the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Jan 2024
Nordic Seas Deep-Water susceptible to enhanced freshwater export to the subpolar North Atlantic during peak MIS 11
Abstract. Recent investigations into Marine Isotope Stage (MIS) 11 (424–403 ka), an unusually long and warm interglacial of the Quaternary Period, have found that the Atlantic Meridional Overturning Circulation remained strong while background melting of the Greenland Ice-Sheet (GIS) was high, and resulted in a fresh and cold surface ocean in the Nordic Seas. These investigations support the hypothesis that deep-water formation may not be as susceptible to future GIS melting as previously thought. Here we test this hypothesis and present a palaeoceanographic investigation of a freshwater-related abrupt climate event recorded in the eastern North Atlantic during peak interglacial conditions (~412 ka), when the GIS was as small or smaller than today. Using sediment core DSDP-610B recovered from the western Rockall Trough we reconstruct the evolution of Nordic Seas Deep-Water (NSDW) using benthic carbon isotope, Neodymium isotopes, and grain-size analysis paired with end-member modelling. Further, a combination of planktonic foraminiferal assemblage census and Ice-Rafted Debris counts allow us to reconstruct surface water properties including temperature and the movement of oceanic fronts throughout this event. Our results demonstrate that a reduction of NSDW only occurs once GIS melt and polar freshwater reaches subpolar latitudes. We hypothesise that the reorganisation of fresh and cold surface waters from the Nordic Seas into the subpolar North Atlantic was responsible for an AMOC-related cold event centred at 412 ka. Placing our results in the palaeogeographical context of the North Atlantic Region we tentatively propose that the ocean-atmosphere climate dynamics linking the Nordic Seas with the subpolar North Atlantic played and will play a crucial role for the stability of NSDW formation in the future, considering the enhanced melting and overall hydrological cycle at high Northern latitudes predicted for future climate scenarios.
- Preprint
(1427 KB) - Metadata XML
-
Supplement
(115 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on cp-2023-101', Anonymous Referee #1, 30 Jan 2024
Curran, M.J. et al., Nordic Seas Deep-Water susceptible to enhanced freshwater export to the subpolar North Atlantic during peak MIS 11. cp-2023-101
General Comments.
Basic premise set out on lines 151-178: The supposition that Feni drift is created and strongly influenced by WTOW is not supported by hydrographic data. Many authors have uncritically repeated the suppositions of Ellett & Roberts (1973), notwithstanding the fact that Dickson and Kidd (1987) had shown that Feni was controlled by Deep Water not the overflow in Rockall Trough. The base of the drift is at ~2500 m in northern Rockall Trough, deepening to >3000 m to the south. WTOW does not affect sediment transport at these depths, a requirement for focussing sediment into a drift on the Rockall margin. Overflow at the Wyville-Thomson Ridge has been supposed by some (e.g. New & Smythe-Wright, 2001) to contribute to deep flow along Feni Ridge, but later work demonstrated that this water (WTOW) mixes so intensely with surface water in its passage over Wyville-Thomson Ridge that it is not dense enough to flow along the bottom below 2200 m, the crest-depth of Feni Ridge at 56° N (which deepens to the south and is at 2417 m for ODP Site 610) (Johnson et al., 2010; 2017). Johnson et al (2017) note that zones of erosion in Northern RT “.... are seen over a depth range (800–2000 m) coincidental with that of deep WTOW ...” Below 2000m the recent work of Dubois-Dauphin et al., (2023) using Neodymium isotopes demonstrates that ‘deep WTOW’ lies above 2000 m while NEADW and LDW occupy the Trough bellow that. So WTOW is not a significant player below 2000 m. If there was a larger amount than present of interglacial freshwater the overflow would have been even less dense.
It is much more likely that Feni is caused by the cyclonic circulation of Lower Deep Water mixed with Southern Source water (traced by silicate concentration) in the Deep Northern Boundary Current (DNBC) of McCartney (1992). WOCE data in Kolterman et al., (2011) show that bottom water (Lower Deep Water, LDW) is about one third SSW farther south at 4500 m. This mixes with overlying Northeast Atlantic Deep Water (NEADW) and enters the cyclonic circulation in Rockall trough along the British Irish margin, exiting around the SE corner of Rockall Bank and Feni Drift (e.g. maps of Knutz et al., 2001, 2007).
