Editor’s note on tc-2017-241

Thanks to the authors for their revisions to this paper in response to the referees’ comments. Looking over the comments, I feel that the manuscript would benefit from restructuring to address referee 2’s comments that the text tends to repeat itself, and that the figures show too many versions of very similar model fields. The authors’ responses have defended the repetition, but I recommend reducing the number of panels in figures 5 and 9 (see below) and moving figure 7 to the supplement. The corresponding descriptions in the text should also be reduced and/or moved to the supplement.

Both reviewers commented on the choice of a non-zero sea level. The authors need at least a sentence to explain their choice of WGS84, rather than a geoid, as their reference surface, and to evaluate the magnitude of the problems this might cause. This is a potential problem in calculating the driving stress (which should be determined by the surface gradient with respect to the geoid) and in calculating heights above flotation (as is relevant to the companion paper). Given that the authors are using WGS-84 as their reference surface, it is not obvious that the height of the sea surface in BEDMAP-2 is the appropriate reference surface, so the authors should refer to the BEDMAP-2 documentation to see how what the ocean surface there is intended to represent: is it a mean sea-surface height or the height at some tidal phase (i.e. mean low-low tide)?

At least one referee commented on the linear sliding law. The authors should make clear in the text that the sliding law is only used as a numerically convenient tool for calculating the basal shear stress.

I recommend that the authors convert figure 5 to include only two versions of the temperature initialization (the one selected, and a second version, of their choice, that demonstrates unacceptable behavior), and only the first and last steps of the temperature iteration. The intermediate steps and the full range of models tested can be described, very briefly, in the text.

I recommend that the authors remove the right-hand column of figure 9 (the relaxation height difference). This column is given only cursory attention in the text.

For all figures (5-9) with multiple panels, the rows and the columns of the figures should be labeled, so that the reader does not need to refer to the letter key in the captions to interpret each panel.

These changes are somewhat substantial, and I would like to look over the authors’ response to them before I consider whether it needs a re-review or not.

Best
Ben Smith

Thanks for your responses to my comments on the second version, and please excuse my inattention in missing your response to the referee about the nature of the basal-friction coefficient. I see that you answered that question well in your response.

Please consider moving the third column of table S5 to table 2 in the main text of the paper. The RMSE of the surface velocity seems like an important metric of the success or failure of each inversion scenario, and helps place information needed to interpret the results of the paper in the text rather than in the supplement. Once this is done, I think the paper should be ready to go.

The paper looks good- I think it's ready to go.