Dear authors,

Thank you for your careful and detailed reply to the two referees comments. On the whole, you have been able to take into account the comments that they have made, and indicate that there are ways that you can address their concerns.

A key area of concern for both reviewers is how you explain the climate driving mechanisms for your data. You make a number of suggestions in your author comments which indicate that you have considered ways to address these, and to make some quite detailed changes to the manuscript. There is a relatively large number of corrections or clarifications indicated by reviewer 2, which you have noted can be addressed. In order to evaluate whether these revisions do address the reviewers concerns, please incorporate your proposed changes into a revised manuscript and upload for further consideration.

Best wishes,
Erin McClymont

Dear authors
thank you for your careful replies and changes in response to the comments from both reviewers. The manuscript reads very well, provides a clear discussion of your data and the implications of the results, and I am satisfied that you have addressed the reviewers concerns. I have some very minor corrections to suggest for the final version of your manuscript:
- page 1 line 28-29 in the tracked changes document (in your abstract) “The increased humidity... likely resulted from decreased Antarctic sea ice, which led to southward retreat...”. Reviewer 2 was cautious about the links you were making directly to Antarctic sea ice. To counter this, I would recommend that this sentence is rearranged so that you start with the local/regional mechanism first, and then lead to the sea ice as a possible driver. For example: “The increased humidity... likely resulted from a southward retreat of the westerlies ... , which would be consistent with a decrease in Antarctic sea ice” (can you also clarify if you mean “Antarctic sea-ice extent”?).
- Likewise, in lines 23-26 (tracked changes) you state “...due to increased Antarctic sea-ice extent”: could you be more cautious and note “which we infer was linked to increased Antarctic sea-ice extent”?
- Likewise, can you also apply the principles of the two comments noted above to your conclusions section? In the conclusions you still put Antarctic sea ice first, as a “known” change, but it would be better to first link your data to the local/regional and then hypothesise a link to the sea ice data. Note that the sea ice data also has uncertainty in its interpretations, so it would be better to have a strong local/regional interpretation with the sea ice viewed as a potential hypothesis to be tested.
- line 55-56 (tracked changes): can you state “to the present day” and then remove the comma after “present day”?
- line 61 (tracked changes): “modelled LGM precipitation patterns for the region are highly variable...” (add ‘patterns’?)
- a minor point, but the annotation text on your figure 4 (e.g. “more C3”) is very small and quite difficult to read. Can you increase the font size for these?
- lines 401-405 (tracked changes): can you edit the i, ii, iii etc to take into account the change in numbering from removing “ii) less precipitation”
- lines 414-415 (tracked changes): perhaps some of the later confusion with the (sea surface) temperature/isotope links (noted especially by reviewer 2) might be mitigated if here you say “consequently changes in air temperature from the LGM...”.
- line 446-447 (tracked changes): to emphasise your inter-related precipitation/ET control, could you add to this sentence “..., which operate in the same direction and are not easily disentangled” (or something similar?)
- line 561 (tracked changes): correct “Chevalier”

Dear authors

Many thanks for making the corrections and for replying to my comments and those of the reviewers.

Note that there is one very minor technical correction to be made: during the final upload of this version of the manuscript equation 1 seems to have been lost in formatting. Can you please upload a revised pdf of the manuscript where this equation is visible?

Best wishes,
Erin McClymont