Because the authors’ data have nothing to do with WTOW, the explanations and discussion must be recast in terms of the history of a more likely water mass, namely NEADW. As this contains some ISOW (which includes NSDW) from the S Iceland Basin (plus SSW), there may be elements of the authors’ arguments that remain applicable in a rewritten account. ‘WTOW’ pervades the Discussion which should be removed and the account recast in terms of more likely water masses.
Specific Comments
Intro is OK down to line 116 but needs editing as its too long at ~1500 words. It reads like a partially digested piece of Thesis, and >150 references are not all neccessary.
271-279. Cutting out the sand data above an arbitrary 211 µm on the basis of a paper by soil science workers who found a high coefficient of variation for their sand percentage results is ill advised. The cited work by Polarovski et al. records high CVs for sand which bias the results for the three samples (out of only 13) with sand percentage less than 10%. The reason for removal of the sand is apparently to avoid interference by presence of air bubbles, but if this were a universal problem nobody would ever make measurements of sand with a laser sizer anywhere. This is not the case.
280-289. An alternative to the end member system for assessing current-controlled sorting of sediment is the plot of sortable silt mean size versus percentage to assess sorting, most recently shown by McCave and Andrews (2019). The EM ratios can be conradictory. McCave and Andrews (2019) pointed out that EMs do not always discriminate well- from poorly-sorted records. Jonkers et al. (2015) proposed that their ratio EM2/EM1 provides a current proxy with no influence of IRD (they say ”… it is possible to correct for the contribution of IRD and obtain an estimate of changes in bottom current speed by using the ratio of EM2/ EM1 ….“). McCave and Andrews (2019) observe that in the case of very slow current and abundant IRD input, resulting in unsorted fines, this EM ratio is simply a grainsize indicator free of IRD influence, not a speed indicator. The ‘mean size ‘ in the range 7.64 to 66.9 mm (Fig. 6) is not far off the Sortable silt mean size range of 10-63 mm and a cross-plot of this mean size versus the EM ratio could show whether the EM ratio stands up as a flow speed proxy here.
308-313. This is Thesis intro style and not needed here.
318. BP is not appropriate here. 351-361 inc. table 2. This would be better put into Supplementary material. And do we really need to be told that it is a fortran 77 program ?
360 and Table 2. What is the parameter that defines ‘WTOW’ ? No mention has been made of this previously: If it is the d18Oben then it is not correct as WTOW is not at the bed now (and even less likely in the past if made less dense by meltwater). If it is the EM ratio then the same applies; it is not a record of the flow speed of WTOW, but perhaps of NEADW.
381 This ‘similarity’ is disputable. Fig 4 shows maxima at ~412, 410.5, 409.5 and a broad belt between 407.5 and 406
432-4. Gives justification for line 360 and Table 2, but too late. However as WTOW is not at the bed here this is an invalid statement. It cannot be a proxy for WTOW. Regretably this also applies to the published paper in CP by Holmes et al (2022).
Figure 6. Why is this on a depth axis rather than age as with all the other figures (3-6, 7, 8). It is impossible to correlate information with other figures.
546-56. This terrestrial discussion has only a tenuous connection to Hi-lat meltwater.
579-80. Fram Strait is the gateway into the Nordic Seas from the Arctic so channelling Arctic FW via Fram St would INCREASE its export into the Nordic seas.
603-4. Not NSDW export into RT.
607-8. The statement that log (Ti/Ca) data record surface ocean properties is disputable. Obviously it is a record of sediment properties, but given a balance between terrigenous and calcareous (ex surface productivity) components the statement may not be correct. The authors say it is a proxy for variations in lithogenic/biogenic inputs. Change could be entirely lithogenic, i.e. not surface ocean.
618. Dubois-Dauphin 2023 have data at depth of site 610 and deeper but others cited do not. (Citations repeated). See Fig #5 from D-D et al,’23
620. Most authors refer to this as SSW; Southern Source Water.
The length of the discussion, at nearly 3000 words for a brief excursion in MIS 11, seems grossly too much. Much of the ~900-word section on climate forcing and Ocean atmosphere teleconnections is not really key to the point at hand, analysis of a piece from 415 to 402 ka within the long (~424-374 ka, (LR04)) MIS 11.
