By Chengyan Liu

General comments:

This paper presents an investigation of the ocean-cryosphere interactions off the Sabrina Coast of Wilkes Land, East Antarctica. Based on a coupled ocean–sea ice–ice shelf model, the authors studied the sea ice evolution, the basal melting of ice shelves, the properties of water masses and circulations, modified Circumpolar Deep Water (mCDW) intrusions over the shelf, the oceanic heat and volume transports, and the meridional overturning circulations within the sub-ice-shelf cavity around the Sabrina Coast.

The state-of-the-art topography data over the Sabrina Coast have been constructed and introduced in the coupled model, and the overturning ocean circulations within the sub-ice-shelf cavities are also shown in this study for the first time. The mechanism responsible for the differences in temporal variability between the basal melting of the Totten Ice Shelf and Moscow University Ice Shelf has been discussed, and the authors found that both mCDW intrusions and sea ice production contribute to the regional differences between the two sub-ice-shelf cavities. More interestingly, the model has captured an eastward undercurrent over the continental slope, which may significantly regulate the simulated seasonal variabilities of onshore heat transport.

It is very topical because ocean-cryosphere interactions around the Sabrina Coast are key processes for the marine ice sheet instability around East Antarctica, which has global implications for climate change and the sea level rising. I believe that this manuscript is very interesting to the Antarctic science community. My comments are given below, and I recommend the manuscript for publication in TC after minor revision.

Note: In the following, "L" means Line.

Specific Comments:

L165: ‘We used observation-based coefficients of the thermal and salinity exchange velocities for the ice shelf (γt=1.0×10⁻⁴ ms⁻¹, γs=5.05×10⁻⁷ ms⁻¹, Hellmer and Olbers, 1989)’

A fixed frictional velocity has been employed by the model. It would be nice if the authors could make a few discussions about the potential discrepancy of the fixed frictional velocity for the thermal and salinity exchanges at the ocean-ice shelf interface, by comparing it to the parameterization of the velocity-dependent scheme.

L485-500: The analysis of inflow and outflow transport across the southern boundary of the Slope Box is missing (Fig. 18). The Slope Box is different from the Sabrina Depression box since the Slope Box has an open southern boundary. Therefore, the transport balance between the inflow and outflow of the Slope Box can not be explained by the calculation confined within the western, northern, and eastern boundaries. The author may add the calculation and description of inflow and outflow at the southern boundary of the Slope Box in Fig. 18.

L490: The authors calculate the inflow and outflow from the surface to 800 m.

Does the vertical transport across the bottom boundary at 800 m depth have some influence on the balance of the inflow and outflow? It would be nice if the authors could have a short discussion on this.

Technical Corrections:

L115: ‘there remains large uncertainties’ should be ‘there remain large uncertainties’

L120: ‘It has been suggested that the inflow of mCDW onto continental shelf regions is related to the Antarctic Slope Front/Current (ASF/ASC) system on the upper continental slope region (Nakayama et al., 2021; Thompson et al., 2018; Silvano et al., 2019)’.

The study of Liu et al. (2013) described the dynamic mechanisms responsible for mCDW intrusions regulated by the ASC/ASF system in East Antarctica, and it might be suitable to be cited here.

Liu, C., Z. Wang, X. Liang, X. Li, X. Li, C. Cheng, and D. Qi, 2022: Topography-Mediated Transport of Warm Deep Water across the Continental Shelf Slope, East Antarctica. J. Phys. Oceanogr., 52, 1295–1314, https://doi.org/10.1175/JPO-D-22-0023.1.

L395: ‘with the seasonal peaks occurring from May to September in the TIS and eTIS’ and ‘The wMUIS reaches its peak between April and May, while the MUIS reaches its peak between October and December’

It would be nice if the authors could also identify these periods directly in Fig. 13 by using boxes or something else.

L445: ‘with the remaining 0.1% being explained by the sum of ice-shelf basal melting from the southern boundary (i.e., ice-shelf fronts), sea-ice production, transport, and melting over the SD box’

Is the ocean model used in this study a volume-conserved model? If the ocean model is volume-conserved, the ice shelf basal melting and the sea ice evolution only change the salinity rather than the volume transport. The remaining ‘0.1%’ may be attributed to the truncation error in the volume transport calculation, and it is so small that such remaining can be omitted without particular attention.

