Revisions

[adele-morrison]

Adele: Reviews are in! All very positive and constructive, but quite lengthy and a number of extra figures requested.

Editor:

Thank you for this interesting and well-written contribution. We received three positive reviews, including a number of constructive suggestions to improve the manuscript and clarify several points. From my read of the reviews and the manuscript, it feels in between minor and major revisions, but I tagged 'major revisions' to give you more time and scope to address the suggestions. Reviewer 3 (R3) also provided the most in-depth review, and requests 'major' revisions with respect to the introduction/framing of your study and consideration of additional diagnostics for the overturning circulation. R2 also notes a pair of very recent studies that should be considered in the framing. I agree with these suggestions, and it will be helpful to ensure that your study is embedded in the context of this recent literature in the introduction and discussion/conclusions, to be clear about what is new in your contribution. It is my sense that the reviewer suggestions are very constructive and should be relatively straightforward to manage. We look forward to the revised submission.

Due: April 12th

[adele-morrison]

Reviewer 1:

The authors conducted sensitivity experiments using the ACCESS model and show how the Antarctic coastal region responds to meridional and zonal winds. The manuscript reads well. I do not have much concern about their analyses and conclusions. I suggest a minor revision.

Minor comments

Lines 103-115: I am aware of Kiss et al., 2020. However, I still recommend authors show (1) the difference in their CONTROL compared to Kiss et al., 2020 and (2) model evaluation compared to observations. I recommend showing simulated T and S for coastal regions, comparison of sea ice extent, sea ice formation rate, AABW production, etc. Even if it is similar to Kiss et al., model evaluation is crucial for readers to understand the meanings of sensitivity experiments. If authors use the same run compared to Kiss et al., I would like authors to repeat this information for new readers and cite papers including figure numbers.

Figure 3 Similar to my first comment I would like to see simulated bottom (or mid-depth) temperature and salinity are well.

Figure 9 Same as above but for temperature

Figure 10 Nice Schematics.

[adele-morrison]

Reviewer 2:

Comments on "Sensitivity of Antarctic shelf waters and abyssal overturning to local winds" by Adele Morrison, Wilma Huneke, Julia Neme, Paul Spence, Andrew Hogg, Matthew England, and Stephen Griffies.

This work is useful for the community as it improves our general understanding of the processes affecting Antarctic shelf water temperatures, which has important implications for ice shelf melting and sea level rise. The paper is well written, the figures are clear and the experiments are well designed. I think that a major novelty from this work is the importance of the meridional wind component which has previously been overlooked. I therefore recommend this manuscript for publication in JCLI. I only have a moderate comment, and a few very minor comments and suggestions.

Main comment:

Further discussion on the choice of the wind perturbation would be very useful. First, the perturbation to the wind field was applied south of the dividing line between the annual average easterly and westerly winds, so basically only the easterlies. Increased anthropogenic emissions should also modify the westerlies north of this line. How accounting for changes in westerlies would change the story? Second, the perturbation consists of increasing/decreasing winds by 10%. How large is this compared to PMIP/CMIP projections? (both for the zonal and meridional wind components).

Other minor comments:

Introduction, L. 90-93: FYI, two recent studies give a slightly more complex picture of the variability of currents at the shelf break in the Amundsen Sea:

• Silvano, A., Holland, P. R., Naughten, K. A., Dragomir, O., Dutrieux, P. and others (2022). Baroclinic Ocean Response to Climate Forcing Regulates Decadal Variability of Ice‐Shelf Melting in the Amundsen Sea. Geophys Res Lett, 49(24), e2022GL100646.

• Caillet, J., Jourdain, N. C., Mathiot, P., Hellmer, H. H. and Mouginot, J. (2022). Drivers and reversibility of abrupt ocean state transitions in the Amundsen Sea, Antarctica. J Geophys Res Oceans, e2022JC018929.

Section 2 (Methods): Very few studies explain what is done with the coarse JRA55do atmospheric forcing along the Antarctic margin. I think that it would be important to provide this information, especially in a study focusing on coastal winds. For example, some groups extrapolate the atmospheric properties from JRA55do's ocean points to the ocean model grid points that would otherwise be under JRA55do's cold and dry over-ice-sheet conditions. The interpolation method itself (e.g., linear, cubic) is also worth a mention.

