the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Can rifts alter ocean dynamics beneath ice shelves?
Mattia Poinelli
Michael Schodlok
Eric Larour
Miren Vizcaino
Riccardo Riva
Abstract. Land ice discharge from the Antarctic continent into the ocean is restrained by ice shelves, floating extensions of grounded ice that buttress the glacier outflow. The ongoing thinning of these ice shelves – largely due to enhanced melting at their base in response to global warming – is known to accelerate the release of glacier meltwater into the world oceans, augmenting global sea level. Mechanisms of ocean heat intrusion under the ice base are therefore crucial to project the future of Antarctic ice shelves. Furthermore, ice shelves are weakened by the presence of km-wide full-thickness ice rifts, which are observed all around Antarctica. However, their impact on ocean circulation around and below ice shelves has been largely unexplored as ocean models are commonly characterized by resolutions that are too coarse to resolve their presence. Here, we apply the Massachusetts Institute of Technology general circulation model at high resolution to investigate the sensitivity of sub-shelf ocean dynamics and ice shelf melting to the presence of a km-wide rift in proximity of the ice front. We find that (a) the rift curtails water and heat intrusion beneath the ice shelf base and (b) basal melting of a rifted ice shelf is on average 20 % lower than for an intact ice shelf under identical forcing. We therefore posit that rifts and their impact in the sub-shelf dynamics are important to consider in order to accurately reproduce and project pathways of heat intrusion into the ice shelf cavity.
- Preprint
(17346 KB) -
Supplement
(18174 KB) - BibTeX
- EndNote
Mattia Poinelli et al.
Status: open (until 28 Mar 2023)
-
RC1: 'Comment on egusphere-2023-75', Anonymous Referee #1, 02 Mar 2023
reply
General comment
The authors present here numerical simulations showing how the presence of an ice shelf rift can act as a barrier to the inflow of off-shelf water, thereby reducing ice melt rates when compared to an ice shelf cavity with no rifting. This is a process which is not currently represented in large scale models and is likely to have implications for accurate predictions of future climate scenarios. Of particular note is that this reduction in melting is found to be concentrated in the proximity of the grounding line, an area of particular sensitivity regarding ice dynamics.
I find the manuscript to be generally well written, with an easy-to-follow methodology and results clearly presented. I have no major issues for the authors to address but do have several smaller comments. Assuming the authors address these I recommend publication.
Specific comment
In the paragraph beginning L 358 the authors mention some limitations of their modelling framework regarding sea ice. I feel this section could be improved by also including some mention of how the presence of frazil ice formation in supercooled water in the rift would affect their results. For example, the Jordan et al 2014 paper they cite found that the majority of freezing in their simulations was occurring via frazil ice formation rather than directly at the ice-ocean boundary. Would this potentially have the effect of enhancing the barrier to off-shelf water inflow?
When discussing the reduction in melt rates in section 3.2 the results are shown in absolute changes in melt rate for the anomalies plots. I think it might be good to also show the relative change in melt rates as a percentage of the reference case. This would more clearly show if there were a change in the melt pattern or just the magnitude of the melt pattern.
I also think that the finding that the biggest reduction in melt rates is seen in the grounding line proximity should be highlighted more in the introduction and conclusions. From an ice dynamic perspective, we know that melt rates near or at the grounding line are the most important, and so highlighting this finding would be of interest to the glaciological community.
When comparing the relative behaviours of the S-N and E-W set ups, would it not be a fairer comparison to have the same total flux of inflow water in each case? Unless I have misunderstood, the S-N case inflow happens over a 15 km section, whilst the W-E happens over a 50 km wide section. It could be argued that the W-E case therefore has a greater supply of water above/below the freezing point before any consideration is made to its direction of flow.
I would also find it more intuitive to have the ice shelf orientated North-South rather than East-West for the descriptions of scenario labels, though I admit this is a personal preference and not a requirement by any means.
Technical corrections
L 34 - "Albeit critical, ...." Reword for clarity, as it currently reads as if the pathways themselves are elusive rather than the observations of them.
L 55 - First use of Tfr, need to define.
L 56 - atmosphere -> atmospheric
L 57 - relative _> relatively
L 64 Would read clearer as "water formed within them"
L 93 I assume intensity is referring to inflow velocity?
L 95 "as a discontinuity"
L 98 " Despite the fact ice "
L169 "that computed”.
L169 "under an identical"
L198 "a negligible"
L 205 "to a progressively "
L 234 "the inland-most"
L 237 "freezing within”.
L 253 "being the same”.
L 318 "tends to flow”.
