Dear Authors
Thank you for taking the time to address the reviewers concerns. I appreciate the effort you have made to improve the paper and note that the amount of work requested was considerable.
I have read through the response in detail and feel you have addressed the majority of the issues. However, given the recommendations from the reviewers of a major revision, I feel I must ask for additional input. I therefore recommend a second review.
I hope the original reviewers will confirm that they are satisfied with the change and I will be able to accept the paper soon.
Kind regards
Liz

Dear Cameron O’Neill,

I have now received two Reviewers’ reports of your revised manuscript. They acknowledge the improvement in this version, but still raise a few concerns. I invite you to submit a revised version, taking into account their comments as well as mine detailed below. Looking forward receiving your revised manuscript.

Regards,
Laurie Menviel


1) In general, I think the text could be tightened as there are a lot of repetitions throughout the different sections (e.g. P33, L9-10 is similar to P5, L. 28-31).
I would also reiterate Reviewer 1 comment to tighten Section 2 by more directly describing what you have done.

2) Data analysis section: I am not sure what the goal of this section is, and I think it needs to be amended. I think it provides an incomplete review of all the changes in the carbon cycle occurring during the last glacial-interglacial cycle. This data is used to constrain the model, so has to be introduced in section 2, however the processes leading to changes in these variables across the last G-IG cycle should be compared to your results in the discussion.

3) Please make sure to define the acronyms (GOC, AMOC, NADW, AABW…) in their first appearance and then use them throughout.
Please be consistent in the way you refer to figures, and remove the parentheses around the panels of figures in the text: i.e. change Fig.X(A) into Fig.XA throughout.

4) Supplementary material: Currently only Table S8 is expressly called in the manuscript. Please make sure the supplementary material includes only appropriate material and that each table and figure are called in the main text when necessary.

Line by line comments:
P 5, L. 10: remove the comma after “study”
P 5, L. 13: remove the commas before and after “at 100m depth”
P5, L. 28: Please remove the reference to Menviel et al. 2012 here, as the land-atmosphere CO2 fluxes were imposed in that study.
P7, L. 15: Please move this information to the code availability section.
P8, L. 13-14: I don’t think this sentence is necessary.
P8, L. 31: Please remove “serve to”
P12, L. 20: Please add concentration to “carbonate ion” and define as [CO32-]
Table 2: Change last row to: [CO32-] as deduced from B/Ca
P15, L. 7: Please replace “littered” by a more appropriate verb.
P18, L. 2: It seems you are showing ocean – atmospheric D14C, no?
P19, L. 9: “its”
P20, L. 3: It might be good to add the depth that you are referring to here, and also add a reminder about the fact that there is no distinction between north and south in the abyssal Atlantic (i.e. 1 box).
P20, L. 4: “its”
P21, L. 1: “shows”
P23, L. 3: How can you be so sure this is just the result of CO2 fertilization? As CO2 decreases so do temperature and precipitation.
P25, L. 4,5: “Last Interglacial”
P26, L. 10-12: I don’t see the link between this sentence and the previous ones.
P26, L. 35: remove “and” before “rich”
P26, L. 35: Add “enhanced” in front of “biological”
P27, L. 2-5: I am not sure this sentence is really correct. I would suggest to revise and correct the reference issue.
P27, L.8: Please be more specific here, and add processes and direction of changes instead of the repeated “other changes”.
P28, L. 16-17: Please remove “observed during the last glacial termination” as this was not studied in Menviel et al., (2015), which was a broad sensitivity study (whereas the last glacial termination was studied in particular in Menviel et al. (2018)). You can reformulate the sentence as follows “in the context of past changes in atmospheric CO2 and d13CO2.”
P28, L.20-21: What do you mean here by “but also quite a variation across …”?
P30, L. 26: “Last Interglacial”
P 30, L. 30: Replace “cooler SST” by “lower SST”
P30, L. 33-34: What do you mean here by polar sea-ice? Only changes in Antarctic sea-ice cover are included in Figure 14, and from Figure 14, an Antarctic sea-ice increase always seems to lead to a small (2ppm?) pCO2 decrease.
P 34, L. 13: Please replace “d13C negative CO2” by “13C depleted CO2”
P 34, L. 16-17: The grammatical structure of that sentence is odd. Please rephrase.

Figure 14: Please remove “proxy” after “Antarctic sea-ice”

Dear Cameron O’Neil,

I have now read your answers to the Reviewers’ comments as well as your revised manuscript. In principle I would be pleased to accept your manuscript for publication in Climate of the Past after minor revisions. Some suggestions are detailed below.

