the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Jun 2026
Deep-water hydrographic variability in the Northeast Atlantic during MIS 4
Abstract. Understanding the evolution of deep ocean circulation and chemistry over the last glacial cycle is key for elucidating the ocean’s role in modulating atmospheric CO2 changes on millennial and orbital timescales. Marine Isotope Stage (MIS) 4 is a key paleoclimatic interval of the last glacial inception for assessing the role of the deep-ocean carbon storage in driving atmospheric CO2 levels, because it is characterized by a large decrease of air temperature and a rapid atmospheric CO2 drop of ~40 ppmv, likely linked to changes in ocean circulation, and includes several millennial climatic events, for example Heinrich Stadial (HS) 6. Although previous proxy-based studies have suggested a weakened Atlantic overturning during MIS 4, and particularly HS 6, basin wide changes in circulation remain poorly constrained. Here, we present high-resolution deep-water hydrography reconstructions from the mid-depth Iberian Margin (core MD01-2444, ~2.65 km water depth), based on benthic foraminiferal Mg/Ca-derived deep-water temperature (Tdw), benthic δ18O, and associated δ18Odw records, alongside SS- based flow-speed record. Our data reveal three distinct deep-water hydrographic ‘regimes’ at the Iberian Margin: 1) warm “interglacial-like” mode during MIS 5a (including Greenland stadials C19 and C20); 2) a colder glacial circulation mode during early MIS 4 (pre- HS 6); and 3) a “Heinrich” circulation mode during HS 6. A stronger influence of colder southern-sourced waters is inferred at the Iberian Margin throughout MIS 4 and intensifies during HS 6, whereas MIS 5a stadials are characterised by a pronounced subsurface warming. The warm stadials C19 and C20 were likely driven by a southward displacement of convection sites in the North Atlantic and appear to be unique to MIS 5a as they are not observed during MIS 3 Dansgaard-Oeschger events at this site. We also find evidence of “millennial-type” variability and, similar to recent studies for MIS 2, a persistent contribution of northern sourced waters during MIS 4 (pre- HS 6). The contrasting impacts of similar millennial-scale DO-type variability on North Atlantic deep-water hydrography and circulation strength during MIS 3, 4, and 5 highlight the strong dependence on the background climate state.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Climate of the Past.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(2835 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2026-3334', Anonymous Referee #1, 07 Aug 2026
-
RC2: 'Comment on egusphere-2026-3334', Anonymous Referee #2, 16 Aug 2026
This paper presents a new sortable silt record and a new benthic foraminifera Mg/Ca-derived deep water temperature record from MD01-2444 on the Iberian Margin from MIS 5a and MIS 4. This is a well-studied core site and region, and the authors utilize both previously published and newly reported data to make the valid and substantial conclusion that three different deep-water hydrographic regimes occurred throughout this interval. The scientific methods are sound, but aspects of the methods section should be revised according to the comments below. The assumptions and uncertainties associated with the data are outlined for the Mg/Ca measurements and temperature calibrations, but not are not discussed for the sortable silt flow speed records. The discussion of the assumptions and uncertainties associated with the Mg/Ca data and temperature calibrations could also be improved for clarity. See the comments below. The authors should also choose distinct symbols to represent standard deviation and standard error, as they currently use sigma for both. The authors should consider changing the title of the paper to include both MIS 4 and late MIS 5, which would be more representative of the content. The authors did not provide any supplemental material and indicate that the data associated with the paper will be uploaded to public repositories, but it is currently not available for review.
Section & Figure Comments:
Section 2: I recommend adding a map figure and/or section plot of modern oxygen at the Iberian margin that shows the vertical distributions of these different water masses, with the depth and location of MD01-2444 labelled. That will make the paper more accessible to those unfamiliar with this region.
Section 3.1. Chronology: How different is the newly developed chronology from the original? This could be good to include in a supplement. Also, I believe the age model tie points and information should be reported in a supplement, repository, or pre-print prior to publication to make sure the figures are reproducible? There may be an exception if the other publication mentioned is in review elsewhere, but since the tie points are shown in Figure 1 maybe those could be reported in this study.