Technical Corrections: typos, etc.A few in Refs: e.g. 1081, 1167,
References (New)
Dickson, R.R. and Kidd, R.B., 1987. Deep circulation in the southern Rockall Trough - the oceanographic setting of site 610. pp 1061- 1074. Init. Repts. DSDP, 94, US Gov. Printing Office, Washington DC,
Dubois-Dauphin, Q., Colin, C., Elliot, M., Förstel, J., Haurine, F., Pinna, R., Douville, E., Frank, N., 2023. The spatial and temporal distribution of neodymium isotopic composition within the Rockall Trough. Progress In Oceanography, 218, Art# 103127, doi: 10.1016/j.pocean.2023.103127
Ellett, D. and Roberts, D., 1973: The overflow of Norwegian Sea deep water across the Wyville–Thomson Ridge, Deep-Sea Res., 20, 819–835,
Knutz, P. C.,Austin, W.E.N. and Jones, E.J.W., 2001. Millennial-scale depositional cycles related to British Ice Sheet variability and North Atlantic palaeocirculation since 45 ka B. P., Barra Fan, U.K. margin, Paleoceanography, 16, 53– 64.
Knutz, P.C., Zahn, R., Hall, I.R., 2007. Centennial-scale variability of the British Ice Sheet: implications for climate forcing and Atlantic meridional overturning circulation during the last deglaciation. Paleoceanography 22, PA1207.
McCartney, M. S. (1992), Recirculation components to the deep boundary current of the northern North Atlantic, Prog. Oceanogr., 29, 283– 383
McCave, I.N., Andrews, J.T., 2019. Distinguishing current effects in sediments delivered to the ocean by ice. I. Principles, methods and examples. Quat. Sci. Rev. 212, 92–107.
Koltermann, K.P., Gouretski, V.V., & Jancke, K. (2011). Hydrographic Atlas of the World Ocean Circulation Experiment (WOCE). Volume 3: Atlantic Ocean (eds. M. Sparrow et al.), International WOCE Project Office, Southampton, UK, 221 pp.
Citation: https://doi.org/10.5194/cp-2023-101-RC1 -
AC2: 'Reply on RC1', Audrey Morley, 16 Apr 2024
The comment was uploaded in the form of a supplement: https://cp.copernicus.org/preprints/cp-2023-101/cp-2023-101-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Audrey Morley, 16 Apr 2024
-
RC2: 'Comment on cp-2023-101', Anonymous Referee #2, 31 Jan 2024
General comments
I am an observational oceanographer, so my review focuses on the modern-day oceanography, assumptions made and quality of the manuscript.
This manuscript clearly represents a large body of work. It has two main parts: firstly detailing how the surface properties have changed at the core site, and secondly linking this to possible changes in deep water and then drawing conclusions about the AMOC. I think the first part is fair and do not have any general comments on this. However, linking the observed SST change to deep water and the AMOC is, in my opinion, not clearly shown by the authors and I have concerns about this part of the manuscript.
I am not sure that the assumption that the sedimentary record at Feni Ridge is representative of changes in WTOW, and therefore the AMOC is fair because:
(1) We do not know whether WTOW is the bottom water mass at the Feni Ridge, and the papers the authors cite only shows its presence further north. Logically, I think it must flow south, but I do not know how large an influence it is at the Feni Ridge and whether it is the bottom water mass in contact with the feature. The authors could explore this more.
(2) Two reports not cited by the authors suggest that changes in the Feni Ridge record reflect a lateral redistribution of water masses. I think that the isotope work is interesting and goes some way to possibly indicating that this is not a lateral redistribution but I think this needs to be explored further.
(3) WTOW is the smallest component of the Greenland-Scotland overflow waters.
(4) WTOW flow into the Rockall Trough is very variable. If the changes at the Feni Ridge are due to WTOW, are the changes in WTOW representative of a change in the AMOC strength? Or is it more related to dynamics in the Faroe Shetland Channels changing the amount of overflow water entering the Rockall Trough rather than the Iceland Basin through the Faroe Bank Channel?
My other major concern is that the main finding of the paper and the title hinges on a single finding – that a high-resolution surface record shows a change at 412.29 ± 0.01 ka while the lower resolution grain-size analysis (which is attributed to WTOW) shows a reduction at 412.86 ± 45 ka. While these are outwith errors, a lower resolution SST record does not show the same offset. Nor I believe do the foram records. I’m uncertain whether the isotope records (εNd and δ13C), which are also used to infer WTOW changes, also show the lag. Additionally, I am curious whether if the grain size or isotopes were sampled at a similarly high-resolution, the offset between the surface and deep would still occur. I have pretty big concerns that the manuscript premise, title and large sections of the discussion are based on this single finding when others are contradictory.
