the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 10 Oct 2025
Dynamic Thinning and Grounding Line Retreat in Porpoise Bay, Wilkes Land, East Antarctica
Abstract. The East Antarctic Ice Sheet (EAIS) is often considered less vulnerable to climate change than the West Antarctic or Greenland ice sheets, but some regions of the EAIS have been losing mass over recent decades. In particular, mass loss in Wilkes Land, which overlies the Aurora Subglacial Basin, is thought to have accelerated over the past two decades. However, whilst several large outlet glaciers drain this region, few have been studied in detail. Here, we present new data on the recent ice dynamics of four outlet glaciers that drain into Porpoise Bay, Wilkes Land, which includes Holmes East, Holmes West, and Frost glaciers. We use optical satellite imagery, differential synthetic aperture radar interferometry, and a range of previously published datasets to describe changes in the ice-shelf front, grounding line position, ice surface velocity and ice surface elevation over the last three decades. Our results reveal evidence of dynamic changes in the region, characterised by thinning of grounded ice and grounding line retreat, albeit with large uncertainties. We find an indication of Circumpolar Deep Water proximal to the continental shelf break that could access the glaciers through deep cross-shelf troughs, which is consistent with previous estimates of high rates of basal melting beneath their floating tongues/ice shelves. Our results also support previous observations of near-synchronous ice-shelf calving across Porpoise Bay's ice shelves, following the break-out of multi-year sea ice, and find an additional recent calving event, further highlighting the vulnerability of this region to ongoing and future changes in ocean and sea-ice conditions.
- Preprint
(2969 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 06 Jan 2026)
- RC1: 'Comment on egusphere-2025-4100', Anonymous Referee #1, 18 Nov 2025 reply
-
RC2: 'Comment on egusphere-2025-4100', Anonymous Referee #2, 25 Nov 2025
reply
General comments:
This manuscript presents an ambitious ensemble of datasets in an attempt to weave together a complex story of glacier dynamic change, as potentially driven by mCDW intrusion and sea ice breakup events, in the Porpoise Bay region of East Antarctica. The authors combine time series observations of ice front position, ice velocity, grounding line position, and bed, glacier, and ice shelf elevations with sea ice observations and ocean temperature/salinity observations/objective analysis. The authors are combining sometimes disparate datasets into decades long time series for each process, which they then compare in order to compile evidence for causality between climate drivers and components of dynamic change.
The paper carries two central narratives. The authors argue that the conditions exist for mCDW transport to Porpoise Bay grounding zones, and that best estimates of grounding line retreat suggest it could be occurring. They also separately argue on the basis of timing that sea ice breakup causes calving events, and that calving can precede upstream dynamic changes. Overall, this manuscript has the potential to be good, but as it stands, there are a lot of moving parts that would benefit from careful reorganization and a touch more rigor. Two elements of the paper that needs attention are (detailed in specific comments):
(1) the uncertainty quantification. Currently, this is limited to a brief discussion in the methods section and frequent references to uncertainty without quantitative support.
(2) the narrative organization and logical flow. This needs attention through reorganization and the addition of a supplement to capture supporting ideas and streamline the central narrative.
With the proposed changes, this paper could meaningfully contribute to a growing body of literature concerning the growing importance of the EAIS to Antarctica’s mass balance and sea level contribution.
Specific comments (in no particular order):
This text combines numerous datasets, which occasionally express uncertainty ranges upwards of an order of magnitude. In order to make it more clear which portions of the timeline are contributing most to uncertainty, I recommend that error bars be placed in all applicable figures (3, 4, and 8) and reported alongside key estimates in the results. Also, I recommend a scan for any mention of “high uncertainty” (I’ve identified a few) and try to replace with something more quantitative (e.g., percent error). These uncertainties should be addressed when qualifying the hypothesized processes. For example, do upper bounds in uncertain bed topography drastically alter the likelihood of mCDW intrusion?
In its current state, there are two competing narratives in this work: (1) dynamic thinning and grounding line retreat from mCDW intrusions, and (2) the interactions between sea ice, calving, and upstream glacier flow. I would suggest taking steps to centralize the narrative, such that these are two functional branches in the broader scheme of glacier dynamic change. This might look like more emphatic statements to link the two processes and some reorganization around those two central threads. Alternatively, suppress or remove (2), since it already seems to be secondary (it does not appear in the title) and is difficult to support without some modelling effort (see my comment on line 663).
In line with the last suggestion, some figures don’t seem to be essential for the narrative, and thus might be better off in a supplement (e.g., figures 10, 12, 14, possibly 7).
