the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Impact of Greenland Ice Sheet Disintegration on Atmosphere and Ocean Disentangled
Abstract. We analyze the impact of a disintegrated Greenland Ice Sheet (GrIS) on the global climate through steady-state simulations with the MPI-ESM (Max Planck Institute for Meteorology Earth System Model). This advances our understanding of the intricate feedbacks between the GrIS and the full climate system. Sensitivity experiments enable the quantification of the individual contributions of altered Greenland surface elevation and properties (e.g., land cover) to the atmospheric and oceanic climate response. Removing the GrIS results in reduced mechanical atmospheric blocking, warmer air temperatures over Greenland and thereby changes in the atmospheric circulation. The latter alters the wind stress on the ocean, which controls the ocean-mass transport through the Arctic Gateways. Without the GrIS, the upper Nordic Seas are fresher, attenuating deep-water formation. In the Labrador Sea, deep-water formation is weaker despite a higher upper-ocean salinity, as the inflow of dense overflow from the Denmark Strait is reduced. Our sensitivity experiments show that the atmospheric response is primarily driven by the lower surface elevation, whereas altered Greenland surface properties mostly amplify but also counteract few of the changes. The lower Greenland elevation dominates the ocean response through wind-stress changes. Only in the Labrador Sea, altered Greenland surface properties dominate the ocean response, as this region stores excessive heat from the Greenland warming. The main drivers vary vertically: The elevation effect controls upper-ocean densities, while surface properties are important for the intermediate and deep ocean. Despite the confinement of most responses to the Arctic, a disintegrated GrIS also influences remote climates. The altered climate in response to a GrIS disintegration also constrains a potential ice-sheet regrowth to high-bedrock eastern Greenland.
- Preprint
(11581 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on esd-2024-24', Anonymous Referee #1, 19 Sep 2024
This manuscript investigates the climatic impact of a disappearance of the Greenland ice sheet, mainly at high northern latitudes. Sensitivity experiments are performed to disentangle the drivers of these impacts: on one hand, the atmospheric response and the oceanic response and, on the other, the effects of the reduction in surface elevation and the change in surface properties.
I believe such a study should be interesting for a broad sector of the community. However I have major issues with the text that need reworking. These have to do mainly with the Results section. First, it is extremely lengthy and descriptive, which makes the reading a bit tiring, so shortening this section would improve the readability. I provide specific suggestions below. In this line, there are many sections that repeat information that was already given before. Second, some reorganization is also needed. For instance, changes in sea ice have a very large impact on surface temperatures but appear very late in the paper. Finally, and importantly, many results are mentioned and described but not shown, especially in relation with ocean changes which seems to be the focus of the paper, e.g. the AMOC. There is also a brief section focusing on remote impacts which are actually not shown.
Having said that, the text is well written and the figures are excellent. I therefore provide below a list of specific comments which I think need to be addressed before the manuscript can be accepted for publication. As you can see they are quite detailed at the beginning because I think the first sections read well but more general towards the end, where I think the problems I mentioned before are more severe.
Abstract, line 1: you do not really investigate the impact on the global climate but rather mainly focus on high northern latitudes.
l 10: “whereas altered Greenland surface properties mostly amplify but also counteract few of the changes”: this could be removed.
l 11-12: “Only in the Labrador Sea, altered Greenland surface properties dominate the ocean response”: I am not sure whether this is grammatically fully correct, please check.
l 14: “Despite the confinement of most responses to the Arctic, a disintegrated GrIS also influences remote climates”: this is not really investigated. See below.
l 22: “the interplay between GrIS characteristics and the broader climate system is imperative to understand”: I think this is not grammatically correct.
l 24-27: you provide a very long list of references here just saying “These studies found considerable climatic changes”. I suggest summarizing briefly the insight provided by those studies.
l 50: “Here, we extend those studies by examining the interactions of the GrIS with the entire climate system, including the deep ocean”: I do not think this is true, you mainly focus on the Arctic. You do comment on changes in the deep ocean and in the AMOC but you do not show the latter
l 57: remove comma
l 95: Could you be more explicit about the surface parameters that you changed? Eventually a figure to illustrate these would be good.
l 101: Since you are using a coupled atmosphere-ocean model which is supposed to be one of the main steps forward of this work, the fact that you do not take into account freshwater associated with the removal of the Greenland ice sheet (GrIS) should be mentioned more explicitly, in particular in the Discussion as a caveat.
l 105: This might be a bit subtle but do you interpret the *_atm experiments as the response of the atmosphere (alone) to the disappearance of the GrIS or the contribution of the atmosphere (only) to the climate response to the disappearance of the GrIS? I would tend to see it as the latter case and I think this is how you frame it below.