As well as these scientific issues, I feel that the authors need to do some work to condense certain bits of the manuscript (e.g. the discussion, maybe some of the methodology and introduction) and improve the figures. For example, the captions do not match with the figures and there’s a lack of (a), (b) etc labelling of multiple panels making them hard to understand.
Major science comments
(1) This manuscript assumes that the sediment at Site 610 is representative of WTOW. I am not sure this is reasonable.
- Deep WTOW has been observed in the northern and central Rockall Trough hugging the western boundary (e.g. Johnson et al., 2017), but this deep WTOW has not been observed further south than 57.5 N. This is likely because studies have not examined further south than 57.5 N. The WTOW observed at 57.5 N must travel southwards, but the depth at which it is at is unknown.
- The authors slightly mis-cite the literature e.g. L176. Ellett et al. 1986 and Johnson et al., 2017 shows the presence of deep WTOW in the northern and central Rockall Trough. Neither show the presence of WTOW at the Feni Ridge latitude as suggested.
- WTOW may not be the water mass in contact with the seabed at Site 610. Data from the southern Rockall Trough show that the deepest water mass originates from Antarctic Bottom Water (e.g. McGrath et al., 2012, New and Smythe-Wright, 2001). This may not be true at the depths of the Feni Ridge but the authors need to look at this further.
New and Smythe-Wright, 2001, Aspects of the circulation in the Rockall Trough, CSR, doi:10.1016/S0278-4343(00)00113-8
McGrath et al., 2012, Chemical characteristics of water masses in the Rockall Trough, DSR, doi: 10.1016/j.dsr.2011.11.007.
(2) The manuscript also assumes that variations in sediment at Site 610 are representative of changes in WTOW strength. Two important missing references (Dickson and Kidd, 1987; Kidd and Hill, 1987) suggest that sedimentary changes at the Feni Ridge appear to be linked to the dominance of southern (i.e. AABW, NADW) origin waters rather than changes in the intensity of NSDW alone. These are reports and may have since been discounted, but I think the authors need to discuss this. Especially as it is fundamental to parts of the manuscript talking about deep water and the AMOC.
Dickson and Kidd, 1987, http://www.deepseadrilling.org/94/volume/dsdp94pt2_36.pdf
Kidd and Hill, 1987, http://www.deepseadrilling.org/94/volume/dsdp94pt2_48.pdf
This may be particularly important because, as the authors state, WTOW is also a variable water mass and well as variability in its flow speed (and therefore transport) there are periods when the water mass is not identifiable (e.g. Johnson et al., 2010).
Johnson et al., 2010, Wyville Thomson Ridge Overflow Water: Spatial and temporal distribution in the Rockall Trough, DSR, doi: 10.1016/j.dsr.2010.07.006
(3) A particular interesting area to me is the isotope work (εNd, δ13C, δ18O) which suggests that the sediments show the presence of a northern water mass. I think this part of the manuscript needs to be developed slightly. As you refer to δ18O multiple times I think this should be included on Figure 4.
(4) I also have some concerns about how representative WTOW is of variations in the AMOC. WTOW is only a small component of the AMOC lower limb and it is variable in nature (e.g. Sherwin et al., 2008, Østerhus et al., 2019). Changes in WTOW in the Rockall Trough may represent temporal variability in overflow at the ridge (e.g. due to dynamics in the Faroe-Shetland Channels) and a shift in the distribution of overflow water between the Rockall Trough and Iceland Basin (e.g. Stashchuk et al., 2011), rather than changes strength of the lower limb of the AMOC.
Sherwin et al., 2008, Quantifying the overflow across the Wyville Thomson Ridge into the Rockall Trough, DSR, doi: 10.1016/j.dsr.2007.12.006
Stashchuk et al., 2011, Numerical investigation of deep water circulation in the Faroese Channels, DSR, doi: 10.1016/j.dsr.2011.05.005
Østerhus et al., 2019, Arctic Mediterranean exchanges: a consistent volume budget and trends in transports from two decades of observations.
(5) As mentioned above in the ‘general comments’ I have concerns that the finding that the surface conditions change before the AMOC is based on the relationship between two records when they are of different temporal resolution and the result is not repeated in any other of the records examined.