L555: ‘The southward heat transport timeseries’ should be ‘The southward heat transport time series’

L575 and L675: ‘This means that a positive value in the SAM index leads to weaker coastal winds.’

It would be nice if the authors could add some references corresponding to the weaker coastal winds in a positive SAM index.

L635: ‘in the surface layer, but this study’ should be ‘in the surface layer, this study’

L640: ‘The present model results shows that’ should be ‘The present model results show that’;

‘at mid depths’ should be ‘at mid-depths’

L650: ‘From modeling perspective,’ should be ‘From a modeling perspective,’

L655: ‘where the Antarctic Circumpolar Current (ACC) deflect southward’ should be ‘where the Antarctic Circumpolar Current (ACC) deflects southward’

L1095: The labels of the vertical overturning circulation in Fig 14b are missing.

REVIEWER'S REPORT FOR MANUSCRIPT tc-2023-78

GENERAL COMMENTS

The manuscript fits within the scope of the journal. By bringing a long-term perspective (70 years) based on a new bathymetric compilation that is considerably more realistic than in earlier modeling studies, it represents substantial progress beyond our current scientific understanding of this region. The purpose of the work is is clearly articulated in the text, I did not find substantial flaws in the methodology, and the conclusions of the study are adequately supported by the figures/tables of the manuscript. The results of the manuscript are appropriately discussed in the context of the existing literature and the authors abundantly cite other related work. Overall, this is a substantial contribution providing considerable detail on the melt of ice shelves in a region storing the equivalent of 3.5m of global sea level, which is obviously relevant to the large fraction of the population living along the world's coastlines.

SPECIFIC COMMENTS : MAJOR

(1) The one major flaw that I'm noticing is that the authors apparently made no effort in limiting the number of figures and tables (24 figures, 3 tables, which is something I've never seen before). So the manuscript fails the journal's requirement of "scientific results and conclusions presented in a clear, *concise*, and well-structured way". This being said, if the authors are reasonable, it should be easy to address this problem by following some, or all, of the following suggestions:

(a) This journal allows for a "Supplement", and I would encourage the authors to use it extensively. While the new bathymetric compilation is a major outcome of the study, I would think that Figure 2 represents more information than the average reader is interested in seeing, and so it could be moved to "Supplement". Figure 4 is mostly methodological in nature and could be moved to "Supplement". Figures 5-6 demonstrate the realism of the model but they are not necessary to support the scientific results (and therefore Figs.5-6 could go to "Supplement"). The timeseries of Fig.9 look so much like a sinusoid that one could say that they are already appropriately described by the panels inside Fig.8 (making Fig.9 a bit redundant). Figure 11 is barely discussed in the text (most of the text about Fig.11 is spent describing the Methodology behind it) and could be moved to "Supplement" as far as I'm concerned. Figures 19-22 appear (to me) to describe the same story in different ways, with Figs.19-20 being only very briefly mentioned in the text; given that, it's hard to believe that all of Figs.19-22 deserve to appear in the body of the manuscript.

(b) Couldn't Tables 2-3 be combined together? This would facilitate a comparison between this study (Table 2) and the earlier studies (Table 3).

(c) Another possibility is to use "(not shown)" when a result is interesting enough to be mentioned but it isn't critical to the scientific demonstration. For example, I felt like the 8 year cycle apparent in Fig.11(a) was a bit interesting and worth briefly mentioning in the text, but I didn't feel the need to see it as one of the manuscript's figures.

SPECIFIC COMMENTS : MINOR

(2) The journal requires the "experiments and calculations" to be "sufficiently complete and precise to allow their reproduction by fellow scientists (traceability of results)". Based on the authors' "Data availability" statements, they plan to share the model results supporting their conclusions "after acceptance". So I wasn't able to access the model results, but the authors seem to be willing to share them, eventually. The editor can decide whether this is good enough (or not).

