Thank you for your revised paper, which has been seen by one of the original reviewers and myself. The paper is almost ready for publication but there are a few minor issues that still need attention:

Page 4, line 5. "This assumption is supported..". Please explain how the accumulation rate in Yiou et al was derived, if not by 10Be (otherwise this statement is a circular argument).

Page 4, line 7. "framework of the Ph D of Cauquoin". I don't feel this statement is appropriate, unless the Ph D is published in which case give the reference.

Page 12, line 5. The discussion here is a little confusing. The SD is 8.8% and that is useful to know, but this doesn't address the question whether the constant flux assumption is correct or not. For example if there is a higher flux when the snow accumulation is higher (because there is some limited wet deposition) then you r calculation cannot address that. Please revise the wording a little to acknowledge the potential for this to be a problem.

Page 14, line 15. The names you use for the models are differnet to those on the figures (here you use the institute name rather than the model name in fact), making it hard for the reader to follow. Please standardise on one set of names.

Page 15. You note that ECHAM accum-delD agrees with your data; howeverECHAM is way off the line for acc vs T_s. This implies that either there are two compensating errors in ECHAM (in which case the agreement is luck) or that T_S is estimated badly in the data. Please be clearer on this point.

Also page 15. Since ECHAM gives a good acc-delD relationship, you can't then say that the models that track it closely in acc-T_s space (MIROC, IPSL)"are not yet accurate at simulating changes in pptn". If that is so then ECHAM is also bad, but actually (as per last para) ECHAM is good at reproducing the actual data. An equally likely statement is that the models that track ECHAM are good at accumulation (but that there is either a problem with derived T_s or the models are bad at temperature). This would equally imply that the other models (which match the black line) are not necessarily good at accumulation. As written the discussion is quite misleading.

Same comment for page 16, line 24.
Same comment for abstract, page 2, line 25.

Fig 2 caption needs some work to explain the different panels (especially I don't follow what panel b is).

Thank you for the changes you have made, especially to the part about the models.

I am now ready to accept the paper, subject to some minor technical corrections listed below:

Page 4, line 7. In the latest version you have removed the statement "primarily by dry deposition". However this statement is crucia: if the method of deposition is not primarily dry then the 10Be method will definitely not work. Please reinstate this phrase.

Page 4, line 24. I still find this way of referring incorrect, as it fails to send the reader to the citation. Please change from "measured in the framework of the PhD of Cauquoin" to "measured, as alreday reported (Cauquoin 2013)". This then correctly sends the reader to the correct citation.