An equilibrium simulation of Marine Isotope Stage 3 (MIS3) climate with boundary conditions characteristic of Greenland Interstadial 8 (GI-8; 38 kyr BP) is carried out with the Norwegian Earth System Model (NorESM). A computationally efficient configuration of the model enables long integrations at relatively high resolution, with the simulations reaching a quasi-equilibrium state after 2500 years. We assess the characteristics of the simulated large-scale atmosphere and ocean circulation, precipitation, ocean hydrography, sea ice distribution, and internal variability. The simulated MIS3 interstadial near-surface air temperature is 2.9
Marine Isotope Stage 3 (MIS3), a period about 60–30 kyr BP (thousand years before present) during the last glacial, was characterised by millennial-scale abrupt climate transitions. These events are known as Dansgaard–Oeschger (D-O) events, as revealed by the Greenland oxygen isotope ice core records
While there have been significant advances in our understanding of the dynamics behind D-O events in recent years, the key mechanisms triggering these abrupt climate transitions remain elusive. A leading hypothesis is related to a switch between strong, weak, and off modes of the AMOC
Although prescribed external forcing are often introduced into the model to trigger D-O events
Most studies of D-O events apply a coupled model configured with Last Glacial Maximum (LGM) or pre-industrial (PI) boundary conditions. Very few model studies apply MIS3 boundary conditions. These MIS3 studies include models with varying resolution and complexities, including an atmospheric general circulation model forced with fixed sea surface temperatures (SSTs)
In this work, we present a MIS3 interstadial equilibrium simulation employing a new version of the Norwegian Earth System Model (NorESM) designed for multi-millennial and ensemble studies
We configure the model with boundary conditions characteristic of 38 kyr BP, immediately following the onset of Greenland Interstadial 8 (GI-8). Spontaneous occurrence of D-O-like abrupt climate transitions is not simulated in the model during the 2500-year integration using MIS3 boundary conditions. Instead, the experiment serves as a baseline simulation for evaluating equilibrium interstadial climate states during MIS3. A satisfactory quasi-equilibrium state is reached before the end of the long integration, and we assess the interstadial climate state of the atmosphere, ocean, and sea ice as represented by the model.
The significance of this study is that the MIS3 baseline simulation is configured with realistic MIS3 boundary conditions and integrated for multi-millennia with a climate model featuring relatively high resolution and reasonable performance in simulating AMOC and sea ice. This experiment is then used as a baseline for glacial sensitivity studies. Given the small number of existing MIS3 model studies, we aim to improve our understanding of the MIS3 climate, especially the baseline climate and the sensitivity of D-O-event-related variability to external forcing within a MIS3 configuration. The structure of the paper is arranged as follows: in Sect. 2, we give a brief overview of the NorESM, including details of the version used in this study, followed by a description of the MIS3 experimental configuration; in Sect. 3, we assess the equilibrium state of the MIS3 long integration with NorESM, followed by details of the simulated mean MIS3 interstadial state of the atmosphere, ocean, and sea ice. Discussions on the simulated AMOC and model response to changes in GHG and ice sheet height are presented in Sect. 4. The main conclusions are summarised in Sect. 5.
The NorESM family is based on the Community Climate System Model version 4
The basic evaluation and validation of the Climate Model Intercomparison Project Phase 5 (CMIP5) version of NorESM (NorESM1-M) is documented by
Compared to NorESM1-M, the model complexity of NorESM1-F is reduced by replacing CAM4-Oslo, which uses emissions of aerosols and explicitly simulates their life cycles, with the standard CAM4 that uses prescribed aerosol concentrations. The coupling frequency between atmosphere–sea ice and atmosphere–land is reduced from half-hourly to hourly, allowing the use of an hourly base time step for the sea ice and land components matching the radiative time step of the model; the dynamic subcycling of the sea ice is reduced from 120 to 80 subcycles. The last two changes result in a model speed-up of
In the MIS3 setup (Table
Forcings and boundary conditions for the MIS3 and PI simulations.
The insolation anomalies of MIS3 at 38 kyr BP relative to present day (W m
Concentrations of GHGs are set according to ice core measurements of typical interstadial conditions following
The ocean bathymetry is adapted based on an estimated sea level lowering of 70 m below present day
Land–sea mask (light grey/blue shading; dark grey indicates modern land), ice sheet extent (white line) for the MIS3 experiment, and the difference of ice sheet orography relative to PI (red contours with an interval of 250 m; the 1000 and 2000 m isolines are highlighted with bold red lines).
The configuration of global ice sheet extent and height (Fig.
For the configuration of land–sea mask in the Barents Sea, care needs to be taken, as this region is an important pathway allowing warm and saline Atlantic water travelling north; therefore, opening or closing it has significant consequences for Arctic ocean circulation and climate
The land surface vegetation type in the MIS3 configuration is set equal to the pre-industrial values, and the extra land points caused by sea level lowering are assigned as tundra (20 % grass plus 80 % bare ground).