(3) Another requirement of the journal is that the language be "fluent and precise". Overall, I thought the whole manuscript was well-written. My main concern is that some portions of the text felt unecessarily long. For example, lines 45-75 are fairly general and would fit better in an introductory paper (or a graduate student's thesis) than in a research paper where conciseness is key. Another example is line 591-616, which acts as a repetition of what the reader already saw inside Sections 3-6. I personally don't see the need for lines 591-616; the Abstract already provide a summary of the key results.

(4) Line 52: "...and iceberg caving at ice-shelf fronts..." (Typo: "caving")

(5) Line 89: "Previous observational and modeling studies have inferred that the interannual variability in glacier/ice sheet variable is strongly controlled..."

Something is wrong in the sentence; maybe delete the word "variable"?

(6) Line 127: "...heavily influenced by their artificial lateral boundaries, the conditions which were often derived from a different coarse-resolution ocean model."

The sentence could be interpreted by a reader as "all regional models are flawed by having artificial boundaries", which would be inaccurate. (The literature is filled with successful regional model implementations using highly-realistic lateral conditions.) Removing the word "artificial" from the sentence would make it a lot better; it gives more weight to the second part of the sentence and ties the criticism to something specific and quantifiable (the coarseness of the oceanic dataset used for the lateral conditions).

(7) Section 3 (Sea-ice extent and production): I could not find where the "observed" "sea ice concentration and extent" were from. What datasets were used? SSMI? Something else? This needs to be specified.

(8) Line 393: "The upper panels in Fig.13 show the net inflow volume transports..."

The word "net" is doing more harm than good in this sentence. "Net" is often used to represent "inflow minus outflow", while in this case the authors are really referring to the inflow. I assume you added "net" to refer to the fact that the inflow was summed across the ice shelf front, but I believe it is creating more confusion than anything. Could you take out the word "net"?

(9) Line 514-516: "We calculated the transports of the surface current by integrating the westward transport in density layers lighter than 27.7 kg/m3 and that of the undercurrent by integrating the eastward transport in density layers denser than 27.6 kg/m-3."

Why would you have such an overlap in your definition of the surface/undercurrent? As things stand, the density range 27.6-27.7 kg/m3 is accounted in both the surface and the undercurrent. Is it because these two currents reach their maximum magnitude at different times in the seasonal cycle, and the overlap was necessary to capture the full magnitude of the surface/undercurrents at those moments? Please clarify.

(10) Caption of Figures 7, 15: In an ideal world, the reader would be able to look at the figures and understand them without having to read the full manuscript. In the caption, could you replace the acronym "SD box" by "Sabrina Depression (SD) box"?

(11) Caption of Fig.10: Please define acronym "CKDRF" in the caption; this particular acronym only shows up a few times in the whole manuscript, so we shouldn't expect the reader to know or remember what it stands for.

(12) Figure 10: As the authors point out in their manuscript, this is one of the first studies to provide a multi-decadal (70 years) perspective on glacial ice loss, so there's an opportunity to provide new insight to the community. If we assume the 69-72.6 Gt/yr grounding line flux of Rignot et al. 2019 is valid over the period 1979-2017, and that the modeled basal melt of Fig.10 is accurate, then what fraction of the grounding line flux is due to calving? How does this fraction compare to the one reported in Fig.1 of Rignot et al. 2013 (although the time periods are different)?

Similarly, if we assume that the calving is constant over time, how does the trend apparent in Fig.10a over 2008-2018 compare with the trends in Miles et al. 2022 over the same period?

(13) Caption of Figure 12: Something is wrong in the sentence: "...velocity along a 5-km off along a line section of...".

(14) Caption of Figures 12, 17: Specify the units for potential density anomaly (kg/m3).

(15) Caption of Figure 13: See my earlier comment about "net".

(16) Caption of Figure 14: Clarify that the stream function is computed from laterally-integrated velocities, clarify that the units for the stream fucntion are mSv.

(17) Caption of Figure 24: Mention what geographical area the "wind speed" is representative of (presumably the yellow box of Figure 3, but this needs to be clarified).