You are encouraged to submit a revised version of the paper reflecting the changes described in your Author Comments of the open discussion. To summarize some key requests:

Please add the new figure (drafted in response to Reviewer 1) summarizing the average C increment and transfers between forest pools and with the atmosphere. Include corresponding new discussion similar to that in Author Comment 1. Please also explain why dead wood and litter shows such a large reduction over the time period when living biomass shows such a large increase and soil carbon is relatively stable.

Briefly note in the manuscript specific information about what this study offers beyond previous work (as drafted in the response to Reviewer 2).

Address all specific comments of reviewers as promised in the Author Comment responses to individual reviews.

Reduce and focus text wherever possible.

Address minor English errors throughout. Also work to improve text flow as suggested by Reviewer 3.

RC2, L40: include this explanation in a revised version.

RC,2, L138-139: clarify along the lines of the response.

As discussed in AC3, add at least some descriptive comparison of the approaches taken by CMB26 and EFISCEN.

As discussed in AC3, consider adding discussion of the possible limitations of having adopted administrative, country boundaries to organize the modeling effort rather than ecoregional or physical/biological attributes (as discussed in AC3).

The revised version satisfactorily addresses all of the reviewer's concerns. AS a final step prior to acceptance I ask that you please make the following minor revisions.

L257 - 259: Edit "probably influenced by the (average) effect of the natural disturbances occurred during 2000-2012" to clarify what about natural disturbances changed during this interval or relative to the recent past to cause the negative balance for dead wood and litter. Clarify if there were fewer or more natural disturbances during the 2000s relative to prior decades.

L259-260: This sentence seems to suggest that fire is the dominant pathway by which DOM is released to the atmosphere, rather than heterotrophic respiration. Please correct this. Even if disturbances have killed material and transferred it to DOM, and even if fire has been the dominant disturbance type (has it? - probably not), fires typically lead to less direct emission as combustion and more indirect emission from decomposition of disturbance-killed material (e.g. Ghimire et al. 2012 JGR-B).

L264-267: If this negative Delta Soil C value is due to the transfer of C from forest to non-forest rather than release to the atmosphere, Fig 2 displays this incorrectly by omitting carbon transfers to non-forest lands and instead displaying all of it as release to the atmosphere. This must be corrected.