the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Oct 2025
A model intercomparison of radiocarbon-based marine reservoir ages during the last 55 kyr including abrupt changes in the Atlantic Meridional Overturning Circulation
Abstract. Changes in the marine reservoir age (MRA) of the surface ocean are important information used for radiocarbon dating of marine sediment cores or archaeological artifacts. MRA changes are expressed relative to the atmosphere, and as such are dependent on the prevailing atmospheric radiocarbon calibration curve. The most recent estimate for evolving global average MRA for latitudes approximately < 50° is incorporated into the marine calibration curve Marine20, which was directly calculated from the atmospheric ∆14C record, IntCal20, using the carbon cycle box model BICYCLE, taking into account observed changes in the carbon cycle. These simulations did not consider changes in the strength of the Atlantic meridional overturning circulation (AMOC) related to Dansgaard/Oeschger and Heinrich events. A recent study using the successor BICYCLE-SE suggested that abrupt AMOC changes would lead to changes in MRA of less than 100 14C yr in the non-polar surface ocean (about < 50°). To better support previous model-based MRA and to further constrain the impact of AMOC changes on MRA, we here assess transient simulations of the last 55 kyr performed by two Earth System Models of Intermediate Complexity (EMICs), LOVECLIM and Bern3D, and compare them to the published BICYCLE-SE box model results and previous output from the LSG ocean general circulation model. The setups within this MRA model intercomparison (MRA-MIP) are not identical, but all models are forced by atmospheric CO2 and ∆14C to have the surface ocean carbon cycle state as close as possible to reconstructions. Simulations with abrupt AMOC reductions during stadials display a rise in in MRA in the surface northern Atlantic (>50° N) and the deep Atlantic, for example during Heinrich stadial 1 of 300–1250 and 500–1300 14C yr, respectively, roughly in agreement with their reconstructed rises by about 1200–1300 14C yr. We find that the changes in the mean non-polar surface MRA (< 50° latitude) during abrupt AMOC changes in LOVECLIM are also in the order of ±100 14C yr, while in Bern3D simulating changes are up to ±200 14C yr. While the models tend to agree that a reduced AMOC leads to lower MRA in the low-latitude surface ocean, under some conditions the opposite is found (e.g. simulations with LOVECLIM across Heinrich stadial 1). Spatially resolved results of the models show that changes in surface MRA during stadials depict the general pattern of a radiocarbon bipolar seesaw (older surface water in the high north, younger in the high south and in the Indo-Pacific), in agreement with previously published reconstructions, but with model-specific details in the non-polar Atlantic. Throughout the last 50 kyr, the change in the multi-model mean in non-polar MRA of the 2 EMICs when compared with Marine20 is less than 100 14C years and within the uncertainties of Marine20. Furthermore, changes in the MRA of the high latitude Southern Ocean (> 50° S) are extremely model-dependent and for most times between 18 and 43 kyr BP the changes in the multi-model mean MRA are larger than the 95 % confidence interval of the non-polar MRA depicted in Marine20, making the construction of a similarly numerical model-based calibration curve for this region a challenging task.
Competing interests: LM and LCS are members 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
(7457 KB) - Metadata XML
-
Supplement
(38228 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2025-5136', Patrick Rafter, 13 Dec 2025
- AC1: 'Reply on RC1', Peter Köhler, 13 Jan 2026
-
RC2: 'Comment on egusphere-2025-5136', Anonymous Referee #2, 23 Feb 2026
General comments
In the paper the authors present a model intercomparison of radiocarbon-based marine reservoir ages (MRAs) over the last 55 kyr BP. The previous estimate of the global averaged MRA for the non-polar regions (Marine20) was obtained with the box model BICYCLE. However, in this previous estimate abrupt AMOC changes that occurred during the last glacial period were not considered. Here, the authors make use of two models of intermediate complexity (Bern3D and LOVECLIM) that can represent abrupt AMOC changes when forced with freshwater flux anomalies. They compare simulated MRA changes to published results obtained with the box model BICYCLE-SE and to outputs from an OGCM (LSG), and to reconstructions. The study aims to explore the robustness of model-based MRA estimates, in particular the sensitivity to AMOC abrupt changes. A second objective is to underline the regional variations of the MRA changes, in particular the authors present MRA changes in polar regions (that are not covered by Marine20).The author found that (a) including AMOC changes produce non-polar MRA changes with magnitudes of the order of 100 and 200 14C yr (LOVECLIM and Bern3D, respectively) (Figure 3). (b) The multi model mean (of the two EMICS) of the non-polar MRA is close to that of Marine20 throughout the last 55 kyr BP, with less than 100 14C yr difference (Figure 7). Regarding polar regions, (c) the polar MRA north of 50°N is larger than Marine20 during the simulated Greenland stadials, the polar MRA south of 50°S is larger for most of the time period in comparison with Marine20 (Figure 7). This underlines the limits of using a global calibration curve for specific regions and time periods. (d) During the simulated Greenland stadials, the MRA changes in the polar regions depict a bipolar seesaw behavior (Figures 5, 7).
