Dear Dr. Hodgson et al.,

Thanks for your revised manuscript and response to the two reviewers’ comments. Your response is thorough and thoughtful, further strengthening your submission. This submission is timely, given the Brunt Ice Shelf’s apparent vulnerability, and you provide the context needed to better understand its future. I learned quite a lot reading it, and the expert reviewers are also generally satisfied with it.

Below I have several minor comments for your consideration prior to a final decision on your manuscript. My main concern at this stage are the quality of the figures, which in general could be improved, but I have a couple of broader questions also.

Thanks again for considering The Cryosphere for your work,

Joe MacGregor
NASA GSFC

103: Here are elsewhere, it is mentioned that the future of the BIS depends on the propagation direction of Chasm 1, but it is never stated what direction it is currently propagating and therefore which scenario might carry greater weight. This is a missed opportunity. Of course, any mention thereof should be carefully qualified, because that direction could change, but as it stands it introduces an unnecessary degree of ambiguity into the manuscript.

151-3: I am confused by the description of this final step of the inversion here and in the supplementary material. To solve inverse problems, one generally does not seek a model that fits the data (where available) exactly, as that risks over-fitting the data. It’s more prudent to seek the minimum amplitude correction necessary to fit the data within their uncertainty bounds, with a suitable tradeoff factor based on expert assessment of the datasets and forward model. This concern seems particularly relevant given the different spatial scales that aerogravity data are sensitive to as compared to seismic and swath bathymetry. Please assess this concern or at least state clearly in the text whether the uncertainty in the data is effectively ignored.

Figures:

In general, I find figures are better served long-term if they can be interpreted as independently as possible from their captions. This requires both legends and labels. Presently, the authors favor a caption-heavy approach, but I encourage you to explicitly reconsider whether that is the most appropriate approach for your figures in the long-term.

Please use a consistent symbol for each station throughout the figures.

Figure 1: Label panels (a) and (b). Is the light blue where the rifts are entirely artificial? I’m assuming it is, as I do not recall such unusual coloration in my recent flyover of this region, but it’s not mentioned in the caption. Separately, despite the graticule, an inset map showing the location in Antarctica would be appropriate to include here.

Figure 2: These photos really need labels and their approximate locations ought to be labeled in Figure 1a.

Figure 4. The color range used for the bathymetry undersells the range of topographic features that is discussed in the manuscript, especially the inferred former grounding lines. Further, the thin blue and black lines representing aspect of the sub-ice-shelf cavity and bathymetry are nearly invisible. Please reconsider colors used in this figure.

Figure 5. I like the way the comparison between the airborne and satellite data is done. Label physical property represented by color bar. Label meaning of flowlines (that they’re shown in Figure 6). Because of the physical importance of the small area where the BIS still contacts the McDonald Bank (and the weight that the manuscript lends to it), an inset panel should be generated that zooms in on that region. As it stands, I think they’re only shown (but without being referred to in the text as such) in Figure 4?

Figure 6. Instead of solid/dotted/dashed lines, for which the latter two are nearly indistinguishable, used three shades of the same color to distinguish between the different flowlines. “Distance from MIR” should instead by “Distance upstream of MIR”. In panel a, label ice flow direction.

Text:
29: ice shelf depends on
38: 74ºS
40: 80 km
97: including:
102: outcomes depend initially on
125: is ground, airborne
126: aerogeophysical
132: no need to qualify data as “innovative” if they have already been described in a previous study
230-1: clarify that the “orientation” of the icebergs presumably refers to their long axis?
232: was examined in detail along
257: McDonald Bank
289: is neither thick nor extensive
326-7: values of depths should not be negative
341: of the BIS depends not only
408: …in a region of Antarctica that is not presently experiencing a rapid increase in atmospheric temperature or a known intrusion of circumpolar deep water, the BIS…
408-417: merge paragraphs
416: Chasm 1

Typos in Supplementary Material:
- Cochran not Cochrane
- kg m–3 not kgm–3
- respectively not respectivley

Thanks for your contribution to The Cryosphere and congratulations on a timely and interesting study.