Figure 1. Could you include the other tie points from previously published studies as well? Between 65 and 55 ka, MD01-2444 and ODP 1057 and 1055 seem to be poorly aligned. Why would that be the case? Also, to make the plot easier to read I would suggest putting all of the MD01-2444 planktic d18O labels on the same side of the plot (the left side) and coloring the y-axis labels by the same schemes as the time series.
Table 1. Please provide the complete ODP sample codes (Site-Hole-Core-Section-Top Offset) for the tie points, in addition to meters composite depth.
Section 3.4. Trace Metal Analysis: You later mention measurements of Al, Mn, and Fe concentration in addition to Ca and Mg. I’m assuming you measured those all by the same method, but you should mention this explicitly. Also please report the associated uncertainty or relative standard deviations of replicates measured for those, if you have them.
Figure 2: Round the values for the equations and the r squared values, and report the p-value.
Section 3.6.1. Calibration: You mention two calibrations, but only use one. Could you add some material to the supplement that shows how some of the reported Tdw values change between the calibrations? It Is unclear where the 2% difference comes from. If that is demonstrated in a previous study, you could also cite that.
Section 3.6.3. Uncertainty Modeling: This section was difficult to follow. You alternate between referring to sigma as the standard error and the standard deviation. I would prefer if you only used the symbol sigma to refer to standard deviation, and used SE as a short-hand for standard error. Switching between reporting 1 sigma and 2 sigma values is also confusing (such as in Line 264). Since you mostly report 2 sigma, I think you should stick with that in the text. Also please modify Lines 268-270 for clarification. I think you mean that the error bars in subsequent plots represent only the error from the data (with the MC modeling) and not the error from the calibration, but that isn’t clear right now.
Figure 3. Add a legend to describe what the error bars and shading are for each plot. I think it is confusing that you use the shading both to show the distribution of MC simulations and the uncertainty. Can you plot the MC simulations as individual points? The figure caption is difficult to understand at the moment, and a legend on the plot would help.
Section 3.6.4. d18Odw and d18Odw,anom reconstructions: What is benthic d18Oanom? Can you explain what this is and how It was calculated in the text?
Figure 4. Somewhere on the plot or in the figure caption you should indicate that all of these records are from MD01-2444.
Section 3.7. Sortable Silt: The uncertainty associated with Tdw is carefully considered in the previous section, but the uncertainty associated with the sortable silt measurements is not discussed. Can you provide any estimates of the uncertainty associated with the reported values?
Figure 5. Is the decrease in the running correlation around 60-65 ka due to the presence of IRD during HS6? Why do you think the two methods disagree between 70-75 ka? It looks like a sampling resolution bias to me.
Section 4. Results: You mention comparisons to the Holocene here and elsewhere, but where is the Holocene data? Could you include that comparison in a plot, or mention specific temperatures?
Figure 6. Do you comment on why the Tdw signal is so much smaller in HS6? How does it compare to the magnitude of the changes across HS during deglaciations?
Section 6: Conclusions: A schematic figure of the three circulation regimes would be a great addition to this paper.
Additional line-by-line comments:
5-10: Using the phrase ‘is key’ twice, consider rephrasing one of the sentences
16: Write out ‘sortable silt’ instead of using the abbreviation here.
20-25: What is millennial-type variability? This may be too vague for the abstract. Could you be more specific about the type of variability?
35-100: This section provides a good overview of the prior work on North Atlantic Ocean circulation from the LGM, but doesn’t go into any detail about prior work that has been done using the sortable silt proxy. I think it is worth including a bit about this in the introduction, since it is a large part of your study.
109-110: Eastern North Atlantic Deep Water (ENADW) or Northeast Atlantic Deep Water (NEADW)? You mix the initialisms and the names here.
138-140: Please write out the full name of GICC05, before using the short-hand.
142: ODP should be written out as Ocean Drilling Program upon first occurrence in the paper. I think there typically is no hyphen between ODP and the Site name, so ODP-1055 should be ODP 1055 or ODP Site 1055.