(6) The authors define NSDW as ‘Nordic Seas Deep Water’. This is a term I’ve not come across before as in observational oceanography NSDW refers to Norwegian Sea Deep Water.
(7) At multiple points in the manuscript that authors refer to a ‘two-step event’. I find this confusing as the manuscript is focussing on the 412ka event whereas the second step appears to be at ~409ka. I suggest the authors consider changing the wording.
(8) More generally, I found that the manuscript needs to decide whether to focus purely on the 412ka event or also the 409ka event (or to focus on the wider temporal changes and then narrow down to 412ka). At times I felt it jumped around a little.
(9) The authors need to make sure to refer to figures/subplots at all appropriate points in the manuscript. This is sometimes missing (e.g. Sections 5.3 and 5.4).
Minor science comments
(1) L104: This needs rewording. Caesar et al. and Thornally et al. refer to present times while this sentence appears to be relating to MS11.
(2) L148-149 – while measurements of the AMOC in the North Atlantic began in 2004 (RAPID, with OSNAP post-2014), measurements have been made at the exit of the Labrador Sea in the Deep Western Boundary Current at 53 N since 1997 (e.g. Zantopp et al., 2017) and there have been long measurements of overflows at the Greenland Scotland Ridge (e.g. Østerhus et al., 2019).
Zantopp et al., 2017, From interannual to decadal: 17 years of boundary current transports at the exit of the Labrador Sea, doi:10.1002/2016JC012271
Østerhus et al., 2019, Arctic Mediterranean exchanges: a consistent volume budget and trends in transports from two decades of observations.
(3) L163-164: A more pertinent reference than Johnson et al., 2017 is Sherwin et al., 2008.
Sherwin et al., 2008, Quantifying the overflow across the Wyville Thomson Ridge into the Rockall Trough, DSR, doi: 10.1016/j.dsr.2007.12.006
(4) L164-165 – a more up-to-date paper looking at fluxes across the Greenland-Scotland Ridge is Østerhus et al., 2019.
Østerhus et al., 2019, Arctic Mediterranean exchanges: a consistent volume budget and trends in transports from two decades of observations.
(5) L168-169: Holliday ea 2000 and Ellett and Martin, 1973 are not appropriate to reference here as neither investigate whether the Feni Ridge is related to WTOW. I don’t think Ellett and Martin, 1973 mention the Feni Ridge – do the authors mean Ellett and Roberts, 1973? Holliday et al., 2020 cite this paper.
Ellett and Roberts, 1973, The overflow of Norwegian Sea Deep Water across the Wyville Thomson Ridge, DSR, doi:10.1016/0011-7471(73)90004-1
(6) The authors use WOA98 to reconstruct SST (L213). There’s been five releases of WOA since then – why have the authors not used e.g. WOA2018? Does this make any difference?
(7) L394-395: the wording suggests that G. glutinata is shown on Figure 5 but it isn’t.
(8) L401-410: I also see a big decrease in NP and the coiling ratio that isn’t mentioned.
(9) Section 6.1: It would aid the reader to briefly say where each core site is (e.g. eastern subpolar North Atlantic, eastern Nordic Seas etc) as well as referring back to Figure 2 (which you do sometimes but not always).
(10) L848-486, L493-494, L531: To me saying that Site U1305 is downstream of the East Greenland Current implies that it is directly influenced by it - which I don’t think it is. From Figure 2 this site appears to be more in the central Labrador Sea whereas the EGC flows down the eastern side of Greenland and then continues as the West Greenland Current flowing up the western side.
Comments on Figures
(1) Figure 4 needs improving
- the different colours mentioned in the figure caption don’t exist in the figure
- what are the shaded yellow vertical bars?
- the x-axis should be the same as other figures in the paper (e.g. Fig 6) to enable easy comparison between the two
(2) I was flicking between Figure 4 and 5 a lot. I think the subplots within the figures need re-organising. Adding SST to Figure 5 would aid the reader as the text compares the SST and foram records. The last two subplots on Figure 4 (εND and δ13C) aren’t referred to in the text until after Figure 5, the authors maybe better re-ordering the text, or changing how the figures are displayed.
(3) Figure 5 – the colours referred to in the figure caption again do not match those in the figure. I also suggest the authors mention in the figure caption when y-axes are reversed to aid the reader.
