During glacial climates, the Atlantic thermohaline circulation (the ocean's density-driven overturning, today expressed as the Atlantic Meridional Overturning Circulation, AMOC) varied on both glacial-cycle (orbital, ~10^4–10^5 yr) and millennial (Dansgaard–Oeschger/Heinrich, ~10^3 yr) timescales. Liu (2006) synthesized observations and model simulations into a timescale-dependent dominance hypothesis: because the Southern Ocean responds slowly (millennia) and the North Atlantic fast (centuries), Southern Ocean forcing — via winter sea-ice growth, brine rejection, and enhanced Antarctic Bottom Water (AABW) production, coupled to the ocean carbon cycle and dust — controls the glacial thermohaline circulation at orbital timescales, while North Atlantic forcing controls it at millennial timescales. Millennial-scale interpolar phasing has since been pinned down (Greenland leads abrupt changes, Antarctica responds with a ~200-year oceanically propagated lag), but the ultimate trigger of the millennial AMOC events, the quantitative closure of the sea-ice–brine–AABW pathway and its carbon-cycle coupling, and the robustness of the dominance picture across model formulations all remain open.

Background

The ocean's thermohaline circulation (THC) is the global overturning driven by density contrasts in temperature and salinity; in the Atlantic it is usually quantified as the Atlantic Meridional Overturning Circulation (AMOC), the northward transport of warm upper waters and southward return of North Atlantic Deep Water (NADW). Ice-core and marine-sediment records show that during glacial periods this overturning was persistently weaker and shallower than today, and that it fluctuated on two distinct timescales. On millennial timescales, Greenland ice cores record repeated abrupt warmings followed by slow coolings — Dansgaard–Oeschger (D-O) events — interleaved with massive iceberg-discharge (Heinrich) events and cold Heinrich stadials, while Antarctic cores show gradual warmings and coolings that are antiphased with Greenland: cooling in Antarctica when Greenland warms, and vice versa. This interhemispheric asymmetry is the 'bipolar seesaw', formalized by Stocker and Johnsen (2003) as a heat-reservoir model in which a collapse of AMOC heat transport cools the north while the stored excess heat gradually warms the south. Blunier and Brook (2001) had established the Antarctic–Greenland asynchrony on methane-synchronized chronologies, but Wunsch (2003) argued that the apparent Greenland–Antarctic phase relations at millennial periods were not statistically robust, so the observational inference of which hemisphere drives the variability was contested. On orbital timescales the picture was equally contested: Shackleton (2000) found that glacial-cycle atmospheric CO2_2 and southern-hemisphere temperature led global ice volume, while Alley et al. (2002) argued for a northern lead in the orbital band.

Coupled-model studies then supplied a mechanistic bridge between the hemispheres. Shin et al. (2003a) completed an equilibrium Last Glacial Maximum (LGM) integration of the NCAR Community Climate System Model, and Shin et al. (2003b) showed that Southern Ocean sea-ice expansion under glacial conditions — through brine rejection (salt release during ice growth) densifying shelf waters and invigorating Antarctic Bottom Water (AABW) — increased oceanic vertical stability and reduced North Atlantic Deep Water production: a Southern Ocean control on the glacial North Atlantic overturning. Liu et al. (2005) further showed in the same model lineage that lowered glacial atmospheric CO2_2 alone could reproduce much of the glacial THC weakening and the reversed deep temperature–salinity distribution, implying a positive feedback between the Southern-Ocean-regulated overturning and the glacial carbon cycle. On the deglacial side, Knorr and Lohmann (2003) used an ocean model to argue that Southern Ocean warming and sea-ice retreat triggered the resumption of vigorous Atlantic overturning during deglaciation.

Against this background Liu (2006) proposed the unifying, explicitly speculative view at the center of this problem: because the Southern Ocean's response time is slow (order millennia) and the North Atlantic's fast (order centuries), the North Atlantic thermohaline circulation is controlled predominantly by Southern Ocean forcing at long glacial-cycle timescales but by North Atlantic forcing at short millennial timescales. The author acknowledged that the millennial-scale phasing and coherence of the Antarctic and Greenland records were then statistically ambiguous; that testing the hypothesis required additional coupled model experiments, process studies, and more robust statistical analyses of proxies; that the underlying equilibrium LGM integration used deep-water acceleration and a simple precipitation-scaling rather than explicit river runoff, and that model formulation, resolution, eddy/Antarctic Circumpolar Current (ACC) parameterization, and the absence of an interactive carbon cycle could alter the magnitude and timing of the Southern Ocean response; and that the strength, timing, and parameter dependence of the sea-ice–brine–AABW mechanism and its coupling to dust and biogeochemical feedbacks required quantification. Here 'dust/biogeochemical coupling' denotes the proposed iron-fertilization and nutrient/stratification feedbacks through which glacial dust supply and Southern Ocean physics set atmospheric CO2_2, which in turn feeds back on the overturning. The Paleoclimate Model Intercomparison Project (PMIP) is the coordinated protocol under which modeling groups simulate the LGM with common boundary conditions.