Section 3 (Results): DSW are somewhat defined L. 240, but this comes too far after Figure 2 where DSW export is plotted. Furthermore, I find the definition of the DSW threshold (L. 240-247) difficult to understand. What is the "peak transformation"? Would it be possible to either provide more details in the text or add a panel to Fig. 4 to illustrate the method?

L. 251-258: It may be worth mentioning that the rejection of the katabatic hypothesis is not so clear for the Ross box, for which the purple bar is closest to the black bar than to the blue bar.

Hypothesis 2: How are calculated Ekman velocities? From the stress exerted by air and sea-ice at the ocean surface, or from the open water stress, or from the air stress at the ocean and sea-ice surface?

L. 355: to support the claim that "northward sea ice advection is primarily governed by the meridional winds", you could add that southern sea ice typically drifts with a 20-25° angle to the left of the wind vector (e.g., Nansen 1902; Brunette et al., 2022). This also explains why approximately 1/4 of the enhanced northward sea ice export is due to increased zonal winds (Fig. 8c).

• Brunette, C., Tremblay, L. B. and Newton, R. (2022). A new state-dependent parameterization for the free drift of sea ice. The Cryosphere, 16(2), 533-557.

• Nansen, F., 1902: The Oceanography of the North Polar Basin, The Norwegian North Polar Expedition 1893-1896. Scientific Results, Vol. 3, Kristiania, 427 pp.

L. 362-375: The meridional wind perturbation explains only ~3/4 of the enhanced sea ice export (Fig. 8c). Yet, it seems to explain entirely the enhanced DSW export (Fig. 8b). An explanation for this should be provided.

L. 387-389: Is the cooling only related to Ekman dynamics or also to changes in sea-ice production on the continental shelf?

[adele-morrison]

Reviewer 3:

Review of: Sensitivity of Antarctic shelf waters and abyssal overturning to local winds
Authors: A. K. Morrison, W. G. C. Huneke, J. Neme, P. Spence, A. McC. Hogg, M. H. England, S. M. Griffies

This manuscript uses a suite of high resolution (0.1°), global ocean-sea ice simulations to assess how perturbations in surface wind stress, localized around Antarctica, impact the formation of Dense Shelf Water (DSW) and the abyssal overturning circulation. I greatly appreciated this study and its aims as it addresses an important gap in our understanding of the ocean circulation around Antarctica. Many studies have focused on dynamical responses to changes in surface forcing over the continental slope and at the shelf break around Antarctica; less attention has been dedicated to dynamics broadly over the continental shelf and in particular near the coast. At the shelf break, perturbations in the zonal wind stress have consistently been shown to impact the frontal structure of slope currents and the associated onshore heat transport. Yet, there is growing evidence that these processes can be (and likely are) decoupled from the physical mechanisms that set the rate of
water mass transformation over the continental shelf. Thus, the key result of this manuscript, that the meridional wind stress has a much larger impact on dense shelf water formation than zonal winds, is important to share with the larger community. The authors note that this result has been anticipated by previous studies, but this is, to my knowledge, the first study to explore this relationship in a circumpolar model.

The manuscript is well written and the figures are clear — it was a pleasure to read! My main comments below relate to the discussion of the study in the context of previous work as well as some suggestions for providing additional diagnostics of the overturning circulation, a key target of this study. I hope the authors find the comments below constructive and useful for revising the manuscript.

Major comments

1. While the introduction is well written, I feel that this paper could have a greater impact if this section was revised to be more specific about what is and is not known about the sensitivity of the Antarctic circulation to surface perturbations. You mention at a very high level that previous studies have shown that regional wind forcing can impact the circulation. However, there has been a very focused line of research that has linked localized surface wind forcing (easterlies in a very focused area) to glacial melt rates in the Amundsen Sea. While these are regional studies, this thinking has had a strong influence on the Antarctic climate community. Yet, as mentioned above, there is a growing appreciation that coastal and continental shelf processes can decouple dynamics at the shelf break from regions of water mass modification, which are rarely located at the shelf break. It would be nice to see this addressed more directly. I provide a more specific example of
   how the introduction could be revised in my minor comment below (line 94).

