Review of “Brief Communications: Tides and Damage as Drivers of Lake Drainages on Shackleton Ice Shelf” by Sommers et al.

This paper presents a brief study of surface lake occurrence and drainage on Shackleton Ice Shelf, Antarctica. It examines spatial relationships between lake locations and drainage, and the locations and orientation of surface fracturing. It also examines the temporal relationship between lake drainage and tides.

Lake drainage through ice shelves is a potentially important mechanism to understand and this work examines some previously proposed controls on it (tidal flexure) and proposes new ideas (that the ‘activeness’ of damage controls lake drainage). For this reason, a study like this could be a valuable contribution.

Main points:

One issue is that the study is quite limited in scope, both in terms of space and time. The generalizability of the conclusions is very limited by the fact that this study only covers three years and one relatively small ice shelf. Given the computing platform the authors used (google earth engine), it seems like it could have been relatively simple to extend the work to other locations and, within the restrictions of the datasets used, to more melt seasons. The main novelty of the paper is the consideration of ‘activeness’ (i.e. how perpendicular is ice flow to the predominant fracture orientation). And one could imagine a paper on this topic (1) proposing activeness as an important factor and (1) thoroughly test if this is the case. However, without an extension in the temporal and spatial coverage of the analysis, I think it is difficult to conclude much about the importance or otherwise of activeness for lake drainage. For example, as I note below, the authors state “…it is clear that at least one of these factors, damage or activeness, must be present for lake drainage to occur.” (albeit with a caveat that the exact relationship needs further investigation).  I think, given the results presented here, this is a much too strong conclusion to draw at this stage. That leaves the main contribution of the paper as (1) proposing activeness as an important factor. It is worth making it clearer that this is the main contribution of the paper. Or alternatively extending the analysis so that it can achieve (2) as well. This would involve extending to other ice shelves and/or other time periods.

Other important potential extensions that are missing include between lake drainage locations and ice-shelf stresses and/or vulnerability to hydrofracture as proposed by Lai et al. (2020). These are obliquely discussed in the context of introducing the idea that how close fractures are to perpendicular to ice flow (activeness) is an important factor. However, no comparisons are made to them directly. This would seem like a more direct way (or at the least a complementary way) of testing the ideas that lie behind the activeness proposal.

Specific comments:

L14: “only some of these lakes drain.” This needs citations

L17: missing “an”

L44: reference should not be in parentheses

L44: explain what “Median image mosaics” are in more detail

L46: comma before ‘which’

Section 2.1: Inconsistent tense. Are and was are both used. Change to be consistent.

L48: not clear what grid is being referred to here. Is this a spatial reference grid? If so, give its name. Or maybe it is not needed here, unless it’s important for the analysis that it was this grid for some reason.

L52: rephrase “Antarctic ice flow velocity observations of 2019”

L53: delete “and kept constant during the studied period.”

Figure 1 caption: add ‘the’ before MEASURES.

L62: rephrase “combining all lake masks of a season.”

L78: “damage maps of 300 m” should be “300m-resolution damage maps”

L83: I am unclear what “(tolerating a deviation of 15◦)” means.

L98: delete “under-“

Figure 2 caption: rephrase “show NeRD damage and activeness of summer 2019/20” and “pixel quality” (I am not sure what quality refers to here).

L103: explain what “record summer” means explicitly. I am guessing it means record high temperatures and/or melt volume/extent. But which of these, I am not sure.

L104: delete ‘such as those by’ and put the citations in parentheses

L105: replace ‘the’ with ‘our’

L100: should this reference figure 3 as well as or instead of figure 2?

L105-107: “However, the findings indicate minimal lake formations in this region, suggesting immediate drainage of meltwater into the ocean in highly damaged areas (Figure 1 and 2).” This is not clear. Which part are you referring to as the north of the ice shelf? The north-west-most part is not covered by the damage map and the region to the south of that is not ubiquitously highly damaged, so I am not sure this is a fair conclusion to reach from these two maps (damage and meltwater).

L116: how did you decide on these values for the categories no, medium and high?

Figure B2: I think this would be useful to have in the main paper.

L118: replace “a detailed listing” with “a list”

L127: this is not an indication of where the ice is deforming, only of where it is flowing perpendicular to fractures. The ice is actively deforming essentially everywhere.

L128: replace “distributed in a bell-shaped curve” with “normally distributed”

L131: clarify what “the glaciers,” refers to here.

L132: delete ‘this’

L133: “tends to concentrate inland, away from the ice shelf edge” I do not see this spatial distribution in the figure. Can you explain this in more detail? What exactly do you mean by concentrate?

L136: would “fast flowing areas’ be more precise than “glacier zones”?

L138-139: “Unlike damage values, the activeness parameter does not appear to be directly related to the distribution of accumulated meltwater” This implies that the distribution of meltwater is directly related to damage. This is mentioned briefly at the start of section 3, but is this what is being referred to here? This should be made a little clearer and perhaps this statement softened somewhat, given that the distributions of damage and meltwater accumulation have not been explored in detail and it has not been established that there is a close connection (see my comment on L105-107).

L145: In what sense is the ice shelf a prototypical example? This paper provides no comparison to other ice shelves, so if it is typical, this needs to be discussed. And I think prototypical refers to this example in some way being the originator, or the original version of something, which, unless I am missing something, it is not.  

L147: I am not sure what “the glacier tongue” is referring to. Please clarify.

L157-159: “Although the exact relationship between these metrics requires further investigation across different ice shelves and with more drainage events, it is clear that at least one of these factors, damage or activeness, must be present for lake drainage to occur.” As mentioned in the main point above, this statement should be softened. There are examples in Greenland of fracture perpendicular to the background flow direction directly draining lakes on grounded ice. It seems likely the same is possible on ice shelves. A more precise statement restricted to what this dataset tells us about this ice shelf over these three years, given the limitations of the remote sensing datasets and your analysis, is needed here. In other words, not only does the quantification of the exact relationship require further investigation, so does establishing that activeness and damage are a requirement at all.

L161-163: I suggest deleting this opening paragraph.

L166-167: I suggest deleting the opening sentence of this paragraph.

L171: Delete “Our data reveals a trend:” A trend implies something changing over time. Also, it is unnecessary. I suggest just describing the relationship.

L177: delete “not only confirms but also extends the work of” and put the citation in parentheses. It’s not clear to me that this analysis confirms that work. Maybe you can say that it is broadly consistent with it.

L180: delete “highlight the multifaceted nature of drainage events and”

L182: “intricate”

L192-193: “the temporal resolution of satellite passes may not be sufficient to capture all drainage events, especially those of short duration.” This is a little repetitive of earlier in the paragraph.

L108-109: “produce different results” can you be more precise with this statement. Changing the thresholds would of course change the results quantitatively, but could it change things qualitatively too?

L226: this sentence mentions a predictive model for the first time. It isn’t clear what this is referring to. What would be the purpose of such a model? I was assuming it would be some kind of parameterization in an ice-sheet model, but the need for real-time data confusing me in that case.

The discussion: One limitation to the idea of activeness being a control is that fractures could advect into areas that are compressive, but the fractures could remain still perpendicular to flow. This scenario would yield non-zero activeness, but may not be conducive to hydrofracture. This underlines the utility of comparing lake drainage locations to ice shelf stresses and fracture orientation to principal stresses orientations.

Data availability: It would have been good to have access to the code and data for the review process.

Please, find my comments on the attached document.