Problem Statement

Determine which hemisphere's forcing dominates the Atlantic thermohaline circulation during glacial climates at each timescale — where dominance is judged as the hemisphere whose forcing predominantly controls the glacial overturning at that timescale, not by the magnitude or timing of the forcing alone — testing Liu's (2006) timescale-dependent dominance hypothesis that Southern Ocean forcing (winter sea-ice growth, brine rejection, enhanced Antarctic Bottom Water production, coupled to dust and ocean biogeochemical/CO2 feedbacks) controls the overturning at glacial-cycle (orbital) timescales, while North Atlantic forcing controls it at millennial (Dansgaard–Oeschger/Heinrich) timescales. The question is adjudicated on the evidence types the source itself uses — proxy/observational records and coupled model simulations — without excluding other adequate evidence. The overarching objective is a validated (or refuted) mechanistic, timescale-resolved attribution of glacial overturning variability to northern versus southern forcing. The attribution problem encompasses these central unresolved facets: the statistical phasing and coherence of bipolar climate records in the millennial and orbital bands (with significance assessments robust to red-noise and chronology criticisms); the strength, timing, and parameter dependence of the Southern Ocean sea-ice–brine–AABW mechanism and its coupling to dust and biogeochemical feedbacks, against glacial and deglacial proxy evidence; and the robustness of any such attribution to model formulation — resolution, sea-ice and eddy/ACC parameterization, and inclusion of an interactive carbon cycle — including where and why the dominance picture may fail.