The authors should also include at least one dedicated paragraph in the introduction that discusses previous studies that have identified the importance of meridional winds on shelf overturning rates. The authors mention Timmermann et al. (2002) and McKee et al. (2011), but the approach and key results from these studies should be described and put in context with the present simulations. Additionally, the results of Hazel and Stewart (JGR 2020, not cited) are quite relevant to this manuscript. Using a Weddell Sea-only model, with ice shelf cavities, they find a strong sensitivity of the shelf overturning circulation and ice-shelf melt rates to the meridional wind stress, rather than the zonal wind stress. The mechanism supporting this sensitivity is similar — a modification of the surface buoyancy forcing due to wind-driven changes to the sea ice distribution. The Hazel and Stewart study provides a thorough mechanistic interpretation of how why the overturning adjusts
by appealing to water mass properties over the shelf. Again, I appreciate that the present study considers circumpolar changes, but I feel it is important to provide a more detailed discussion of these earlier papers.

2. Please discuss the lack of ice shelf cavities and the potential impact of these missing dynamics earlier in the study. Ideally, this would be included in the Methods section.

3. I am somewhat confused about how katabatibc winds are distinguished from the meridional wind experiment. Hypothesis 1 and 3 both focus on changes in water mass transformation in the DSW formation region due to meridional winds, related either to katabatibc winds (Hypothesis 1) or the total winds over the shelf (Hypothesis 3). Is DSW formation diagnosed in the same regions across both the WIND+ and WIND+_no_local? If so, I do not completely understand why the two simulations are so different. Perhaps for Hypothesis 3, the perturbed surface properties need to be advected into the DSW formation region, but this is not stated in the text. This links to my next comments about supplying the reader with more details about how DSW production is diagnosed.

4. This study suggests that in the simulations DSW forms in limited regions near the coast (e.g. Figure 3). However, this is never explicitly shown in the paper. The analysis would benefit from one or more sections across the shelf that document that convection is indeed limited to coastal regions and not distributed broadly over the shelf. Additionally, although this paper focuses on the overturning circulation, the magnitude and structure of the overturning (volume transport in density space) is never presented. It would be nice to add this information either using the 1000-m isobath as a northern boundary (similar to the DSW export, but resolved in density space) or by looking at local closed overturning cells in various shelf seas by diagnosing volume transport across sections that span two coastal boundaries.

5. Figures 10 and 11: I like Figure 10 as a summary of the mechanistic response to a change in meridional or zonal wind stress. However, I feel that Figure 11 is unnecessary and I suggest removing it. First, the figure essentially provides the same information as Figure 10 — the forcing and response just change signs. More importantly, Figure 11 and its caption seem to suggest that this is the response expected due to a global warming-induced weakening of the easterlies. Yet, a projected future weakening should also account for the changes in atmospheric forcing and sea ice state, which are likely to differ significantly from the initial states applies in these experiments. This difference in initial state could modify the Antarctic climate system's response to a perturbation.

One additional thing to consider — in both of these figures, the warming and cooling signature are shown as uniform over the depth of the water column. However, the temperature structure is stratified such that the upper part of the water column (the Winter Water layer) should already be close to the freezing temperature and therefore relatively insensitive to temperature changes following isopycnal heave. Overall, the manuscript should be clearer about where in the water column warming signals are expected to occur.

Minor comments

Line 81. In the discussion of Stewart and Thompson (2012), you indicate that changes in the Ekman transport modify the strength of the abyssal overturning circulation. You should note explicitly that a change in the strength of the overturning requires a modification in water mass transformation rate and thus some additional change to surface or interior transformations. Since the temperature flux is held fixed in this study, the change likely arises from a modification of the surface density distribution. The discussion in the introduction should be cautious about attributing a perturbation in the abyssal overturning to winds alone as even in these idealized studies, there are complex interactions with the surface buoyancy forcing.

Line 93. The new GRL paper, Silvano et al. (2022), should also be discussed here in this paragraph. Also, I felt that the discussion here could be handled more carefully; it feels like the results from a number of papers are being convolved (although I sympathize that the literature on this topic has not always been very clear!). You state, "the Amundsen Sea has shown to be sensitive to the local easterly winds at the shelf break." What has been shown to be sensitive? How have these studies built on each other, and what assumptions have they relied on? For instance, the Holland et al. (2019) paper attempts to link wind variability in the tropics to wind variations at the shelf break (there is nothing about heat transport at the shelf break in this study), whereas the Jenkins et al. (2016) paper considers changes to the thickness of the CDW layer. The Silvano paper only focused on frontal structure at the shelf break and does not connect to changes closer to the coast.
I feel it is important to describe these difference clearly.

Section 2: It may help the reader to present a small table that summarizes the simulations that were conducted for this study.

Figure 1b: Is there a reason for plotting positive zonal winds as negative? This seems to be a non-intuitive choice.

