Dear Dr. Hernández-Almeida and co-authors,

First, apologies for taking so long to arrive at an editor decision. Partly this was caused be the required handover from editor Andrea Dutton to me. The rest is due to me not finding the time to go through all documents for a full overview.

However, I finally have read the manuscript, the reviews, and your author response. The referee reports had been relatively favourable. You addressed some of the points the referees mentioned and rebutted others. I think the manuscript is close to being acceptable. However, there are many small points I would still like you to address and one major point that relates to key point of publications, i.e. how well the data support the stated findings and conclusions. I would therefore ask you to revise the manuscript once more.

Thank you, with best regards,
Thorsten Kiefer


* THE MAJOR POINT *

Even after your revisions I still share the concern of both referees who noticed that the relationship between IRD, T, and d18Osw is not as clear in the data as the text might imply. The revised text has changed a little to mention some exceptions, but you don’t provide any statistical analysis. Moreover, figure 2 is not very transparent. I tried to follow what the vertical red-yellow bars are supposed to mean. The Fig. 2 caption says they “indicate IRD discharge associated with subsurface warming.” This is too vague and subjective, and accordingly was done in an inconsistent way. For example, around 960 ka the bar seems defined by the d18Osw, at 980 ka by the IRD peak. As a minimum increase of transparency I ask therefore you to re-draw the vertical bars in a consistent way, either hinging all on d18O or all on (probably more logically) IRD, so that at least readers can assess more easily for themselves how robust the described relationship is. Figure 3 needs to be adjusted then as well to be consistent with Fig. 2. Moreover, as some semi-quantitative statistical visualisation I suggest that you colour-code the new table 1. You can, for example, colour fields in red where a significant warming occurred together with the IRD. You could even distinguish between a light red and darker red depending on how significant the warming occurs. Same with d18Osw and d13C. If doing such a significance analysis, please explain exactly how you did it, so that one can see that the analysis is clear, objective, and reproducible.
For inspiration, I recommend to look at Lang and Wolff, 2011, CP http://www.clim-past.net/7/361/2011/cp-7-361-2011.html



* MINOR BUT RELEVANT POINTS TO ADDRESS *


ABSTRACT

p1 line 17: remove or fill the empty "()"

p1 line 18 "at Site U1314": add location info like "from a sediment core from Gardar Drift in the subpolar North Atlantic" in stead of site name, or in addition if you think the site number is relevant in the abstract.

p1 line 21 "temperatures and salinities": Where? *subsurface* I assume, but please be unambiguous by saying so.

p1 lines 22/23 "during periods of weaker Atlantic Meridional Overturning Circulation (AMOC) reduction": I guess you mean either "weaker AMOC" or "AMOC reduction" but NOT "weaker AMOC reduction".


TEXT

p2 lines 9/10 "insulating effect of extensive ice-shelves": Not sure what is meant here. How can the ice shelves themselves insulate? What do they insulate from what from what?
line 10: Replace "through" with "to".

p2 line 11 "in the dynamics of the North Atlantic subpolar gyre that controls the meridional heat and salinity transport": That sounds incorrect or at least imprecise. The meridional transport is controlled by the MOC, the subpolar gyre alone just circulates it in the northern North Atlantic. Maybe you mean the transport to the high-latitude northern North Atlantic or to the Nordic Seas? Please me more precise.

p2 line 19/20 "This mechanism involves the coupling of the 20 AMOC with ice-sheet dynamics, by an increase of the heat and salt export from low latitudes”: This is the key hypothesis you are testing with data. Following the concern of Referee#2, I ask you to be clearer in describing the mechanism hypothesised. This does not need to be elegantly formulated, but clear. The missing link at the moment seems to be why a weaker AMOC should necessarily result in heat and salt export from low latitudes (in the subsurface layer, I suppose …). Either explain the hypothesised mechanism or give an indication that it is observed but not understood.