I find the study of high-interest and relevance for the community. It is well written and well organized. The motivations of the study are well justified and the results are clearly presented. The figures are also very clear, especially given the relatively large amount of information that is presented. Please note that I am not an expert in biogeochemistry or MRA. My research focuses more on climate modelling and the abrupt events of the last glacial period.
I have only a few comments/remarks that could be addressed before the paper is published. I hope that these are helpfull. First, I think that the introduction could be improved by adding some basic information and background knowledge about marine reservoir age. For example, the term is never defined in the introduction. Then, regarding the experimental setup, I think providing more information on the simulations involving additional FWF as well as on the glacial simulations for the LSG would make the paper more self-contained (in the sense that readers would not need to read other articles to understand the setup). To finish, I find that the mechanisms described to account for the ΔMRA changes could be more detailed. In particular, more emphasis could be put on the mechanisms responsible for ΔMRA decrease and increase in sections 3.2 and 3.3 (around lines 261 and 293). I think that clarifying the causal pathways underlying these changes would substantially improve the readability and interpretation of the model results.
Specific comments
Abstract:
- line 16: “in agreement with their reconstructed rises by about 1200-1300 14C yr”. I find this sentence a bit confusing. While Bern3D and LOVECLIM show MRA rises in the surface North Atlantic and in the deep Atlantic, Figure 6 suggests that the reconstructed MRAs do not exhibit such rises during HS1. For example, in Figure 6a, the reconstructed MRA appears relatively high between ~27 and 16 kyr BP and then it decreases. I therefore suggest rephrasing this sentence.
- lines 22-23: “but with model-specific details in the non-polar Atlantic”. I find this sentence a bit vague. I suggest removing it or precising the details within a few words.Introduction:
- I find the introduction very clear, but I think it could benefit from background information on MRA. I suggest adding a few lines to define and provide the reader with basic knowledge about MRA. Maybe between lines 32-33?
- line 40: Please define piston velocity within brackets.Methods:
- line 93: “since this agrees best with 14C reconstructions in the deep Atlantic ocean”. Is there a reference for this? Otherwise, the phrasing can sound speculative. I suggest adding the reference of a previous study or rephrasing the sentence.
- line 111: “meltwater of up to 0.3 Sv is added into the North Atlantic”. I wonder where exactly the freshwater flux is added. Can you indicate the latitudinal and longitudinal extent of the corresponding area? Also, I suggest adding “flux” after “meltwater”.
- Regarding Figures S1,S2,S3 and Figure 3: How is the AMOC strength defined exactly? Is it the same definition for all models? If yes, I suggest adding the definition to the caption of Figure 3.
- line 128: Same as for LOVECLIM, where are the freshwater fluxes added?
- line 142: What does 'constant ocean circulation' mean? Does it mean that all oceanic parameters are identical in the nine simulations? I suggest adding some lines to explain how the glacial scenarios are achieved and how a weaker AMOC is obtained. Also, would it be possible to have the equivalent figure of Figure S1 to S3 but for LSG?
- line 177: Do you have ideas for a particular effect?
- line 180-185 (Figures S1 to S3 and Figure 6): How many cores are included to produce the time series for the North Atlantic surface and Atlantic depths for the data? It may be interesting to know?Results and Discussion:
3.1 Pre-industrial MRA compared to GLODAP
- line 201: Regarding Figure S5, why is it only the JFMJAS months that are used in Bern3D to display the mean MLD?
- line 203: of the “mixed layer” depth range?
- line 217: Is 500 14C yr a multi-model mean? If not, could you be more specific and provide the values for each model? I think that this could be relevant here, see my following comment.
- line 218: ”with BICYCLE-SE and LSG showing predominantly larger values, …”. It does not seem to be the case looking at Figure 2. For instance, LOVECLIM presents larger values than LSG in the Atlantic at 2 kyr BP.
- line 222: “a similar offset to GLODAP is missing in the also 1000 m deep northern high latitudinal boxes, pointing to an alternative explanation”. I am not really convinced because almost all the BICYCLE-SE boxes seem to have the same surface MRA value at 2 kyr BP, so it could still be a box effect?
3.2 MRA variations across the last glacial period
- line 234 (and 236): What is meant by adjusted? Were the applied freshwater fluxes adjusted beyond what is described in section 2.3 (and 2.2)? If yes, maybe the freshwater flux scenarios could be added to the Supplement Information.
