the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Feb 2025
Large scale climate response of the Southern Ocean and Antarctica to reduced ice sheets
Abstract. The Antarctic and Greenland Ice Sheets (AIS and GIS) are critical tipping points in the Earth's climate system. Their potential collapse could trigger cascading effects, significantly alter the global climate patterns, and cause large-scale, long-lasting, and potentially irreversible changes within human timescales. This study investigates the large-scale climate response of the Southern Ocean and Antarctica to the isolated effect of reducing ice sheet extent. Using well-documented paleogeography of the Late Pliocene (LP) from the PRISM4D reconstruction, where the West Antarctic Ice Sheet (WAIS) and the northwestern GIS were significantly diminished, we conducted 1450-year simulations with the EC-Earth3 model at atmospheric CO2 concentrations of 280 ppmv and 400 ppmv.
Our results reveal that the implementation of the LP ice sheet configuration leads to a 9.5 °C rise in surface air temperature, approximately 16 % reduction in sea ice concentration, and a 0.63 mm/day increase in precipitation over Antarctica and the Southern Ocean. These changes far exceed those driven by CO2 increase alone, which result in a 2.5 °C warming and a 9.3 % sea ice decline. Throughout the simulations, the positive phase of the Southern Annular Mode (SAM) persists, intensifying westerly winds and contributing to sea ice export and deep ocean warming. The combined effects of ice sheet reduction and CO2 forcing initially weakened Antarctic Bottom Water formation, with major implications for the Global Meridional Overturning Circulation due to stronger water column stratification driven by surface freshening and subsurface warming.
By isolating the direct albedo effect of reducing GIS and AIS extents, this study offers critical insights into the mechanisms driving atmospheric and oceanic variability around Antarctica and their broader implications for global climate dynamics. Although the complete Late Pliocene boundary conditions may serve as a valuable analogue for current and future climate change, excluding the orographic changes allows us to specifically assess the primary role of surface reflectivity. This targeted approach lays the groundwork for future research to explore how the combined effects of albedo and paleogeographical changes could influence interhemispheric climate feedbacks under scenarios of future ice sheet collapse or instability.
- Preprint
(4644 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Review of egusphere-2024-4061', Anonymous Referee #1, 14 Mar 2025
Power et al. assess the high southern latitude climatic impacts of changes in albedo resulting from a smaller (Late Pliocene) Antarctic ice-sheet. This is an interesting study, relevant for past climates and potential future. Some revisions are however necessary before publication.
- It should be made clearer in the title and first part of the abstract that only albedo changes and not orographic changes are taken into account.
- Improved methods: It might be helpful to include a table of the experiments performed with the forcing included. I am actually surprised to see changes in albedo over the East Antarctic ice-sheet in fig. 2d. It is my understanding the ice-sheet is a forcing in this simulation, and thus the albedo over the ice-sheet imposed. Can the authors explain why albedo is changing over land areas where the ice-sheet forcing should not have changed?
- Improved analysis:
106: the authors state that “when ice-sheet reductions are included there is a persistent positive phase of the SAM, which intensify the westerly winds”
On Figure 4, timeseries of SAM indices are shown for all the experiments (note that the x axis is missing), and in all cases the SAM index is much lower in the Ei than E case. So the SAM index is less positive when the ice-sheet reductions are taken into account contrarily to the statement above. I note that even in the PI case (E280), the SAM index is centered on 0.2, which seems odd. The authors should double check their results and include a description of their SAM index calculation.
114-115: the authors state that there is a “deepening of the Amundsen sea low with stronger winds and stormier conditions”, but this is not shown and it sounds to me more like a simple description of what is expected under positive SAM than in the simulations presented. If you want to keep that statement, then please show the changes in SLP or geopg in a more convincing way as this is really not evident from Fig. 3 c,d. In addition, the two statement above are contrary to the statement L.163-164: “our simulations do not indicate significant changes in the wind regime over Antarctic and the Southern Ocean.” These contradictory statements and figures suggest that the authors need to carefully look at their results and assess whether there are significant changes in the winds or not. Please show the wind changes.
118: The authors suggest that the poleward contraction of the westerlies amplifies the upwelling, which accelerates sea-ice melting from below. Related to the comment above, are there significant changes in the westerlies in your simulations? Please show the wind and upwelling changes.