l 105-110: I am not sure how the *_atm experiments are done: do they really include the full coupled atmosphere-ocean model with SST and SSS nudging? Why not just doing an atmosphere-only run with prescribed SSTs? Do you really need 3000 model years in this case?
l 139-141: This description is very detailed, I don’t think you need to go so much into how much every region warms. You could spare a few lines here.
l 141: replace -6K by 6K
l 143: here and in many places in this section you refer to sea-ice changes without showing them. They only appear much later, in Fig. 10. I think these should appear right away and be discussed already here since they are crucial to understanding the temperature response.
l 145-146: Also, I would replace “atmospheric circulation changes” by “the atmosphere”. And I would say the dipole pattern is caused in part by the atmospheric response, and in part by the ocean. In relation to my comment above, I see these changes as contributed to by the atmosphere and/or the ocean.
l 148: I would write “limits temperatures to values below the freezing point”
l 152 and below: sea-ice changes are mentioned here again without referring to figures.
l 155: I think here where you write “feedback” you mean “effect”. The effect is opposite to what you describe in lines 153-154 but the feedback is the same.
l 157: I think here again you mean negative temperature response rather than feedback.
l 160: I think it is premature to say the atmosphere changes are associated with circulation changes. You can nevertheless say you will show in Section 3.1.2 that is the case.
l 166: remove the comma.
l 165-168: Why not comment on the ocean response briefly?
l 170-175: Please show the annual mean changes too.
l 185: 10-m winds are (in general) not reversed but decreased.
l 185 - 198: Here again you mention often changes in sea ice that should be shown. Also, you refer to temperature changes without referring to the figures, which would help the reader to follow the argument. Since the atmospheric circulation changes are only shown on an annual mean basis, showing the annual mean temperature changes would help too.
l 189: You mention here as a cause of the warming the reduction in sea ice but Figure 3d shows (for summer) a larger warming role of the atmosphere-only compared to ocean-only. Again I think it would be better to compare this with the annual mean temperature change.
l 190: Remove comma.
l 193: Do you mean insulation rather than insolation?
l 199-216: Here I repeat the same arguments as above. You very often mention sea-ice changes, which need to be shown, and you relate changes with temperature changes, which I think need to be shown and discussed on an annual-mean basis.
Figure 4c,d: a logarithmic scale would be better because the changes over Greenland are so large that the rest of the arrows and values are muted.
l 213: remove comma
l 216: remove “both”
l 225: include “not shown” at the end
Section 3.1.3: I think this section very much repeats what has been said in the previous ones so I don’t really think it is necessary. The only thing I think is really new is the description of how the surface properties change when the GrIS is removed, but since these are relevant from the beginning of the discussion they should probably be shown earlier. Figure 8, however, is a nice summary, so it could be kept but very much reducing its description.
Section 3.2.1: This section seems extremely long and descriptive to me. There are about six pages describing changes in the Arctic. I think there is no need to go in detail over each of the five basins as is done now, but rather try to synthesize the main results. In this line, the introductory paragraph (l 305-317) is not needed and the rest should not be arranged on the basis of individual basins but as a whole, much more succinctly. Therefore I do not go in detail over the text. Another important problem of this section is that many results which are mentioned are not shown. For instance, changes in deep water formation is mentioned in lines 376 or 443 but not illustrated in the figures. The same applies to the overflow in the Denmark Strait (l 433) and changes NADW and AABW (l 439).
Section 3.2.2: This whole section refers to AMOC changes but they are not shown. One wonders whether a figure has been forgotten but there is no reference to one. You should either focus on ocean circulation changes and show them or not.
Section 3.3: This section is supposed to focus on remote changes but these are not illustrated. A few remote changes are mentioned (cooling over Europe, a reduction in the storm tracks over NW Europe and a shift in the North Atlantic subtropical gyre) but these, again, are not shown. The authors have summarized these and other changes in a very nice figure but without showing the main changes many of the features shown just have to be believed. I would suggest removing the remote changes and focusing on the regional ones, and keeping a short description of the figure (Figure 13), which is a nice summary.
Section 3.4: As before, this section is extremely long and repeats very much what has already been learnt. As I said before, I would just keep a short description of figure 13, which is a nice summary.
Section 3.5: This section is interesting but seems to be out of scope. I am not totally against it but given the length of the manuscript and the results that are omitted the authors could consider removing it. There are many studies with ice sheet models addressing the irreversibility of a Greenland ice sheet disappearance.
