Dear Holly and Christina,

Thanks for your careful reply to both referee. I strongly encourage you to submit an improved version of your manuscript that I will send back to review.

Regards,
Olivier Gagliardini

Dear Holly and Christina,

I have now received two new reviews based on the revised manuscript of your paper. Both reviewers acknowledge for the improvements since the previous version and are globally happy with this new version. They have both nevertheless made a number of remarks that could still be followed to improve your paper. I invite you to submit a revised version of your manuscript based on these two reviews. Based on this new version, I will take a final decision regarding the acceptance of your paper in The Cryosphere.

Best regards,
Olivier Gagliardini

Dear Holly and Christina,

Thanks for this new version of your manuscript and your careful reply to this second round of reviews. I think this new version has greatly improved. I have nevertheless still few minor revisions that should be done to your manuscript before a final decision is taken regarding its acceptance to The Cryosphere. These minor revisions are listed below.

Best regards,
Olivier Gagliardini

- The caption of Fig. 1 is not precise enough: In (a), you should mention to what refer the white dashed line (limit of the FE model domain?). I would suggest to add a square in (a) showing the limit of (b). Also, "grounding zone" should be "grounding lines". In (b), you should specify what is the white dashed line and yellow lines.

- In Fig. 1, on the north part of the shelf, the model domain seems to cut the ice-streams. Why this choice? What type of BC do you apply at the model domain boundary (observed velocity?)?

- line 161: 1000 years is a very long relaxation? Model might reach some steady state already after such a long relaxation period? Usually relaxation periods are of few years to few decades?

- line 321: Negative response of which variable? (The mean instantaneous velocity change over the whole ice-stream is, however, negative?)

- caption Fig. 9: Positive values indicate faster flow for the perturbed model without SCIR and ...

- line 447: may be also conclude that an inversion alone is not sufficient for transient prognostic simulations and that appropriate evolution equations for the ice rheology and the basal friction would be required to have a correct evolution of these fields induced by evolving flow conditions?

Dear Holly and Christina,

Thanks for these last corrections. I am happy to accept your manuscript for final publication in The Cryosphere. Thanks for having choosing The Cryosphere to publish this nice piece of work.

Regards,
Olivier Gagliardini