Dear Yuzhe,

I have received now two new reviews from the same referees. As you can see their reviews are quite contrasted, but I think referee #2 is making an important point regarding the use of the upper part of the temperature profile for calibration. Can you provide a reply to that comments as well as a revised version that might include a discussion on this particular point. On the basis of this new material I will take a final decision.

Regards,
Olivier Gagliardini

Dear Yuzhe,

Based on the new version of your paper submitted on November 2017, I have now two reviews which indicate that indubitably this new version has improved. There is, nevertheless, still an important point raised by the first reviewer that should be addressed relatively easily with all the material you have already. Please consider his point and also the other remarks in both reviews to submit a new version of your manuscript from which I will take a final decision.

Best regards,
Olivier Gagliardini

Dear Yuzhe,

Thanks for this new version of your manuscript which can now be accepted in The Cryosphere. It has been a long process but I really think that the published manuscript has been greatly improved during the review stage.

Thanks for having choosing TC to publish your work.

There are few typos that I found when reading this last version that should be corrected before moving to the production stage.

Regards,
Olivier Gagliardini,

Technical corrections:
- when citing a book, like Cuffey and Paterson (2010), would be nice to give the page number to easily find the information
- you are using K and °C alternatively during the manuscript. It would be easier to keep to °C all along?
- page 13, line 24: and are not able to -> and is not able to
- In FiG. 5 it would be nice to indicate what are the default values for all the varied parameters? Does it correspond to the green curve (which looks the same in all panels?)?