Citation: https://doi.org/10.5194/esd-2024-24-RC1 - AC1: 'Reply on RC1', Malena Andernach, 29 Oct 2024
-
RC2: 'Comment on esd-2024-24', Xavier Fettweis, 25 Sep 2024
This paper presents sensitivity experiments of the global model MPI-ESM with or without the Greenland ice sheet. It is interesting to read, mainly well written and figures are nice. It then deserves to be published in ESD.
However, some discussion about the usual atmospheric indexes (NAO, AO and GBI) and their seasonal variability is missing for me as it will help the readers to put in a larger context the atmospheric impacts of noGrIS. Moreover, the impact on the Jet Stream and its variability should be also discussed for me as it is the main driver of the atmospheric dynamics. As this paper is already too long, I suggest to split this paper into 2 papers with a partim 1 about the atmosphere and associated climate indexes and with a partim 2 about the ocean (including its feedback on the atmosphere).
Finally, the impact of noGRIS on the SMB is a bit out of context here and not enough scientifically robust with respect to other parts of this paper. This part should be leaved for another paper for me where a fully coupled ice sheet – atmosphere – ocean model should be used to evaluate this as a regrow of the ice sheet will impact the atmosphere which is not taken into here while discussed in depth in the other parts of this paper.Citation: https://doi.org/10.5194/esd-2024-24-RC2 - AC2: 'Reply on RC2', Malena Andernach, 29 Oct 2024
-
EC1: 'Comment on esd-2024-24', Michel Crucifix, 04 Nov 2024
First of all, I would like to thank the reviewers for their careful and overall constructive assessment of the manuscript. The authors have already provided a fairly extensive reply, commenting on both editorial and scientific issues.
There appeared to be consensus about the manuscript being already fairly long, and on the hand lacking figures describing changes in ocean circulation. Reviewer #2 has suggested splitting the manuscript in two parts. I respect the decision of the authors to keep a one-part format, while shortening the analysis of the atmospheric response. Now that this option is taken, it is up to the authors to provide a revised version with the suggestions for added material about the ocean circulation and sea ice, while keeping the manuscript size within reasonable bounds.
Before doing so, I'd like to provide a few additional comments and suggestions based on the reviewer’s analysis on my own reading of the manuscript.
-l .24 – 27: I agree with reviewer #1 that one collective citation of 15 studies is not so good practice. While I concede that it is not the place for a full review, perhaps splitting the citation into three groups describing more what they have brought would be better.
Regarding the list of experiments: it still not entirely clear (to me) why the nudging experiments will provide different results than equivalent atmosphere-only experiments. I understand from the response of the authors that the parameter configuration of the atmosphere in the model is not the same as in the coupled version and perhaps that, in this case, the nudging approach was a more straightforward and secure option to isolate the atmospheric contribution (even though a much computationally-expensive). But other than this technical aspect, I don't see how the atmosphere would see something different in a nudging experiment than a an experiment were sea-ice and SST would both be prescribed (if changes in sea-ice are allowed with respect to the control experiments, then we have an experiments considering both atmosphere and sea ice responses and they should be clarified; the only other point I can think of is a change in ocean variability at a period smaller than the nudging timescale). Can you comment on this?
The CTRL_Wind experiment is also a little bit mysterious to me. It would be useful to be more specific about what is meant when speaking about flux correction. It doesn't seem to be a classical heat or freshwater flux correction, but rather, seemingly, a wind correction. Is that correct? Can you provide a little bit more detail?
Both reviewers raised some doubts about the relevance of the discussion about the regrowth of Greenland ice sheet and the related discussion of the surface mass balance. The authors have provided a rebuttal showing that they have considered the comments and their final decision should be respected.
Authors have supplied new figures to answer the questions of reviewer #2 about jet stream indices (Figures 1 and 2). Technically, these figures are now published material (since the review is public) but I understand that adding them in the manuscript would not help the objective to shorten the paper. A brief note about the analysis could nevertheless be added somewhere in the final version of the paper.
Finally, the authors mentioned here and there statistical significances. It is good practice to mention what is the associated null hypotheses (independent samples, equal distribution?)
I know invite the authors to provide the revised version of the manuscript, along with a point by point response to reviewers (which is expected to amply recycle material already provided in the public discussion). Given that there were a couple of minor scientific issues and some more substantial editorial comments, the revised manuscript will be sent again for further examination by the reviewers.
Most sincerely,
Michel Crucifix
Citation: https://doi.org/10.5194/esd-2024-24-EC1
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
284 | 51 | 108 | 443 | 9 | 9 |
- HTML: 284
- PDF: 51
- XML: 108
- Total: 443
- BibTeX: 9
- EndNote: 9
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1