References

  1. Liu, Zhengyu (2006). Glacial thermohaline circulation and climate: Forcing from the north or south? Advances in Atmospheric Sciences 23(2): 199–206. DOI 10.1007/s00376-006-0199-7. https://doi.org/10.1007/s00376-006-0199-7
  2. Stocker, Thomas F., and Sigfùs J. Johnsen (2003). A minimum thermodynamic model for the bipolar seesaw. Paleoceanography 18(4): 1087. DOI 10.1029/2003PA000920. https://doi.org/10.1029/2003PA000920
  3. Blunier, Thomas, and Edward J. Brook (2001). Timing of millennial-scale climate change in Antarctica and Greenland during the last glacial period. Science 291(5501): 109–112. DOI 10.1126/science.291.5501.109. https://doi.org/10.1126/science.291.5501.109
  4. Wunsch, Carl (2003). Greenland–Antarctic phase relations and millennial time-scale climate fluctuations in the Greenland ice-cores. Quaternary Science Reviews 22(15–17): 1631–1646. DOI 10.1016/S0277-3791(03)00152-5. https://doi.org/10.1016/S0277-3791(03)00152-5
  5. Shackleton, Nicholas J. (2000). The 100,000-year ice-age cycle identified and found to lag temperature, carbon dioxide, and orbital eccentricity. Science 289(5486): 1897–1902. DOI 10.1126/science.289.5486.1897. https://doi.org/10.1126/science.289.5486.1897
  6. Alley, Richard B., Edward J. Brook, and Sridhar Anandakrishnan (2002). A northern lead in the orbital band: north–south phasing of ice-age events. Quaternary Science Reviews 21(1–3): 431–441. DOI 10.1016/S0277-3791(01)00072-5. https://doi.org/10.1016/S0277-3791(01)00072-5
  7. Shin, Sang-Ik, Zhengyu Liu, Bette L. Otto-Bliesner, John E. Kutzbach, and Stephen J. Vavrus (2003). Southern Ocean sea-ice control of the glacial North Atlantic thermohaline circulation. Geophysical Research Letters 30(2). DOI 10.1029/2002GL015513. https://doi.org/10.1029/2002GL015513
  8. Shin, S.-I., Z. Liu, B. Otto-Bliesner, E. Brady, J. Kutzbach, and S. Harrison (2003). A simulation of the Last Glacial Maximum climate using the NCAR-CCSM. Climate Dynamics 20: 127–151. DOI 10.1007/s00382-002-0260-x. https://doi.org/10.1007/s00382-002-0260-x
  9. Liu, Zhengyu, Sang-Ik Shin, Robert S. Webb, William Lewis, and Bette L. Otto-Bliesner (2005). Atmospheric CO2 forcing on glacial thermohaline circulation and climate. Geophysical Research Letters 32: L02706. DOI 10.1029/2004GL021929. https://doi.org/10.1029/2004GL021929
  10. Knorr, Gregor, and Gerrit Lohmann (2003). Southern Ocean origin for the resumption of Atlantic thermohaline circulation during deglaciation. Nature 424(6948): 532–536. DOI 10.1038/nature01855. https://doi.org/10.1038/nature01855
  11. EPICA Community Members (2006). One-to-one coupling of glacial climate variability in Greenland and Antarctica. Nature 444(7116): 195–198. DOI 10.1038/nature05301. https://doi.org/10.1038/nature05301
  12. WAIS Divide Project Members (2015). Precise interpolar phasing of abrupt climate change during the last ice age. Nature 520(7549): 661–665. DOI 10.1038/nature14401. https://doi.org/10.1038/nature14401
  13. Liu, Z., B. L. Otto-Bliesner, F. He, E. C. Brady, R. Tomas, P. U. Clark, A. E. Carlson, J. Lynch-Stieglitz, W. Curry, E. Brook, D. Erickson, R. Jacob, J. Kutzbach, and J. Cheng (2009). Transient Simulation of Last Deglaciation with a New Mechanism for Bølling-Allerød Warming. Science 325(5938): 310–314. DOI 10.1126/science.1171041. https://doi.org/10.1126/science.1171041
  14. Peltier, W. Richard, and Guido Vettoretti (2014). Dansgaard–Oeschger oscillations predicted in a comprehensive model of glacial climate: A 'kicked' salt oscillator in the Atlantic. Geophysical Research Letters 41(20): 7306–7313. DOI 10.1002/2014GL061413. https://doi.org/10.1002/2014GL061413
  15. Vettoretti, Guido, and W. Richard Peltier (2015). Interhemispheric air temperature phase relationships in the nonlinear Dansgaard–Oeschger oscillation. Geophysical Research Letters 42(4): 1180–1189. DOI 10.1002/2014GL062898. https://doi.org/10.1002/2014GL062898
  16. Brown, Nicolas, and Eric D. Galbraith (2016). Hosed vs. unhosed: interruptions of the Atlantic Meridional Overturning Circulation in a global coupled model, with and without freshwater forcing. Climate of the Past 12(8): 1663–1679. DOI 10.5194/cp-12-1663-2016. https://doi.org/10.5194/cp-12-1663-2016
  17. Obase, Takashi, and Ayako Abe-Ouchi (2019). Abrupt Bølling–Allerød warming simulated under gradual forcing of the last deglaciation. Geophysical Research Letters 46(20): 11397–11405. DOI 10.1029/2019GL084675. https://doi.org/10.1029/2019GL084675
