Dear Mathieu Morlinghem and co-authors,
This manuscript received two in general very positive reviews that highlighted the importance and the advance and extension of the modelling approach to a whole section of the Greenland ice sheet and thereby providing a significant step towards the highly relevant topic of predicting the evolution of ocean-terminating outlet glaciers in Greenland.
Both reviewer had however also some critical points and suggestions for improvements, in brief:
1) correct all the typos and polish the text (that should have happened before submission), and
2) improve and better justify the methodology and descrition of the methods and experiments.

Regarding point 1) all the editing and languages issues as listed by the 2 referees have been corrected and improved, but unfortunately some new typos/text issues have been added with the revised text (see text below). Again, a careful proofreading would have avoided this.

The authors also added a very useful figure showing the modelled velocity response.

Regarding point 2) most points have been addressed satisfactorily and carefully or justified why nothing has been changed. Regarding some of the valuable questions of referee 2 on the modelling approach/experiment choice (see details in list below) have also been well explained and justified in the response, but as these are general points that are relevant for all readers, I recommend to add a few brief sentences of the justification (see details in list below) also in the text of the paper (not just in the author-response). Specifically, these points concern:

2a) comment by referee 2 regarding why 'calibration of sigma_max for retreat rather than pre-retreat positions': a satisfying explanation is given in response but not added to the text of the paper. This is relevant for all readers not just the referee so please clarify this also in the paper-text (maybe in section 2.3 calving parametrization; something along the lines of your explanation in the response of: 'Another reason is that stable glaciers generally have their terminus on a distinct feature (ledge, ridge, etc) in the bed topography and the numerical model is also stable for a wide range of !max. Constraining the calving threshold, !max, is easier to do for retreating calving front as we constrain the rate of retreat as opposed to just the stability of the glacier. This is something that we noticed in Morlighem et al. [2016] and Choi et al. [2018].').

2b) comment by referee 2 regarding stepwise increase of thermal forcing anomaly and how different would results for linear/gradual forcing be. I can accept this choice of step changes and I am happy with the justification in the response, but please add this explanation/justification (sensitivity analysis rather than projection, to assess glaciers at risk and bulk magnitude oif mass loss) also in the paper-text (in section '2.4 Experiments').

Thus overall this manuscript is very close to publication but these 2 points above regarding methods/experiment choice should be addressed in addition to the general technical/minor points listed in detail below. And please very carefully proofread the final manuscript again!
I congratulate the authors for their very interesting paper and also would like to thank them for their supportive handling of the revision process.

Best regards
The editor, Andreas Vieli, 3 Feb 2019


Rather minor technical and editing issues

p. 3 line 25: maybe 'The model mesh comprises…' is more elegant.

p. 5 line 12: a 'the' is missing here: '…is the temperature of the local freezing point…'

p. 5 line 15: delete one of the 'a' before 'high-resolution'.

p. 5 line 16: style: maybe 'ocean modelling study' is better.

p. 5. L.ine 17: plural 'experiments'

p. 7. Fig. 3 caption: '…(m/day) and calibrated…'

p. 8 line 9-10: there is a 'for' missing: '… as for all other experiments'.

p. 8 line 13-14: awkward formulation, maybe simplify to: '…an artefact that is due to our underestimation of the rate of undercutting for this region'.

p. 8. Line 25: something wrong in this sentence, maybe change to: '…the ice front positions, that were manually digitized from Level 1 Landsat imagery, together with the modeled…'

p. 9 table 1 in caption: add that positions are along the flowline/centerline: '…observed and modelled ice front retreat (in km along the flowline9…'

p.10 fig 4 caption: the 4th word (modelled) should be 'lower case'

p. 10 line 10: '…upstream of the 2007 position',…'

p. 11 Fig 5 caption: should be in plural: 'Modeled ice velocities (solid lines) and ice front positions (dashed vertical lines)…'

p. 13 line 3: style '…mass loss is significantly higher…'

p. line 13: either consistent plural or single tense: '…any glaciers which advance.' or 'any glacier which advances'.

p. 13 line 29: simplify to '…consistently under-estimates ice sheet mass loss…'

p. 13 line 30: '…should therefore be treated with …'

p. 14 line 5: should it not be '…towards ice-ocean-climate coupled model …'

p. 14 line 9: I think there is a 'more' missing: '…is itself more sensititive to…'

p. 14 line 14: '…no numerical ocean model…'