- line 244: Could you please specify the flux corrections that are mentioned here? Maybe between brackets?
- line 253: “show no consistent pattern”. I suggest to precise the sentence because Figure 3f shows that the differences between Bern3D (PallSTD) and Marine20 with respect to 2 kyr BP, as well as the differences from BiCYCLE-SE (A3), decrease during stadials.
- lines 260-261: Can you add a few lines to precise the mechanisms explaining why the global average surface MRA and the mean deep ocean 14C would vary in opposite directions, following AMOC weakening and eddy diffusivity reduction?
- line 279: What are the radiocarbon distributions in the models? Are they similar? Can they be easily compared to observations? I feel that this could be worth exploring in future studies or that this could be a perspective (?).3.3 Details of the impact of AMOC weakening on MRA
- line 293: Could you please briefly describe how the MRA increases in the North Atlantic are connected with reduced MLDs at this location. Also, the acronym MLD was not defined in the main text.
- line 308 (Figure 6): From what I have understood, the LOVECLIM and Bern3D outputs for the entire northern surface Atlantic (+Arctic) and deep Atlantic are compared to the reconstructed time series. I wonder how the comparison would be improved (or not), if the colocated-to-the-data model outputs were compared to the reconstructed time series.3.4 Toward global MRA
Conclusion:
- line 361: in LOVECLIM “in the surface northern Atlantic”? I suggest specifying the location because it seems that Bern3D is closer to the reconstructions elsewhere (Figure 5; PallSTD) and in the deep Atlantic (Figure 6; Pnofwf), during HS1.
- lines 361 to 366: I would suggest inverting the last two sentences of this paragraph.
- line 371: I would suggest removing “mini”? I feel that “mini” minimizes the impact of the study.Technical corrections
- line 14: There are two “in”.
- line 38: There is a coma instead of a point at the end of the line.
- line 135 and 301: Isn’t there a syntax issue with these sentences? Please note that I am not a native english speaker.
- I also find that some phrases are a bit long, and maybe the paper would benefit from splitting some of the sentences into two? e.g., lines 25, 45, 71.Citation: https://doi.org/10.5194/egusphere-2025-5136-RC2
Data sets
Model output from LOVECLIM and Bern3D Peter Köhler, Laurie Menviel, Frerk Pöppelmeier, Timothy J. Heaton, Edouard Bard, and Luke C. Skinner https://my.hidrive.com/lnk/EhfGOFikK
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 415 | 132 | 37 | 584 | 122 | 80 | 81 |
- HTML: 415
- PDF: 132
- XML: 37
- Total: 584
- Supplement: 122
- BibTeX: 80
- EndNote: 81
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Summary
Although I’m not a modeling expert, but I am a user of these datasets and this MIP manuscript is timely and well-written. Given all that I’ve learned in this manuscript (e.g., the “surface ocean” is not consistent between the models), the observation that these models are broadly showing similar values (at least for the lower latitudes) is pretty amazing. This MIP is mostly focused on large-scale observations, with a mostly qualitative discussion of the inter-model differences and/or comparisons with observations. But maybe that's fine for this first of its kind work. Otherwise, I’m happy to see this useful study published.
Line by line notes:
Line 12: Should define LSG earlier in the manuscript
19: “leads to a lower MRA” is too vague; should be supported with values
44: Incomplete sentence here
55: I believe that “LSG” is still undefined at this point in the manuscript!
62: I like “come-as-you-are”
70: The formatting of this equation makes it seem like it is all multiplied by “14C yr” when that is, in fact, the units of the equation. I see this in the later equation as well.
155: yes, once again this formatting could be confusing
178: I think it’s generally understood this publication is using planktic foraminifera radiocarbon compilation, so I think it would help to be explicit about this?
193: Should just be “Results and Discussion”?
213: Statistics could be used instead of “compare well” here
246: I think this run of one sentence paragraphs could be streamlined.
255: A run-on sentence, I believe.
256: “for the same model do we find”?
262: The manuscript likely needs some explanation of why / how there was a scaling of the eddy diffusivity to the AMOC proxy. This is unclear to me and likely other readers.
285: “dynamical behaviour” is not clear text. Does this refer to differences in model physics?
285: Notably, there is no LSG model comparison here—maybe the manuscript should quickly state why?
362: “…which model might be the..” (?)
Figure 2: I like the layout of the panels. However, the caption confused me at first. I might replace “for runs with abrupt AMOC changes” with “for runs with modeled abrupt AMOC changes”. When I first read this caption (before I read the paper), it almost sounded like the AMOC changes developed on their own!