With the adjusted MIS3 land–sea mask and surface topography, a new river-routing map is produced (Fig. S1 in the Supplement). For the ice-free land surfaces, the river routing corresponds to the PI simulation. Where there is new land, due to the lower MIS3 sea level, the river outlets are extended to the ocean. For the ice covered areas, a new map is generated based on the land ice topography, routing the water from the land ice along the steepest gradient, either directly to the ocean or to the nearest river if the ice margin terminates on land.
Salt equivalent to 0.6 g kg
The MIS3 experiment was run for 2500 years and the PI experiment for 2000 years. When comparing the two simulations, the model years between 1800 and 2000 are averaged.
Both the PI and MIS3 experiments reach a quasi-equilibrium climate state after the multi-millennial integration, as indicated by the time series shown in Fig.
Global mean values for the MIS3 and PI experiments (both averaged between years 1801 and 2000).
Time series of
In the following results, the statistical significance of the calculated trends is tested using the Student's
The MIS3 experiment exhibits a small negative TOA radiation balance (
At the ocean surface, sea surface salinity (SSS) in the MIS3 experiment exhibits negligible drift over the model years 1801–2000 (Fig.
While the simulated MIS3 surface properties reach a quasi-equilibrium state, the ocean interior experiences a multi-millennial cooling trend (Fig.
Both experiments exhibit an increasing AMOC at the beginning of the model integration, followed by a gradual equilibration to a weaker state (Fig.
The weakening of the AMOC, after the initial overshoot, occurs concurrently with a shoaling of North Atlantic Deep Water (NADW) and more intrusion of Antarctic Bottom Water (AABW) as a manifestation of an adjustment of the deep ocean. Previous studies relate this behaviour to an expansion of Antarctic sea ice in a colder climate
Simulated annual mean surface air temperature change with respect to PI is shown in Fig.
Simulated MIS3 minus PI annual mean near-surface air temperature (
Simulated global mean near-surface air temperature during MIS3 is 2.9
In contrast to the limited number of MIS3 studies, there is a rich literature on the LGM climate. With both similarities as well as apparent differences with regard to the external forcing and the climate, it can be useful to compare the climate of the two periods: our simulated MIS3 cooling is smaller than the reconstructed LGM global mean cooling of
The elevated surface of the MIS3 Laurentide Ice Sheet modifies the atmospheric stationary waves, rendering an enhanced, meandering wave pattern in the vicinity of the North American continent (Fig.
Simulated MIS3 and PI DJF 500 mbar geopotential height (hm). The black and red contours are for the MIS3 and PI experiments, respectively.
Simulated annual surface wind stress over the ocean for
In the tropics, the northeasterly trade winds are strengthened in the NH, while in the SH the southeasterly trade winds are relatively unchanged. In the Southern Ocean, the westerlies are strengthened in the Pacific Ocean sector and weakened in the Indian Ocean sector. The zonal mean of the westerly wind stress in the Southern Ocean shows a slight strengthening during MIS3 (
In the colder MIS3 climate, simulated global mean precipitation is 0.18 mm d
The reduced atmospheric
Simulated MIS3
AMOC for
The cooling in the North Pacific is partly associated with a reduction of northward ocean heat transport in this region (Fig. S5); e.g. ocean heat transport is 23 % smaller at 30
As northward ocean heat transport in the North Pacific decreases, southward ocean heat transport in the South Pacific and Indian Ocean increases (e.g. 13 % increase at 30
Simulated MIS3 – PI sea ice thickness anomalies (shading; m) for
For the surface salinity (Fig.
Dramatic freshening takes place in the Eurasian sector of the Arctic at MIS3. Here, the Arctic river outlets extend further out to the open ocean, relative to the present-day locations, due to the change of land–sea mask (Fig.
The fresh surface water in the South China Sea during MIS3 is caused by increased runoff from the new river routing in this region due to the change of land–sea mask. In the southwest Pacific, surface freshening is due to a southward shift of the ITCZ and an overall decrease of evaporation minus precipitation in the region. Off the coast of Antarctica, enhanced formation of sea ice (Fig.
Our NorESM experiments show a strengthened AMOC at MIS3 (27.5 Sv) relative to the PI (24.3 Sv) (Fig.
Together with the changes to the AMOC, the deep Atlantic ocean exhibits changes in the distribution of water masses. The zonal mean Atlantic (including the Nordic Seas and the Atlantic sector of the Southern Ocean) temperature (Fig.
Atlantic zonal mean
With more vigorous deep water formation in the NH (associated with a stronger AMOC) and in the SH (as discussed later) during MIS3, enhanced upward motion of sea water is expected away from the sinking regions
The Atlantic zonal mean salinity anomaly (Fig.
In the cold deep ocean, the salinity effect dominates the change of density, manifested by a larger increase of potential density in the Atlantic sector of the Southern Ocean (0.6–0.8 kg m
As documented by
In the Atlantic at MIS3, there is more winter sea ice south of Newfoundland and in the northeastern Labrador Sea (Fig.
