Both referee 1 and 3 have now provided feedback on your revised manuscript. As you will see, R1 recommends a minor revision, but still has some comments which should be adequately dealt with. Most importantly, R1 finds that the discussion and conclusions lack a clear message on the added value of these data for advancing with respect to the work of Walker and Gardner (2017) and Zhao et al. (2017). Although I believe your work provides a valuable contribution to the literature, I agree with this statement, and would like to see such a discussion so that the reader can put your work in the right context. Although the other comments of R1 are less critical, they should be addressed as well; in particular, please consider discussing the velocity changes in subsections of the profile on P9L25.

Although the list of comments is greatly reduced compared to R3's first review, some concerns remain and a major revision is suggested. I believe the concerns of R3 can be addressed with a number of relatively straightforwards additions and changes:

* provide statistics of % removed velocity measurements in the upper areas where possible

* provide an estimate of the uncertainty in the elevation measurements after regridding and discuss how this propagates into uncertainties in elevation changes

* provide the size and range of the vertical registration and characterize its spatial variability

* wheras I agree with you that the error penetration bias is hard to estimate, R3 has a point in the comment that TSX DEM differencing results should be validated against (other) independent measurements. This would provide a rough estimat of the remaining error due to penetration effects. Also, the potential impact of spatial variability in the correction, which appears to be apparent in fig. S4b, should be discussed more extensively and the choice for the 0.2 m/yr should be better motivated.

* further discuss differences w.r.t. Walker and Gardner (2017), following the test suggested by R3

Of course, remaining comments on wording and figures should also be properly addressed.

I'm looking forward to your response.

Bert Wouters - TC/TCD Editor

Dear authors,

It's my pleasure to inform you that your manuscript has been accepted for publication in The Cryosphere, pending a number of minor technical corrections (see list below). Congratulations, and thank you for your patience in the longer-than-expected peer review process. Most of all, thank you for your valuable contribution to our journal.

best regards,
Bert Wouters - TC/TCD Editor

Technical issues to be fixed before publication (pages and line numbers refer to the annotated manuscript in the Author's Response file):

p 3:
line 3: space missing in '2014and'
line 25: 'pining point'
p 4:
line 16: remove 'the order'
p 9:
line 11: 'positive ice thinning rates': a positive thinning rates refers per definition to a negative dh/dt. Is this what you mena, or do you mean ice thickening (which would be a negative thinning rate)?
p 15:
line 27: 'around 2008' is probably more appropriate than 'since', as we did not put an exact date on it.
line 28: I am unable to reproduce the quoted ~35% number from the Hogg paper; 'part of the ice loss' might be more appropriate
line 30: space missing in 2010/2011exceptional?

Suggestions for the figures:
Fig. 2: The acquisition dates on the right axis might be illegible in the final version. You could consider to show only a few dates for each year with a larger font, and refer to the table for the other dates.
Fig 4 and Fig S7: with the current color scale it is hard distinguish positive from negative rates. A contour at zero values, or a color scale with a clearer separation between + and - (e.g., white around zero) might solve this.