The revisions proposed in the response letter go a long way to address the critiques raised by reviewers and a duly revised manuscript is correspondingly welcomed. Adding the nutrient dynamics to the simulation system, and comparing those results to the original C-dynamics-only simulations, will add a good deal of value to the paper. The paper's contributions will also be deepened by including text from the response letters in the paper's results and discussion sections, where appropriate.

The work has the potential to make a valuable contribution, providing useful insights into the model mechanisms that govern terrestrial ecosystem responses to elevated CO2. However, as noted when handling the BG_2018_153 version of this manuscript, care must be taken to ensure that the paper's major interpretations and conclusions are framed to accurately match the methods and findings. My understanding of what this study mainly shows is that beta factors vary (a) across hierarchical levels of C-fluxes and stocks, and (b) across PFTs in a way that is largely attributable to variations in LAI dynamics across PFTs. This finding is framed as if it sheds light on across model spread in beta but that is not correct. The methods and results do not pinpoint LAI as a leading source of across-model spread or uncertainty in the land carbon cycle response to elevated CO2. The authors attempt to make a conceptual and motivational link between the across-model spread in beta factors shown in other studies and the findings presented in this paper however this is misleading. Consider this - it is possible that all of models in some MIP would show the same kind of across-PFT variation in beta, and that those models still having large across-model disagreement in the overall beta factors for different ecosystems and for the global biosphere. Some of the response comments still seem to misunderstand this point. Again, the paper has a contribution to make, particularly with the revisions in response to reviews, but I would reiterate that the introductory framing, interpretations and conclusions may still need to modified to match its methods and findings if the paper is to be ultimately accepted for publication.

The revised version of this manuscript addresses many of the major concerns raised by reviewers. This assessment is supported by a follow-on review which also offers some suggestions for minor revisions. Before issuing a final acceptance I encourage the authors to consider incorporating these suggested revisions in a final version of the manuscript that I will review quickly once more.