Figure 5 does not seem appropriate to discuss AABW changes. I) the AABW in E280 should be shown. Ii) the AABW taken through this calculation seems really low and does not seem to reflect the lower limb of the MOC as seen on Figure 7. Is your value of the global stream function south of 60S and below 500m even negative (ie anticlockwise circulation)? In fact, on your figure 7, the lower limb of the MOC does seem weaker in Ei400 than E400. Coarse resolution models do have issues representing AABW formation, and on Figure 7, there does not seem to be any downwelling branch south of 60S (ie anti-clockwise circulation south of 60S). Please re-assess your statements L. 153-156.
170-174: I am confused about this paragraph. Is this a statement based on what has been shown in previous studies or based on the authors’ results? If the former, then references are critically needed. If the latter, then this should be shown as it does not seem to be what is obtained here (ie SAM is less positive when the albedo decreases over Antarctica), and references to the figures where it is shown should be included.
4. An improved discussion of the results in comparison to previous studies is needed.
For example, previous studies looking into the mid-Pliocene (Weiffenbach et al., 2023, Climate Past, https://cp.copernicus.org/articles/20/1067/2024) and Last Interglacial (Yeung et al., 2024, Com. Earth & Env., https://www.nature.com/articles/s43247-024-01383-x) highlighted the impact of Southern Ocean warming and reduced sea-ice on stratification and AABW formation, with potential feedbacks on the AIS.
The impact of AIS loss on climate was also studied in Hutchinson et al., 2024 (Nat. Com., https://www.nature.com/articles/s41467-024-45501-x). While in that study, changes in orography were also taken into account, it seems relevant to include this study in the discussion as the climatic impact seems much smaller in Hutchinson et al than in your study.
5. Line by line comments:
60, 61: Please include the references to the mid-holocene, Last interglacial and mid-Pliocene simulations.
124: If I understand correctly, there is no ice-sheet model. Marine ice sheet instabilities parametrization is only relevant for ice-sheet models.
Legend of figure 5: Please include the area or latitudinal band of the Southern Ocean that was used for this graph.
131: the reason for the interannual to decadal variability is not shown here, so this seems like a speculation and it does not seem relevant to the study.
163-164: Please re-assess if this correct after looking into the wind changes (see comment 3 above).
203-208: this paragraph might need to be a bit modified. Even though there are some uncertainties associated with Antarctic topography after the loss of WAIS, it seems that it would be more consistent to do so. I have no problem with the study only looking at the impact of changes in albedo, but the authors have to acknowledge that assessing the impact changes in orography is also important.
Citation: https://doi.org/10.5194/egusphere-2024-4061-RC1 - AC1: 'Reply on RC1', Katherine Power, 10 Apr 2025
-
RC2: 'Comment on egusphere-2024-4061', Anonymous Referee #2, 18 Mar 2025
Review of Power et al “Large scale climate response to the Southern Ocean and Antarctica to reduced ice sheets”
In this paper, the authors use the EC-Earth climate model to test the impacts of reduced albedo under Late Pliocene ice sheet configurations to the climate system. Experiments also include the impacts of elevated greenhouse gas levels. While the topic is of great interest, I have several big-picture concerns, specifically about the internal consistency of the experiments presented here.
First, it is unclear what boundary conditions the authors are using for these experiments. In Section 2.2 (Experiment setup), the authors state they are using the PlioMIP 4F Tier 2 experiments (Line 78). PlioMIP uses Pliocene boundary conditions of the ice sheets, including modified ice sheet extent and orography and other changes associated with a reduced ice sheet and warm period, such as changes to vegetation and GHGs etc. However, the previous paragraph (L75) the authors state that they exclude orographic changes and freshwater input associated with the ice sheet extent. If indeed orographic and freshwater changes are not included in these experiments, it remains unclear to me how these changes were implemented in the model. What albedo was chosen in places where West Antarctica or Greenland are longer “ice”? How are these grid cells treated, presumably as land? What albedo is used in these locations? More clarification and explanation is required. Thereafter throughout the manuscript, the authors often refer to using a Pliocene ice sheet configuration, and that such a configuration serves as an analogue of future configurations. The term ‘configuration’ implies that all relevant elements of that configuration have changed, including orography (e.g. L196). Better explanation and justification is needed on experiment design, and care must be taken when using terms like ‘configuration’ and ‘reduced ice sheet extent’ throughout, especially if this is not what the experiment is actually doing.
