The authors have generally answered the points raised by the reviewers, but this will still be rather a provocative paper, given the way the authors determine and use the RDR. Still, given the huge amount of variability in the various experiments summarized here, it seems to be as good a way of defining things as any other. Since there are bound to be continuing arguments about the use of such a parameter, I believe that a specific statement similar to that given in answer to comment 5 in answer to reviewer #1 on the importance of keeping things as natural as possible and comparing only data sets with “control” and “augmented” CO2 concentrations is needed. This should probably be either at the end of section 2.1 or in section 2.2.

If the authors can add such a statement, then the paper can be published.

Thank you for adding this paragraph. It certainly clarifies the reasoning behind your work.