158-159: What Is the sedimentation rate over your interval in MD01-2444 according to your age model? This is the more relevant sedimentation rate for this study.
161: Please write out Globobulimina affinis upon first mention in your paper.
172: wuellerstorfi is spelled incorrectly here.
189-191: The structure of this sentence is confusing.
195: I suggest removing ‘(i.e. inefficient cleaning)’.
196: Should this just be ‘coating’ rather than ‘carbonate coating’?
365: The absence of any signal in d13C outside of HS6 is interesting. Do you have any ideas about why that might be the case?
383: Should this be ‘southern-sourced water (SSW)’? You refer to SSW in the next line.
410-415: You switch between 2 sigma and 1 sigma here. Stick with 1 or 2 for both, otherwise it is confusing for the reader. If 2 sigma is 2 degrees C and the difference between Tdw is 3-5 degrees C, I wouldn’t say that is a ‘much larger’ difference, maybe just ‘larger’.
445-446: Including the quotes around ‘stagnation warming’ and ‘collapsed’ makes it seem like you are referencing some other text. If you are, include the reference at the end of this sentence. If you aren’t, take the quotes out.
477: Is ‘glacial NADW’ a typo? Should it just be ‘NADW’?
563-564: Can you clarify what d18Oanom is in the methods? I get what d18Odw,anom is, but what is d18Oanom?
570: How does benthic d13C compare between the LGM and MIS 4?
616: You say ‘intermittent to SSW’, should there be some other word after ‘to’?
Citation: https://doi.org/10.5194/egusphere-2026-3334-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 86 | 33 | 14 | 133 | 8 | 9 |
- HTML: 86
- PDF: 33
- XML: 14
- Total: 133
- BibTeX: 8
- EndNote: 9
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
The authors present a rich dataset to document deepwater circulation variability during MIS4 obtained with robust methodologies. They suggest that despite different ice volumes, MIS4/3 and MIS1/2 transitions show similarities and that similar DO-type variations might have different impacts on deep-water circulation. However, I do have some comments that I would like the authors to address before considering their manuscript for publication. My comments are as follows:
1) I suggest you take a look at the new paper by Hughes et al. (2026) about stadials-interstadials: https://www.sciencedirect.com/science/article/pii/S0277379126004038#sec6
2) You speak about many water masses. I think you should display them on a map along with a temperature-salinity transect.
3) MOW expansion can reach down to ~1800-1900 m below its lower core range at ~1500 m as diluted parts in some locations. You can check it with GLODAP data. You suggest that because its core range is 500-1500 m, MOW has not affected your site (lines 122-123, line 500), which is ~2.65 km deep, but in line 611, you suggest your site was affected by subsurface warming. That part, I think, you should reassess thoroughly.
4) Trace elements - in continental margins, regardless of cleaning methodology, high Fe values can be observed, possibly due to sediment input (https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023GC010885). The samples you might have potentially discarded as contaminated may not necessarily be contaminated.
5) 3.6.3 Tdw Uncertainty Modelling - I find this section overwhelming. The calibration equation already comes with an error margin. I do not see the point of this extra analysis. If you wish to keep it, I suggest moving the details to supplementary material, including the additional plots.
6) Line 301-307: If you think there are uncertainties caused by RSL, why did you apply this method? You also do not discuss what these uncertainties are or why the 6ka window is any better than the previous method. If you wish to keep RSL calculations, move them to supplementary material.
7) Line 390-391: Brine rejection - this comes out of nowhere and needs a reference. Plus, some studies argue against brine rejection (i.e. doi.org/10.3402/polar.v27i2.6172, https://doi.org/10.1002/2015PA002815). Hence, your interpretation in line 486 may align with earlier interpretations in previous studies.
8) Line 399: "mode of operation of deep-water formation" - what does it mean?
9) Line 487: "However, the origin of this warming remains debated." - needs reference.
10) Lines 610-611: "mode ... of subsurface heat budget variability" - what does it mean?
11) There are typos such as Mccave, Carrrara, and there is also a typo in Fig. 3c.