Figure 2: Please provide a description of how DSW is defined and how its export is calculated.

Line 205: "As explained in Griffies et al. (2014), the resulting change in cross-slope gradient in steric sea level drives a mass redistribution, with mass flowing off the shelf to re-balance the cross-slope gradient in sea level that is set up by the dense water-driven changes in steric sea level." I found this sentence confusing and I was not sure how it contributed to the paper.

Line 210-11: "Fresher and younger waters are indicative of enhanced local ventilation." What is the mechanism for ventilation here? If it is convection (consistent with the deepening mixed layer depths), then the water should become saltier.

Figure 4: It would be nice to see the shape of the water mass transformation curves (in density space), in addition to the integrated volume transport.

Line 413: "In these regions, the increased meridional wind speed brings colder air and also decreases the sea ice thickness over the continental shelf, leading to enhanced heat loss to the atmosphere and increased mixed layer depths (not shown)." I would have expected that salinity would still dominate the surface buoyancy budget, even in warm shelf regions, since the CDW is capped by salinity stratified winter water. Why does cooling here lead to deeper mixed layers?

Line 444: "In the simulations presented here, the zonal easterly winds do not impact the strength of the abyssal overturning circulation." You should note that there is no change in the circumpolar integrated abyssal overturning in response to perturbed easterlies, but might there be compensating changes in warm and cold shelf regimes? Again, it would be interesting to not only show changes in the export of DSW, but also changes (or not) to the patterns of water mass transformation.

Line 440: "which only outcrops close to the Antarctic coast and therefore is not forced by the Ekman upwelling-favourable winds further north." That may be true, but access of LCDW water to the shelf will still depend, to a certain extent, on Ekman dynamics.

[adele-morrison]

How does this sound for an approach to revisions?

1. I create a response doc on overleaf.
2. People volunteer to address particular comments by putting their names on the overleaf doc.
3. I can also delegate (@PaulSpence tells me I need to delegate more...) for remaining people / comments if others would like.
4. We set a deadline (3 weeks from now?) and meet to discuss progress / remaining issues.

[AndyHoggANU]

Approach sounds good.
You do need to delegate more.

[PaulSpence]

While we are into taking advice from me .... stop working on Sunday! I'm on holidays this week buried in Hokkaido snow. Delegate away!!! Thank you Paul

…

On Sun, Feb 12, 2023, 11:46 AM Andy Hogg ***@***.***> wrote: Approach sounds good. You do need to delegate more. — Reply to this email directly, view it on GitHub <#73 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABSWJXEB3MJ6V5QY4FSDEB3WXBFIVANCNFSM6AAAAAAUY4QGRE> . You are receiving this because you were mentioned.Message ID: ***@***.*** com>

[matthew-england-unsw]

Maybe a delegation // volunteer compromise? [Coming from a serial non-delegator]
i.e. Step 1: people volunteer. Step 2: the residual is delegated.
I'm happy to volunteer all comments regarding the introduction text given I wrote it, penance :-).
And can help sort comments regarding the experimental design.

[StephenGriffies]

I will try to read through the draft responses. But my time is overly occupied this semester teaching a new course (GFD II) for the first time.

[adele-morrison]

No worries Steve! We'll reach out to you if we'd like your advice on anything in particular.

For others, please volunteer yourself to address some comments in the overleaf response doc here.

[matthew-england-unsw]

Done....

[wghuneke]

Where can I find the overleaf response doc? The link goes to the manuscript, am I missing something?

[PaulSpence]

There's a response doc folder on overleaf. I struggled to find it also ;) P

…

On Tue, Feb 14, 2023, 6:22 AM Wilma Huneke ***@***.***> wrote: Where can I find the overleaf response doc? The link goes to the manuscript, am I missing something? — Reply to this email directly, view it on GitHub <#73 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABSWJXD72OVR46UOGAUCJX3WXKQ25ANCNFSM6AAAAAAUY4QGRE> . You are receiving this because you were mentioned.Message ID: ***@***.*** com>

[julia-neme]

Is there a deadline for the response? Should we maybe put some?

[adele-morrison]

Yes, we should! Revisions are due 12 April. How about everyone completes their responses/changes in the next two weeks, and we meet during the week of 13-17 March to discuss.

[AndyHoggANU]

I don't think I can do anything before 15 March, sorry. Will hold off for now and pick up something that is left after then.