If I understand correctly, only albedo is being modified in these experiments. I am struggling with the rationale and relevance for such an experiment. Keeping a modern AIS/GrIS topography, but reducing albedo to match Pliocene ice extent results in inconsistent boundary conditions. You cannot collapse WAIS as in Figure 1 but maintain modern topography. If WAIS is ice-free this implies it is collapsed. Therefore, not only is the orography fundamentally different, but also ocean pathways between the Ross, Amundsen and Weddell Seas open up, leading to different ocean circulation, sea ice regimes and atmospheric circulation (e.g. Steig et al (2015), Pauling et al (2023)). Therefore, much of the analyses presented here (changes in sea ice regime, SAM index, albedo, surface air temperatures and precipitation) is limited by the presence of land above sea level in West Antarctica, and drawing conclusions about those elements of the climate system would be inaccurate.
If the goal of the study is to ask how the Southern Ocean and AIS climate responds to reduced ice sheets (as in the manuscript title), this experimental design cannot answer that question. Instead, it seems to me that the authors are asking what the impact would be if WAIS was suddenly painted black (or whatever low ocean/land albedo is being used here). If that is the question, please justify. To isolate the effect of albedo a better tool might be a simple global energy balance model. This would also resolve the authors’ technical issues with modifying the atmospheric component of the model when modifying orography (though I will note this is technically possible and not uncommon (see PlioMIP experiments)). Another option would be to use the current experiments, but rather than comparing to a PI control run, also compare this simulation to a run where the full configuration of the ice sheets is changed (including orography). This would allow the authors to better detangle the role of albedo-only changes, by comparing to cases with and without orography change.
Once these big picture issues are taken care of, there are a number of other issues throughout the paper that I recommend addressing. For example:
- As mentioned above, the title does not accurately represent this work – you are not actually reducing ice sheet extents in these experiments
- The abstract needs work to make connections clearer and providing appropriate motivation for the experiment.
- As mentioned above, the experiment design (Section 2) should include a more comprehensive description of the experiments and implementation.
- Figures need to be better labeled. For example, Figure 1 should have labels, and the caption should explain more. Consider also including a cross section of the orography for the two ice sheets.
- Figure 2 it is unclear why albedo is spatially variable in the Ei experiment. This is confusing.
- There are some claims throughout, such as the stability of ice sheets being increasingly at risk, please provide appropriate citations for these.
- Usually Greenland Ice Sheet is abbreviated as GrIS rather than GIS.
- Include references when discussing EC-Earth’s proven capabilities.
- The discussion requires more work, more context, and some of the concluding claims are not well-supported. For example, it is not clear how this experiment will help identify early signals of ice sheet retreat and possibly ice sheet instabilities because this study does not include a dynamical ice sheet.
References
Steig, Eric J., Kathleen Huybers, Hansi A. Singh, Nathan J. Steiger, Qinghua Ding, Dargan MW Frierson, Trevor Popp, and James WC White. "Influence of West Antarctic ice sheet collapse on Antarctic surface climate." Geophysical Research Letters 42, no. 12 (2015): 4862-4868.
Pauling, Andrew G., Cecilia M. Bitz, and Eric J. Steig. "Linearity of the climate system response to raising and lowering West Antarctic and coastal Antarctic topography." Journal of Climate 36, no. 18 (2023): 6195-6212.
Citation: https://doi.org/10.5194/egusphere-2024-4061-RC2 - AC2: 'Reply on RC2', Katherine Power, 10 Apr 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 165 | 47 | 8 | 220 | 9 | 7 |
- HTML: 165
- PDF: 47
- XML: 8
- Total: 220
- BibTeX: 9
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 93 | 37 |
| Germany | 2 | 33 | 13 |
| China | 3 | 16 | 6 |
| United Kingdom | 4 | 16 | 6 |
| South Korea | 5 | 12 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 93