(4) Figure 6 – the figure caption is confusing – it is better to label the subplots and refer to (a), (b) etc. This is especially true if there are two different x-axes (such as on Figure 6).
- I think the IRD subplot is already shown on Figure 4 (?). If so, does it add anything to repeat it on this figure?
- why do you use depth rather than time as the x-axis on this figure?
(5) Figure 7 – please can the author check that all subplots within this figure are referred to within the manuscript?
- the IRD subplot is impossible to read because it is showing too many stations as solid bars. I suggest either using transparent bars, lines, or removing some stations.
- I think this subplot (and the caption) would again benefit from each subplot being labelled (a), (b), (c) etc.
- it’d be good to use more distinctive colours between the different subplots (if you chose to do this).
- please can the authors check that all the subplots are referred to in the manuscript?
(6) I felt I was missing seeing the δ18O timeseries, could this be added as a subplot to e.g. Figure 4? Or tell the reader in the text (not shown).
Technical comments
(1) L77 – write out CO2 in full first time
(2) L105-107: please check this sentence as it didn’t make sense to me!
(3) L162-163: …. via the Wyville Thomson Ridge…
(4) I thought Table 1 and 2, and maybe Figure 3 could maybe go in the SM as they don’t seem integral to the main manuscript to me?
(5) L366: do you mean i.e. rather than e.g.??
(6) L515-520: this feels out of place to me and possibly not needed.
(7) L533-536: Is a reference needed here?
(8) L541: define SLE
(9) L617-618: double reference
(10) L624: double 610…
(11) General – you have a lot of acronyms and I think some are unnecessary. They can make it harder for the reader, especially if they are non-standard ones. I recommend going through and removing any that aren’t needed.
Citation: https://doi.org/10.5194/cp-2023-101-RC2 -
AC3: 'Reply on RC2', Audrey Morley, 16 Apr 2024
The comment was uploaded in the form of a supplement: https://cp.copernicus.org/preprints/cp-2023-101/cp-2023-101-AC3-supplement.pdf
-
AC3: 'Reply on RC2', Audrey Morley, 16 Apr 2024
-
RC3: 'Comment on cp-2023-101', Anonymous Referee #3, 27 Feb 2024
Review report to “Nordic Seas Deep-Water susceptible to enhanced freshwater export to the subpolar North Atlantic during peak MIS 11” submitted to Climate of the Past by Curran et al.
The manuscript investigates the causes and consequences of a meltwater event in the North Atlantic during MIS11. The authors created new data records of planktic foraminiferal assemblages, XRF, particle size distributions, authigenic epsilon Nd, and benthic foraminifera stable isotopes across the event from Site DSDP 610 at the southern tip of Rockall Bank in the Northeast Atlantic.
The authors interpret the proxy changes they observe to show a deep water current slowing and reduced presence of WTOW at the core site following the freshwater event, and interpret decreased benthic stable carbon isotope data as an increased presence of southern ocean waters. Notably, they interpret differences in the surface water proxies such as reconstructed temperatures and faunal assemblages across the North Atlantic region to indicate that fresh surface waters propagated from the western Nordic Seas into the western North Atlantic and then to the East. They reconcile their findings by assessing that North Atlantic Subpolar Gyre Dynamics are essential for the AMOC behaviour and the freshwater only affected the AMOC by suppressing Nordic Seas deep water formation once it reached the North Atlantic Current entering the Nordic Seas.
I think the manuscript investigates a very important topic that is a good example of how paleoceanography can inform us about climate dynamics and potential future climate evolution. The study site is well suited is in an interesting location and presumably well suited for proxy reconstructions, and the proxy observations appear of high quality. Generally, most observations fit the overall picture of proxy records across the region, supporting their viability. I have a few open questions and concerns about the interpretations, detailed below. While the writing is generally clear, I found it not easy to follow the interpretations, mainly because 1) the figures are not easy to read and 2) many findings from other studies are mentioned but not shown (which, of course, is not always possible). I suggest to improve on these two points wherever possible and added some ideas about the figures at the end.
My main concerns are:
1) The interpretations about changes in the deep circulation. I do not necessarily object the interpretations, but they could be better supported. Several points are listed in the detailed comments below.
2) The sequence of events occurring at the site currently seems insufficiently supported and visualised. An important argument the authors raise is that deep ocean changes occurred before changes at the surface. I am not yet convinced this delay is significant. Maybe showing the records vs. sediment depth or detailing sediment depths, ages, and proxy record changes together with uncertainties could help?