Best regards,
Laurie Menviel

As a general comment, I would suggest to read the manuscript carefully, try and tighten the text by avoiding repetition or information that is not directly relevant. The parts of the text describing the terrestrial biosphere and changes in peats and permafrosts are usually hard to follow. Also I would suggest to be more precise when describing the timing of changes that occur in proxy records. I think you should use “during” to mention changes occurring during a certain period and not “at”: e.g. atmospheric CO2 decreased during MIS 4.

Please use the appropriate tense throughout the manuscript, and make sure you do not switch tense without a reason. For example, the tense of the Methods consistently varies between present and past tense (Methods should be in present tense).

Ideally, the order of the material in the supplementary information should follow their mention in the main manuscript (which is not the case for Table S13 for example). Please consider changing the order of the material shown in SI.



P1, L.1: please amend as follows: “of changes in atmospheric CO2 concentration”
P1, L. 13: Please move the part on MIS5b after MIS5d and thus before MIS4.
P2, L. 2: “in the concentration of atmospheric CO2” (and elsewhere as you feel appropriate as it the concentration of atm. CO2 that changes in time).
P2, L. 31: “biogeochemical”
P3, L.10-11: “A second phase of 40ppm CO2 drawdown between 72 and 65 ka..”
P3, L. 17: “and other proxy records presented in Kohfeld & Chase, 2017, covering the last glacial cycle.”
P3, L. 32: “We use”
P6, L. 17: delta sign.
P6, L. 20: “with respect to”
P6, L. 28: “we undertake” and “biological parameters”
P6, L. 29: “We target” (and please use present throughout the methods as sometimes you switch to the past).
P9, L. 33: “the terrestrial biosphere carbon reservoir and its d13C signature”
Table 2: For atmospheric CO2 and d13C, please make sure the correct references are included: Bereiter et al., 2015 paper focuses on 600 to 800ka whereas your simulation focuses on the last 130ka.
Figure 3 legend: Throughout please simply state “Figure SX” and remove “Supp. Info” in front.
Please remove L. 20-28, p14.
From p15, L. 13 to p16, L. 1-2: Please add appropriate references.
P16, L.2-3: Please be more precise as to when the d13Catm decreases occur.
P16, L. 21 and 22: “between the LGM and the Holocene” or MIS 2 and MIS 1 (but not the reverse).
P18, L.6: Please replace “North Atlantic overturning circulation” by AMOC.
P19, Legend of Figure 6: Please amend as follows:
“DD14C is ocean minus atmospheric D14C” (with the correct Delta sign).
“rather DD14C represents the difference between each ocean data point and the contemporary atm. 14C value.
P20 L. 7, please change to “MIS3, then increases from 16 Sv to 29 Sv between MIS 2 (the LGM) and the Holocene.”

Figure 10: Please move the labels so that they do not cover the data.
P24, Legend of Figure 11: “DD14C is ocean minus atmospheric D14C”
P24, L.6-7 – Figure 12 and legend: How did/can you correct NPP to take into account permafrosts and peats?
I can’t see how this can be justified and would thus suggest to remove that part. I also note that both your NPP and the estimate from Hoogakker et al. do not take into account permafrosts and peats. It might be worth emphasing in the manuscript that there is only one box for the terrestrial biosphere and that the changes in NPP in that box are a simple function of pCO2.

P26, L.2-7: This is an almost word for word repeat of P. 14, L.30 through p15 L. 1. Please amend.
P28, L. 14: “atmospheric”
P28, L. 18: Please remove “in the experiment”
P30, L.5-6: “enhanced Southern ocean biological productivity could account for”
P32, L. 13: The correct ref is Menviel et al. 2012.
P34, L. 4-6: This sentence is not really correct as in Menviel et al 2016 the impact of these different processes on pCO2 was not studied. Menviel et al., 2016 showed that the oceanic d13C, and D14C records were most consistent with a weak GOC and AMOC. Menviel et al., 2016 further showed that this weak oceanic circulation would significantly increase the deep ocean carbon content (and thus significantly contribute to the pCO2 decrease).
P34, L. 8: You should add a sentence of caution here stating that your analysis looks at the mean MIS state. The oceanic circulation was highly variable during MIS3. Your result could also be an artefact of averaging across MIS 3 D-O variability, and HS6.
P34, L. 34 and P35, L.1: I don’t think that is the way to put it. Temperature and precipitation exert a dominant control on the vegetation type and will thus impact NPP. This is simply not taken into account in your model.
P36, L. 21: please remove “transient”