  18. Thompson, Andrew F., Sophia K. Hines, and Jess F. Adkins (2019). A Southern Ocean mechanism for the interhemispheric coupling and phasing of the bipolar seesaw. Journal of Climate 32(14): 4347–4365. DOI 10.1175/JCLI-D-18-0621.1. https://doi.org/10.1175/JCLI-D-18-0621.1
  19. Skinner, Luke, Laurie Menviel, Lauren Broadfield, Julia Gottschalk, and Mervyn Greaves (2020). Southern Ocean convection amplified past Antarctic warming and atmospheric CO2 rise during Heinrich Stadial 4. Communications Earth & Environment 1: 23. DOI 10.1038/s43247-020-00024-3. https://doi.org/10.1038/s43247-020-00024-3
  20. Muglia, Juan, and Andreas Schmittner (2015). Glacial Atlantic overturning increased by wind stress in climate models. Geophysical Research Letters 42(22): 9862–9868. DOI 10.1002/2015GL064583. https://doi.org/10.1002/2015GL064583
  21. Kageyama, Masa, Sandy P. Harrison, Marie-L. Kapsch, Marcus Lofverstrom, Juan M. Lora, Uwe Mikolajewicz, Sam Sherriff-Tadano, Tristan Vadsaria, Ayako Abe-Ouchi, Nathaëlle Bouttes, Deepak Chandan, Lauren J. Gregoire, Ruza F. Ivanovic, Kenji Izumi, Allegra N. LeGrande, Fanny Lhardy, Gerrit Lohmann, Polina A. Morozova, Rumi Ohgaito, André Paul, W. Richard Peltier, Christopher J. Poulsen, Aurélien Quiquet, Didier M. Roche, Xiaoxu Shi, Jessica E. Tierney, Paul J. Valdes, Evgeny Volodin, and Jiang Zhu (2021). The PMIP4 Last Glacial Maximum experiments: preliminary results and comparison with the PMIP3 simulations. Climate of the Past 17(3): 1065–1089. DOI 10.5194/cp-17-1065-2021. https://doi.org/10.5194/cp-17-1065-2021
  22. Lhardy, Fanny, Nathaëlle Bouttes, Didier M. Roche, Xavier Crosta, Claire Waelbroeck, and Didier Paillard (2021). Impact of Southern Ocean surface conditions on deep ocean circulation during the LGM: a model analysis. Climate of the Past 17(3): 1139–1159. DOI 10.5194/cp-17-1139-2021. https://doi.org/10.5194/cp-17-1139-2021
  23. Hasenfratz, Adam P., Samuel L. Jaccard, Alfredo Martínez-García, Daniel M. Sigman, David A. Hodell, Derek Vance, Stefano M. Bernasconi, Helga (Kikki) F. Kleiven, F. Alexander Haumann, and Gerald H. Haug (2019). The residence time of Southern Ocean surface waters and the 100,000-year ice age cycle. Science 363(6431): 1080–1084. DOI 10.1126/science.aat7067. https://doi.org/10.1126/science.aat7067
  24. Sigman, Daniel M., Mathis P. Hain, and Gerald H. Haug (2010). The polar ocean and glacial cycles in atmospheric CO2 concentration. Nature 466(7302): 47–55. DOI 10.1038/nature09149. https://doi.org/10.1038/nature09149
  25. Hain, Mathis P., and Daniel M. Sigman (2024). CO2 in Earth's ice age cycles. Oxford Research Encyclopedia of Climate Science. DOI 10.1093/acrefore/9780190228620.013.879. https://doi.org/10.1093/acrefore/9780190228620.013.879
  26. Zheng, Peisong, Thomas Bauska, and Matthew Osman (2025). Early warming over the Southern Ocean during the last deglaciation. Geophysical Research Letters 52(17): e2025GL117155. DOI 10.1029/2025GL117155. https://doi.org/10.1029/2025GL117155
  27. Huang, Huang, Marcus Gutjahr, Yuanyang Hu, Frerk Pöppelmeier, Gerhard Kuhn, Jörg Lippold, Thomas A. Ronge, Shuzhuang Wu, Patrick Blaser, Lester Lembke-Jene, Samuel L. Jaccard, Yimin Luo, and Jimin Yu (2026). Expansion of Antarctic Bottom Water driven by Antarctic warming in the last deglaciation. Nature Geoscience 19(1): 113–119 (published online 2025-12-01). DOI 10.1038/s41561-025-01853-7. https://doi.org/10.1038/s41561-025-01853-7
  28. Trombini, Irene, Nils Weitzel, Paul J. Valdes, Jean-Philippe Baudouin, Edward Armstrong, and Kira Rehfeld (2025). Atmospheric and Oceanic Pathways Drive Separate Modes of Southern Hemisphere Climate in Simulations of Spontaneous Dansgaard–Oeschger-Type Oscillations. Geophysical Research Letters 52(5): e2024GL111473. DOI 10.1029/2024GL111473. https://doi.org/10.1029/2024GL111473
  29. Lhardy, Fanny (2026). Exploring the links between model biases in Southern Ocean sea ice and deep ocean circulation in glacial simulations. EGU General Assembly 2026 abstract (EGU26-19500). DOI 10.5194/egusphere-egu26-19500. https://doi.org/10.5194/egusphere-egu26-19500
  30. Barathieu, Héloïse, Thibaut Caley, Masa Kageyama, Didier Swingedouw, and Pascale Braconnot (2026). Using ocean surface paleo-density to evaluate PMIP3 and PMIP4 Last Glacial Maximum climate simulations. EGUsphere preprint. DOI 10.5194/egusphere-2026-254. https://doi.org/10.5194/egusphere-2026-254