The argument for mCDW presence relies on two (necessarily) incomplete pieces: EN4 and tagged seal data. EN4 is an objective analysis product, and given the data limitations in this region, some discussion (possibly a supplemental figure) depicting the actual available data points used in this product would provide an honest anchor for the veracity of water temperatures in Porpoise Bay. Additionally, a quick supplemental comparison between the two datasets (such as one plotted on top of the other) would strengthen them both.
Factors which affect ice elevation changes over time (namely precipitation and firn densification) are never discussed. This should be central to introducing the elevation changes as a metric, so that the reader understands where the ambiguities lie.
Even more specific comments/questions:
20: consider dropping the last clause, and adding a concluding sentence that widens the scope back to the broader community contribution.
40-60: Might be worth citing Walker et al., 2024, which captured the first ice shelf collapse in East Antarctica, to highlight that changes are starting to take place.
41: This paragraph could be split in two, talking about EAIS, then about Wilkes Land
75: Thompson et al., 2024 is another key paper about Wilkes Land dynamics.
110: “identification criteria” would be a bit more clear
129: To which satellites are these errors attributed? Here is an order of magnitude error range that makes the good data look bad.
141: Maybe specify that it comes from the ATL15 gridded product that is corrected with firn-models and tides
144: Not clear what you mean by compare the datasets
162: Possible to use a smaller cutoff? 50 percent is quite large, that’s 200 m/yr error for some of your upstream velocities. At least warrants an explanation.
Fig. 3: Dashed lines aren’t necessary because you have markers too. Also, could consider dropping the 1960-2000 to make the era of interest more clear. You could change the axis label to “change from 1960” and possible include vertical lines to show the 1960-2000 min and max since that is the primary benefit of those points in the first place.
Fig. 4: I would recommend aligning all vertical axes, even if the lines get squished a bit. Will help the reader understand the relative significance of each glacier.
170: Maybe state where grounding lines were found as well as where they weren’t. This confused me a bit.
192: Perhaps MOA would be more clear than manual, since you have DInSAR labelled.
3.1.: “Ice shelf front position.” This terminology in particular varied a bit throughout.
230: Might want to explain how EN4 is constructed
233: This sentence is superfluous. Also consider sometimes replacing phrasing of “high uncertainty” with “best estimates”
415: recommend replacing “around” with ~
452: “elevation thinning”
455: “faster (<500 m/yr)” consistent formatting
456: A grounding zone paper to cite: Zhu et al (2025)
Table 3: Maybe add date ranges to the table itself
3.6: This section could use another look to make sure you are highlighting the results you want to highlight. E.g., sea ice is not mentioned here. Maybe that was on purpose.. But I read this section as if it were the intro to your discussion, for what that’s worth.
474: “face value” said a few times, it has the effect of downplaying your estimates.
482: another “highly uncertain”
511: does EN4 depict the temperature and salinity that can be used to identify CDW?
653: another “high uncertainty”
663: It could be argued that wind is the force driving both events, and that the sea ice gives in first, then calving occurs, creating the illusion of causality. Recent modelling work by Surawy-Stepney et al, (2024) explores this on the Antarctic Peninsula.
Technical comments:
There is a shift between active and passive voice (lots of active voice in the discussion).
Citation: https://doi.org/10.5194/egusphere-2025-4100-RC2 -
RC3: 'Comment on egusphere-2025-4100', Anonymous Referee #3, 29 Nov 2025
reply
Please see the attached review.
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 322 | 105 | 23 | 450 | 12 | 15 |
- HTML: 322
- PDF: 105
- XML: 23
- Total: 450
- BibTeX: 12
- EndNote: 15
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Review of “Dynamic Thinning and Grounding line Retreat in Porpoise Bay, Wilkes Land, East Antarctica” by Weatherley et al.,
This study focuses on investigating how ice shelf fronts on Porpoise Bay (Holmes East/West, Frost and the creatively named Glacier 1) have evolved since the late 1990s (or the early ‘60s for ice shelf front) by utilizing a mix of satellite imagery, radar interferometry alongside previously published data on the state of the ice shelf front. This shelf evolution was compared to temperature/salinity data taken in Porpoise Bay from MEOP casts and EN4 reanalysis data to provide evidence of any continental deep water intrusions onto the shelf front, which could support heightened basal melt. The authors also find a previously undocumented calving event on the Holmes East glacier and link the breakup of the protected mélange layer to a loss of local sea ice in the months leading up to calving.
