|This is my second review of the manuscript by Jakob et al. with a focus on the response to both reviewers to allow final submission of the manuscript. The authors have extensively answered all the issues and comments made by the reviewers providing clear arguments how to deal and include the issues into the manuscript. If this is indeed done following the response the manuscript will be ready for final publication. |
The two main points that will especially need to be dealt with and on which I have added some specific comments below are the comparison with other studies, and also the point of what crassaformis habitat really means. This did not show up as a big issue in the first consideration of the manuscript but based on the reviews I feel that this deserves some good attention. Which signal is really recorded by “commonly-referred-to- as thermocline dwellers” as crassaformis, dutertrei, tumida and several others and to what degree can their records really be compared to each other? I do not see this as a critical issue but rather as an opportunity to point out the differences seen here, which may involve comparing apples and pears rather than paleoceanographic change?
Specific comments to responses:
R. 1.2, 1.3 and 2.2
Comparison with the other locations as mentioned in the response would be very helpful. More specifically, the Site 1241 was always considered as representing a more global signal rather than a regional signal as the longterm trends are very similar to the LR04 stack. The glacial-interglacial closure of Panama was more affecting the Caribbean Site 999 as current flow was from the Pacific into the Caribbean. I think the difference in thermocline records may be more related to the differences between deep/permanent thermocline of crassaformis vs seasonal thermocline for dutertrei (see also comment to R. 2.6). Obviously, creating records of crassaformis at Site 1241 and dutertrei at Site 849 may be very interesting in future work.
Agree. But it would make sense that tumida shows warmer temperatures than crassaformis as it still lives a lot shallower, see also your figure 1b.
I would agree that the full consideration on Mn/Ca does not need to be included into the revision; Mn/Ca is a complicated but also very promising proxy after it has been resolved what all the different inputs are and how the signals are extracted. So in that sense thinking a bit further on any new Mn/Ca data brings things forward. But I would include the reduction experiment to show that at least part of the higher Mn/Ca was caused by Mn-oxyhydroxides. My comment to the in-situ, primary signature for the Mn/Ca was in this case more to the Mn, which gets actively incorporated into the primary calcite of crassaformis, so that would mean that it may record changes in oxygen in the deeper thermocline possibly related to an oxygen minimum zone.
Comparison with dutertrei or tumida would maybe not make much sense as these calcify much shallower. The very low temperatures may be due to the calibration specifics, i.e. which specimens were used to make these calibrations in comparison with the ones you used? The Globorotalia species all form a thick crust later in their life cycle which is formed deeper in the water column and depending on “habits” people may pick/select either fully crusted or non-crusted specimens. Compare for example the Cleroux calibration for G. inflata with the Groeneveld and Chiessi (2011) one, there is quite a large offset between them.
So, if this high percentage of crassaformis is the true signal because dissolution did not play a significant role, it may be interesting to find a modern analogue in a database like MARGO which could then provide a good indication for the conditions during this time as they are obviously quite different then from today at the location.
R. 2.6 and 2.7
I wonder if this discussion on if there were thermocline depth oscillations on glacial-interglacial timescales may not be a matter of definition. The habitat of crassaformis deeper in the water column at the base of thermocline reflects the permanent thermocline, which under modern conditions is not showing much variability and may also not have done so on G-IG cycles but would respond to longterm trends in oceanic conditions, like was also shown in the tropical Indian Ocean with the Karas record. The G-IG changes seen in many dutertrei, tumida or obliquiloculata records are all showing variations in the seasonal thermocline. I would not be surprised to see such changes at this site also for these species, although that may be outside the scope of this study.