p2 line 23 “for abrupt climate events such as Heinrich and Dansgaard-Oeschger (D-O) cycles”: This is the first time you mention Heinrich and Dansgaard-Oeschger (D-O) cycles. Two remarks: They should be mentioned in paragraph 1 already. And: cycles are not abrupt. And in fact Heinrich events are not occurring in a particularly cyclic way. So maybe better something like: ”…for abrupt climate shifts such as those associated with Heinrich and Dansgaard-Oeschger (D-O) events”.

p2 line30/31 “However, only few paleoceanographic studies using these proxies have been produced (Jonkers et al.,
31 2010b;Peck et al., 2008;Peck et al., 2006).”: In fact plenty of paired Mg/Ca and d18O studies have been produced. Therefore, please either specify this statement or delete it.

p3 line 2: “out of” does not work. Replace with “other than” or “beyond”.

p3 line 3 “paleo-community”: Avoid such sloppy language in a paper. Paleoscience community or paleoclimate community or paleoceanographic community …

p3 line 7 “high-quality/resolution”: No lazy writing please. This is no ratio, therefore avoid the /. Reformulate and or replace.

p3 line 11 “similar structure transitions”: Do you mean “similarly structured transitions … line in the D-O…”?

p3 line 13 “evaluate”: You don’t evaluate, you study or specify or constrain or elucidate …

p3 line 19: “temperatures”: remove the plural-s

p3 line 25 “preceding”: Change to “prior to”.

p3 line 26 “indicating clear evidence”: Either a suggestive “indicating” or a firm “providing (clear) evidence”, but no mixture, please. My own suggestion would be to definitively avoid the “clear” because you are inly making inferences, not studying the ice shelf yourself.

p4 line 1 “rates average 9.3 cm/kyr from 1069 to 779 ka”: Decide on one abbreviation for year, either a or y or yr, and use throughout.

p4 line 28 “a record d18O of seawater ”: Change to “a record of seawater d18O”.

p4 line 31: replace the with “a” in “the reductive cleaning step” or specify “as originally introduced by …”

p8 line 30 “Atlantic inflow”: Unclear: what inflow exactly?

p9 line 1 “salt inflow in the North Atlantic”: As above, this is unclear. What inflow exactly?

p10 line 11/12 “destabilize ice-shelves creating pulses of ice-rafted debris”: In this description of the scenario please add the logical step of iceberg discharge. The mentioned IRD is only the expression of the discharged and melted icebergs. If ou keep IRD in here, note that no pulses of IRD exist, but layers of IRD or pulses of IRD release (or pulses of IRD deposition).

p10 line 13 “suddenly released”: Released to where?


CONCLUSIONS

The conclusion contains a lot of “would” formulations, which are not suitable here. Please formulate the conclusions you are confident to put forward. Moreover, please reconsider whether the last half-sentence referring to “the 100-kyr climate cycles of the Late Pleistocene” is a good and well-formulated conclusion. It is the first mentioning of the 100-yr cycle in tis entire paper. I think it should either prepared and discussed somewhere earlier in the paper, reformulated in the conclusions chapter, or deleted.


TABLE, FIGURES

Table 1: Consider colour-coding as I suggested. In caption be clearer what the number mean.

Fig. 2: Place the letters A-F more clearly so that they are unambiguously assigned to a curve or panel. Very confusing as it is. And in caption say what the green and grey sections mean in the benthic 13C curve.

Fig. 3 Caption: Mention what is “specific” about the intervals chosen.


SUPPLEMENT
Please (1) add a Table caption that also contains more metadata so that the table becomes more meaningful if used without the paper at hand. (2) Makes sure that the final version of the table fits the width of one table (in the current version the last column is on a separate page).

Dear Dr. Hernández-Almeida and colleagues,

Thank you for making the latest minor revisions to the manuscript. I note that you took on board most of the recent suggestions and feel that the manuscript is now ready to become a valuable contribution in Climate of the Past. I am therefore happy to accept the paper in its present form (just make sure that LR04 has a zero in it, not the letter "O"). Congratulations to you, thank you for your patience related to the change of editors and my own pace, and thank you to the referees for reading through the manuscript and providing their useful comments.

Kind regards
Thorsten Kiefer