Overall, I found this a well written and interesting manuscript on what is both an underrepresented region in literature, and a difficult region to analyze owing to the aforementioned presence of mélange. With a bit of tightening, it should be a very useful paper for anyone looking to undertake research in the region. Also, low-key love the care put into colour schemes throughout.
Minor comments:
Line 20: following the break-out of multi-year sea ice alongside a recent calving event., further…
Lines 30—35: Seeing there have been a few big mass balance papers out recently, I’d be curious to see some comparison to Green et al., 2022 (doi: 10.1038/s41586-022-05037-w) and Davison et al. , 2023 (10.1126/sciadv.adi0186)
Lines 43-45: The start of “However, there is emerging evidence of a longer-term trend in Wilkes land seems at odds with the previous sentence that Wilkes land is the dominant source of mass loss in EAIS.
2.2 Ice surface elevation general comment 1: I found the repeated usage of “the data” or “this dataset” made this section hard to follow.
Ice surface elevation general comment 2: I wonder if it would flow better if the Nilsson et al., part is given its own paragraph, plus a tad more information on the additional data from ICESat-2 and how it meshed with the previous data.
Line 161: I’d like the error range given for each box instead of the range. Especially seeing that is a rather large difference in error.
Line 178: Uncertainty was derived from…
Line 246: Considering the mention of ice-shelf breakup, I’d also mention grounded icebergs if we’re going for potential “errors” in NSIDC sea ice index.
Line 255: Fluctuated from where (especially as you abandon this kind of measure in the next paragraph to go relative from calving)?
General result comment: isn’t slowed down by – number a double negative? I think the results would be improved by sticking to either a sign to indicate change, or text.
Line 286: There is an additional space after 382.3
Line 291 consistently increased by 7-10% over the study period would make it more concise. Same comment on Line 295.
Ice shelf elevation change: I found this section awkward in its layout, and I think it would benefit from figures 5 and 6 being combined (as much as they’re pretty figures). Alternatively, I wouldn’t be opposed to text, figure 5, figure 6 as the layout.
Line 334: The “we observed lower rates…” in a paragraph that continuously reminds the reader that this data is from other sources seems out of place (not that I’m against the we observed, it just came off as jarring).
Line 365: I’m not a fan of “taken at face value”. It’s a given with results.
Figure 8 and 9: Considering how much these figures interplay with each other, could they be combined?
Line 435: This feeds into a comment I have in the discussion, but how does a February-March combination look for concentration anomaly? Seeing they are typically the two lowest months for sea ice extent.
Figure 13: Can you make each tick visible.
Figure 14: Can you align the ticks to their respective bar. Took a few seconds to get my bearings with this one.
Line 449: Holmes West glacier are consistent with dynamic …
Line 456: ice velocity is increasing across …
Line 476: I’m not sure “down a retrograde slope” is necessary.
Line 485: retreat could be less rapid than we record.
Lines 511-515: I think that would benefit being broken over two sentences.
Line 562: glaciers (Fig 8.) alongside the rapid decrease in surface elevation (Fig 6d)., since Holmes …
4.3 Role in sea ice driving ice shelf and glacier flow change general comment: I found the lack of a mention for the Voyeykov an odd decision here, owing to the similar mélange break up and proximity to Wilkes Land.
Line 571: I’d switch Baumhoer et al., 2021 for Teder et al., 2025 (doi: 10.1038/s41561-025-01713-4). Bit of a double whammy where Baumhoer doesn’t go into great detail about fast ice/melange, and Teder et al., does give evidence for a very large calving event being in response to mélange breakup (seeing the perennial fast ice in north Wilkins is a mélange layer).
Line 577: This does feedback to the comment earlier about being curious about the February concentration and combining the sea ice low period. Considering Teder et al., + Arthur et al., showed evidence that there was a persistently anomalous sea ice low in the couple of months leading up to the breakup of mélange, I’d be curious how well this breakup compared to those two regarding timing.
Line 590: With the bulldozing comment, its certainly possible seeing fast ice growth tends to be limited by depth around Antarctica (Fraser et al., 2023, doi: 10.1029/2022RG000770). That said, I’m not sure how feasible it would be to provide bathymetry of where this mélange was pushed too.
Lines 605—610, seeing proximity, how does this compare to what has been observed on the Totten ice shelf (Green et al., 2018, doi: 10.5194/tc-12-2869-2018)
Line 660: we report on a previously unidentified calving..
References: Personally I’m not a fan of the abbreviated journal titles, though if this is the cryosphere template, please disregard.
Admundson et al., becomes justified in its second line (and there are a couple of others in the reference list)
Favier, L. has F. Pattyn instead of Pattyn, F.
Haran et al., 2021 has a duplicate entry