fourteen ice shelves studied.
- Preprint
(3404 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-3544', Anonymous Referee #1, 30 Jul 2026
-
RC2: 'Comment on egusphere-2026-3544', Anonymous Referee #1, 30 Jul 2026
Note, As a referee, I clicked the box that said I was willing to review a revised manuscript. However, the actual poof of the review always indicated that I was *not* willing to review the revised manuscript. This is an error in the review software that needs to be fixed.
Citation: https://doi.org/10.5194/egusphere-2026-3544-RC2 -
RC3: 'Comment on egusphere-2026-3544', Bertie Miles, 26 Aug 2026
This manuscript reports on the impact of the 2022/2023 sea ice lows on swell-induced flexure of a selection of ice shelves around Antarctica. The authors conclude that flexure has, for the most part, increased, but that at present most ice shelves are of sufficient thickness to avoid large-scale calving events. However, they show that based on hypothetical thinning rates, several ice shelves could become vulnerable towards the end of the century. The manuscript tackles an important and arguably understudied topic given the sea ice decline observed over the past decade (and expected future decline), and the results are novel. I believe it is suitable for publication in The Cryosphere, subject to addressing the comments below.
General comments
- I think some of the language and framing used to describe the increases in stress in 2022 and 2023 relative to the baseline does not match the results, e.g. 'far greater', 'substantial', 'major'. While this is true in relative percentage terms, most of the ice shelves appear to have experienced very little change in absolute terms. For example, is a 300% increase at Pine Island really meaningful when it represents an increase from 0 to 0.003 kPa? I think in 12 of the 14 ice shelves the increases in stress are less than 0.5 kPa. The manuscript suggests that 10 kPa is an important threshold for triggering greater calving, so in most cases could it be argued that the changes have been minimal in absolute terms?
- How good is the wave model, and can we actually trust it? Some of the ice shelves included are quite small, with very complex coastlines. The resolution of the wave model is 0.4°: can this really do a good job for the smaller and more complex ice shelves (e.g. Sulzberger, which is fringed by islands, ice rises and other ice shelves)? Likewise, how confident are you in the way it treats sea ice? The 25% concentration threshold seems a little arbitrary. I appreciate that there may be no alternative, but I think some qualitative discussion is needed. Are there any field observations that validate it? Presumably there is a non-negligible chance that the true values are some way off.
Line comments
- Line 8: I am not sure flexural stresses are far greater in absolute terms.
- Figure 1: I think Nickerson and Sulzberger are labelled the wrong way around.
- Line 66: Arguably the most significant changes at any ice shelves over this time period have been at Pine Island and Thwaites, so it reads a little strangely when less significant changes at Larsen C and Sulzberger are discussed but Pine Island and Thwaites are not.
- Line 80: Presumably you did this manually for Thwaites, given the mélange and icebergs in front of it? In many ways it is a strange choice for inclusion among the 14 ice shelves, because it is hardly an ice shelf any more.
- Line 100: It is perhaps worth noting that in complex ice shelves with multiple outlets of differing thickness (e.g. Sulzberger, Shackleton) this might bias the results towards outlets with lower discharge.
- Line 176: What is the justification for using E = 9 GPa?
- Figures 7, 8 and 9: I find it a little difficult to pick out the 2014–2021 boxes because they are so thin. Please make them the same size as the 2022 and 2023 boxes, and perhaps use a different colour.
- Line 215: This compares ice shelf thickness change with ice shelf frontal thickness change, and is therefore not like for like. As an ice shelf front advances, it will typically become thinner; in contrast, most of the thickness change usually occurs near the grounding line.
- Line 292: A reader picking out this line may think there has been a dramatic and meaningful increase in stress at Pine Island and Thwaites, when in fact the changes are negligible.
- Figure 11 caption: 20921 → 2021.
- Line 324: Consider rewriting: arguably there have not been large increases in flexural stress at most ice shelves in absolute terms.
- Line 338: Presumably Denman — which is responsible for the vast majority of discharge from Shackleton — is substantially thicker than the surrounding ice shelf and thus not as vulnerable? see comment reference line 100
- Line 355: 'much greater': arguably they were not, in relative terms.
- Line 372: 'major amplification' — again, consider rephrasing?
- Line 375: If they thin by around 20–30% of present thickness (i.e. thinning at 1% per year)?
Citation: https://doi.org/10.5194/egusphere-2026-3544-RC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 210 | 108 | 29 | 347 | 17 | 17 |
- HTML: 210
- PDF: 108
- XML: 29
- Total: 347
- BibTeX: 17
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments:
This is easily the most compelling, sophisticated and convincing manuscript I have encountered on the topic of ice-shelf/sea-swell interaction. The manuscript uses sophisticated, creative analytical methods and employs the state-of-the-art data sets. The manuscript is well written and well proof read.
I have no significant questions or challenges to bring up in my review. I regard the manuscript as being very valuable and will be likely read, cited and used as inspiration going forward.
specific comments:
line 34, page 2 “sea ice, in the form of both stationary consolidated face ice and moving pack ice, creates a barrier that” should face be “fast”?
page 2, somewhere in the introduction, it might be worth pointing out that there have been some studies supporting speculation that periodic ice-shelf features, like the “rolls” found on the Ward Hunt Ice Shelf in the Arctic, have an effect that prevents transmission of sea swell with certain frequencies into the wider ice-shelf region. There has been speculation that the only remaining elements of the Ellesmere Ice Shelf witnessed by European explorers in the mid 1800’s are the ones that have these rolls and are thus more robustly protected from sea swell associated with changing Arctic sea ice since the Little Ice Age and into the present day.
A reference to the above: Nekrasov P, MacAyeal DR. Ocean wave blocking by periodic surface rolls fortifies Arctic ice shelves. Journal of Glaciology. 2023;69(278):1740-1750. doi:10.1017/jog.2023.58
page 5 line 85. I’m not sure I understand what a “percentile” variable is… is it a single scalar number or is it a set or population of grid points satisfying some criterion. Perhaps an explanation is not needed due to my not being facile with this, but maybe a comment would help.
Actually many of the variables being described in the well written, but terse algorithm sections are subject to potential misinterpretation by people trying to follow in the footsteps of this study. Perhaps a sentence or two (could be in supplemental material) on what constitutes each variable (is it a set, a population, a scalar, a vector, a parameter?) would help.
Figure 2: it is worth pointing out that some attenuation by the arrangement of ice out in front of the ice shelf front may be also due to Bragg scattering, a la the same process as “rolls” on the Ellesmere Ice Shelf.
Page 11 - It may be worth indicating why thin ice shelves can get >100 kPa vs. thicker ice shelves being more like 10 kPa. Why does flexure stress increase with decreasing thickness… just a basic principle or idea can be mentioned for the curiosity of the reader who may not have experience with this.
page 20 - sometimes p-values and sometimes rho-values appears. Are these different?
Conclusion - It might be worth mentioning that seismological methods might be able to assess damage that is associated with increasing flexural fatigue of the various ice shelves, and give a simple citation that interested readers could follow. It may be that seismological monitoring of ice shelves could assess the damage of increasing sea swell effects in the Antarctic ice shelves.