3) The figures could be improved and restructured to support and guide the interpretations better. See also below.
4) The precision of the numbers given appears often too high to be justified and higher than necessary, and generally inconsistent.
Detailed comments:
l. 120ff.: How certain is it that the site is bathed mainly and directly by WTOW, and not e.g. by Lower Deep Water (LDW) or North East Atlantic Deep Water (NEADW)? In their Fig. 1b, Crocker et al. (2016) show a section of Rockall Trough with different water masses indicated, which would suggest that Site 610 lies deeper than WTOW today. While it may still have been pathed by northern sourced waters, a dominance of NEADW might make the discussion of the observed deep water changes more complex.
l. 245: Is Wu et al. 2015 the best citation for the demonstration of the viability of foraminifera-bound Nd as paleo-circulation tracer and the methods? One well cited (review) paper about this is Tachikawa et al. (2014), so why not cite them, or an earlier paper incorporated there? Also, does the method follow the Tachikawa review? If not, it might be interesting to point out significant differences.
l. 453 and rest of the MS: check consistency of epsilon Nd notation (i.e. the epsilon font family and Nd as subscript or not)
l. 488 ff.: I cannot find evidence of SST at Site 1305 of 10°C in Fig. 7. If I interpret the figure correct, then SST ranged between ~ 4 and 8 °C. Similarly, the described drop in temperatures is not obvious. Furthermore, in the face of those temperature uncertainties, I suggest to round all given numbers to the full °C, including at many other text paragraphs.
l. 507: I understand that the (determination of the) timing of the proxy transitions is limited by the record resolution, but not why this means that the durations are maximum estimates.
l. 543: I don’t understand what “-38.7 to +3.9m” of sea level equivalent mean. These should be positive numbers, I reckon? Again here, it might be worth to round the numbers.
l. 544 f.: I don’t understand this sentence. Please clarify how much ice these estimates imply.
l. 560: Please use consistent precision in numbers, e.g. here the age and its uncertainty.
l. 566 f.: The IRD concentration numbers could also be rounded.
l. 578: It is not quite clear what anomalous means here. Is it unlike the Holocene? Or colder than the surroundings? Or colder than earlier? Also, should it mean IN the Nordic Seas?
l. 586: “and therefore 412 ka“ should be “and therefore date to 412 ka”, I think?
l. 596 ff.: Please add references to figures.
l. 605 ff.: The age uncertainties seem very small. Do the authors indeed claim that e.g. the error of 10 years in line 608 is defendable? Maybe adding (or focussing on) the sediment depths of the different discussed signal onsets helps making this discussion more convincing?
l. 611 ff.: Could it be that the changes were brought about by changes in the subsurface to deep ocean and then propagated to the cryosphere and the surface ocean?
l. 614 ff.: I think the authors should be very careful when stating Nd isotopic signatures of water masses during MIS11 when they do not cite proof from contemporaneous archives. It is well established that end member Nd isotopic signatures of different water masses have varied through time, often in accordance with long term climatic changes. Hence, citing modern Nd isotopic values for MIS 11 water masses may be interesting in the face of a lack of better information, but must be taken with caution and this should be reflected by the manuscript text. This applies to other parts of the manuscript, too.
l. 623 ff.: The reductions in benthic δ13C have indeed been interpreted as increased presence of SOW before. However, I am wondering about two aspects:
1) Do the carbon and neodymium isotope values fall in the mixing polygon of the different described end members (i.e. WOTW, LSW, LDW)? Do the changes add up to allow for pure changes in the mixing ratios of these water masses? Or are additional processes necessary to explain the evolutions in these two proxies? A cross-plot of these two proxies may already answer these questions.
