Dear Authors,

Both referees find your work of significant interest for publication in Biogeosciences, but concur that the lack of validation data is troubling. I understand the difficulty associated with validating such model projections however, and your proposed approach is I think sufficient to allow readers to judge the level of confidence that should be given to projections in different regions/PFTs. I would suggest that in future work you consider the suggestion of Reviewer 1, to use the observed sensitivity of vegetation cover over the satellite observational record to inform modeled projections of potential future change.

Best wishes,

Trevor