Citation: https://doi.org/10.5194/egusphere-2023-75-RC1 -
RC2: 'Comment on egusphere-2023-75', Anonymous Referee #2, 14 Mar 2023
reply
General comments
This highly idealised model study investigates the effects of a rift across a floating ice tongue in an ice shelf, using the MITgcm numerical model. As a result, import of relatively warm off-shelf water into the ice cavity is decreased by the presence of a rift and further decreases with increasing width of the rift.
Generally, the manuscript is well-written and the results are convincing. There are many grammatical errors of which I have just noted a few. The design of the idealised model setup needs more motivation. The ice-ocean interface is quite flat (just extending from 250m at the grounding line to 200 m at the calving front), and there is no sill. Some discussion of these simplifications and their impact on the results is needed. The authors criticise existing more realistic and larger-scale models that they do not resolve such rifts. Those rifts are however, as mentioned in the discussion highly intermittent at short time scales of days, and it is unclear how this relates to the simulations performed here that are run for 7 years with a stationary under-ice topography. It would be good if the authors could propose a concept how these subgridscale processes can be considered in numerical models in general.
More discussion and explanation is needed in general. These suggested modifications of the manuscript are largely minor.
Specific comments
16: How can you have an acceleration of 94 Gt/yr ea h decade at a current rate of 134 Gt/yr? Does it mean that the rate was negative until two decades ago?
49: A brief further explanation of the "compressive arch" might be helpful.
53: typo „orders or magnitude“.
63: „limited to a 2-D rift environment“: which two dimensions are meant here? Two horizontal or one vertical and one horizontal?
78: How much can potentially important non-hydrostatic effects at vertical side walls of 200 m thickness be ignored in a 1 km wide ice rift? Did you use any of the entrainment formulations existing for vertical melt water plumes?
86: Please indicate which latitude this Coriolis acceleration corresponds to.
87: With a vertical resolution of 10 m, subglacial plume are not vertically resolved, see the discussion by Burchard et al. (2022). What consequences does this have for the vertical heat transport towards the ice-ocean interface?
91: Please indicate for which ice shelves a grounding line depth of 250 m is characteristic. To my knowledge the large ice shelves have much deeper grounding lines. Also, no sill has been imposed at the bottom. Which impact does this simplification have compared to typical ice shelf dimensions?
98: Better „sea ice“.
Section 2.2: Please mention that no subglacial discharge was prescribed at the grounding line and discuss the implications. Do also mention that ice melt does not move the ice-ocean interface.
140-141: Isn’t cross-section B showing the „horizontal distribution of tr01 across a horizontal cross-section“?
144: „heat fluxes (in the form of melt rate)“, not clear what is meant here. Heat flux and melt rate have a complex non-linear relation, how can heat fluxes then be presented in the form of a melt rate?
185-186: typo „in almost all experiment“.
222: should be „re-freezes“.
224: Shouldn’t it be „upwelling“?
229-230: What do you mean with „stronger off-shelf water“?
234: Do you mean „reach the inland-most section“?
235-237: This looks like an artefact of the experiment. To widen the rift, also the west flank could be moved further west. By widening the east flank towards the calving front, two effects are mixed up: closer proximity to the open ocean and a wider rift.
Fig. 7, panels i-p: Since only one contour line is shown, it cannot be seen where minima and maxima occur. It would be better to show a different contour line in a different colour.
262: What is the thickness of this buoyant plume as calculated by the KPP model and how is it resolved by the vertical discretisation?
268: Also for the case without rift (Fig. 7m), the concentration of tracer 1 is very low near the calving front.
290-291: This indicates a 75 m thick buoyant plume. Is that what you expect under Antarctic ice shelves or is this a model artefact?
304: should be „leads to spreading of saltier and denser water“.
306-307: Isn’t this mainly because the initial tracer contents in wider rifts is much larger than in narrower rifts?
318: should be „tends“
347-350: This shows the potential impact of subglacial discharge that has been neglected here. A doiscussion of the effects of this ommission is needed here.
354: typo „As an precise representation“.
References:
Burchard, H., Bolding, K., Jenkins, A., Losch, M., Reinert, M., & Umlauf, L. (2022). The vertical structure and entrainment of subglacial melt water plumes. Journal of Advances in Modeling Earth Systems, 14(3), e2021MS002925.
Citation: https://doi.org/10.5194/egusphere-2023-75-RC2
Mattia Poinelli et al.
Mattia Poinelli et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 299 | 97 | 10 | 406 | 31 | 5 | 5 |
- HTML: 299
- PDF: 97
- XML: 10
- Total: 406
- Supplement: 31
- BibTeX: 5
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1