2) Galaasen et al. (2020) also analysed a few samples on their B/Ca ratios, which relate to the seawater carbonate ion concentration. In their Fig. S6 B they show the combined evolution of carbon isotopes and B/Ca across the low-δ13C events at Site U1305. They interpret “the association of high (low) C. wuellerstorfi δ13C with high (low) C. wuellerstorfi B/Ca” as an indication of NADW replacement by nutrient rich SOW. However, this association is arguably very weak, with an R² = 0.2 across their data. I am wondering whether these data do not rather argue against a strong incursion of SOW, but rather other geochemical processes, leading to a (northern sourced?) deep water with high B/Ca and low δ13C ? Maybe the process is similar to those observed by Yu et al. (2008) during the Last Glacial?
l. 676 ff.: The authors cite Eldevik et al. (2014) for “since strong deep-water formation in the Nordic Seas requires Atlantic inflow and open-ocean convection”. However, in the same sentence they state that “A strong AMOC seems at odds with the western Nordic Seas covered by meltwater”. Eldevik et al., however, explicitly state that the relationship of open ocean convection with the THC (or AMOC) is not straight forward, and that “It has in particular been observed that a previously inferred causality (Hansen et al., 2001) between northern deep ventilation and dense overflow from the Nordic Seas does not hold (Olsen et al., 2008)”. I think it is important for the whole manuscript to include the premise that this link is not as strong as often assumed. This may be a purely semantic issue.
Fig. 4: Please describe what the yellow bars indicate in the caption. Could they be used also for Fig. 5 to facilitate a temporal comparison?
Fig. 5: It might be helpful to indicate the color-site specifications in the figure or a legend.
Figs. 5 & 7: It is very hard to see in the figures what is described in the text, and in general to decipher the different records in the figures. Some suggestions on how to improve on this (need to be tested to see their usefulness):
- use the same site-specifi colors (and symbols) for all records from each individual site, and also use these in the map in Fig. 2. It might also be good to make these colors systematic, e.g. red-blue depending on latitude (and another color for Site U1305?), or same/similar colors for same regions.
- mention the general location of each site, e.g. Arctic, Rockall Bank, Reykjanes Ridge, Labrador Sea, somewhere near the sites in the plot or the caption; or alternatively indicate the latitudes.
- same or very similar proxies could be plotted on the same axis
- add legends
- annotate figures in the figure panels
- thicker data lines and possibly larger symbols
- vertical bars to point out certain time periods, such as SST cooling at site 610B.
- indications of proxy uncertainties (as error bars next to the records, not necessarily for each data point)
- focus on showing the data described in the text and omit (or move to supplement) other, apparently less relevant data
Fig. 6: Please indicate that these data are from Site 610 from this study. And should the x-axis not show the age instead of core depth?
All figures with proxy records: I suggest to indicate the event (or its sub-phases) with a vertical bar to ease the interpretations.
Figures: I have the feeling that a figure summarising the most important findings across the region as records would be helpful. For example, it could show records of NPS or SST at different locations in the same panel, then current speed and epsilon Nd at Site 610, and then benthic δ13C at different sites in another panel all as one stacked record, with the critical time of the meltwater event clearly marked with a vertical bar. The many individual plots showing records may be important to show the diversity of proxy records, but one core figure as a summary would surely help the reader to follow the discussion.
References mentioned:
Crocker, A.J., Chalk, T.B., Bailey, I., Spencer, M.R., Gutjahr, M., Foster, G.L., Wilson, P.A., 2016. Geochemical response of the mid-depth Northeast Atlantic Ocean to freshwater input during Heinrich events 1 to 4. Quaternary Science Reviews 151, 236–254. https://doi.org/10.1016/j.quascirev.2016.08.035
Yu, Jimin, Henry Elderfield, and Alexander M. Piotrowski. “Seawater Carbonate Ion-Δ13C Systematics and Application to Glacial–Interglacial North Atlantic Ocean Circulation.” Earth and Planetary Science Letters 271, no. 1–4 (July 2008): 209–20. https://doi.org/10.1016/j.epsl.2008.04.010.
Tachikawa, K., Piotrowski, A.M., Bayon, G., 2014. Neodymium associated with foraminiferal carbonate as a recorder of seawater isotopic signatures. Quaternary Science Reviews 88, 1–13. https://doi.org/10.1016/j.quascirev.2013.12.027
Citation: https://doi.org/10.5194/cp-2023-101-RC3 -
AC1: 'Reply on RC3', Audrey Morley, 16 Apr 2024
The comment was uploaded in the form of a supplement: https://cp.copernicus.org/preprints/cp-2023-101/cp-2023-101-AC1-supplement.pdf
-
AC1: 'Reply on RC3', Audrey Morley, 16 Apr 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 498 | 213 | 47 | 758 | 51 | 35 | 34 |
- HTML: 498
- PDF: 213
- XML: 47
- Total: 758
- Supplement: 51
- BibTeX: 35
- EndNote: 34
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1