In September, the simulated MIS3 sea ice retreats and nearly coincides with PI sea ice extent in the Pacific side (not shown) and in the Labrador Sea (Fig.
In the SH, MIS3 shows extended Antarctic sea ice cover in both seasons. The seasonal cycle is large in both MIS3 and PI experiments, with the total sea ice area varying by a factor of 3 and 4 between March and September, respectively (Fig. S2; Table
We briefly evaluate the simulated change of two important climate internal variabilities: the El Niño–Southern Oscillation (ENSO) and the Northern Annular Mode (NAM). Simulated ENSO variability is weakened in the MIS3 compared to the PI simulation. For the NAM, simulated centres of action over the Arctic and North Pacific are both weakened, with the latter much reduced due to the presence of the elevated Laurentide Ice Sheet. More details are given in the Supplement (Sect. S3).
Abrupt climate changes such as D-O events have been shown to involve changes in both the geometry and strength of the AMOC, as indicated by a number of marine proxy reconstructions
Previous studies have argued that the increased production of AABW during glacial times is driven by expanded Antarctic sea ice and enhanced brine rejection during sea ice formation in the Southern Ocean
The simulated ventilation of AABW is enhanced in our MIS3 simulation compared to the PI (Fig. S7). However, in the Atlantic, the volume of AABW is not comparable with that of the LGM, during which benthic foraminiferal
While deep water production in the Southern Ocean has the potential to displace and reduce the strength of the NADW production, competing effects are at play in the North Atlantic. For example, the altered surface westerlies in the North Atlantic caused by the elevated Laurentide Ice Sheet (Figs.
During the last glacial, sea level lowering and the removal of shallow continental shelves (Fig.
The NorESM MIS3 simulation presented above is representative of an interstadial climate, i.e. a relatively warm period during the last glacial; in agreement with paleo-reconstructions, this includes Greenland temperatures only 5–8
It is unclear if the baseline climate during MIS3 should be a stadial or interstadial state, nor is it clear that there is indeed a baseline climate, as the climate states can be inherently oscillatory
In the MIS3 “stadial” experiment, the global near-surface temperature cools by 0.4
With the multi-millennial long integration of the MIS3 simulation presented in this work, only one stable climate state is found, and the model reaches a quasi-equilibrium with a small drift towards the end of the integration. In this section, we explore the potential for model bi-stability associated with the transition between the warm interstadial and cold stadial climate states of MIS3. We do so by perturbing the model with changes in atmospheric
Our NorESM MIS3 simulations agree with that of
The studies by
It is not within the scope of this paper to perform a complete examination of the model response to every combination of
Contrary to the studies cited above, the NorESM MIS3 experiments exhibit surprising stability without any significant changes in Greenland temperature, sea ice, or AMOC (Fig.
Time series of AMOC at 26.5
Further analysis of key relevant metrics, e.g. spatial distributions of SSS, winter sea ice extent, and AMOC geometry for the sensitivity experiments, are included in the Supplement (Sect. S4). As the changes are highly related to the strength of the AMOC, which is only weakly reduced in the sensitivity experiments, changes in these metrics are also relatively small.
The results of the sensitivity experiments to
To trigger a cold stadial-like climate state in the NorESM, other mechanisms including enhanced iceberg calving and freshwater input to the North Atlantic from the Laurentide, Greenland and Fennoscandian ice sheets should be considered. There is a rich literature on applying freshwater flux of different magnitudes and locations to study climate response and transitions
In this paper, we present an equilibrium simulation of Marine Isotope Stage 3 forced by 38 kyr BP boundary conditions, with a recently developed version of the NorESM featuring a horizontal resolution of 2
The reported simulation, with its current length of integration, does not produce spontaneous transitions between colder stadial and warmer interstadial climate states. Rather, we obtain a MIS3 background climate state with a state-of-the-art climate model that can serve as a baseline for investigating mechanisms behind D-O events by discriminating different factors that can invoke abrupt transitions, e.g. freshwater input, changes in GHG concentrations, ice sheet size, orbital forcing, and ocean diapycnal mixing.
Despite a small drift due to ocean cooling, the model reaches a quasi-equilibrium state in terms of both surface properties and deep ocean hydrography. We analyse the large-scale features of the mean climate states and the model internal variabilities, and compare the results to previous studies of MIS3 and LGM climates. The major findings are as follows:
Globally, the simulated MIS3 interstadial climate is 2.9 The global mean SST at MIS3 is 1.2 Despite the uniform addition of salt into the global ocean (by 0.6 g kg The upper cell of the AMOC is deepened and intensified under the influence of competing factors from both hemispheres: the cutoff of freshwater input due to the closed Bering Strait and the strengthened surface wind stress in the NH subpolar region both tend to invigorate the AMOC. In the Southern Ocean, expansion of Antarctic sea ice stimulates AABW production by enhanced salt rejection and deep water production. The results are supported by marine proxy records indicating an AMOC comparable to the present day during the last glacial (except during Heinrich stadials). The enhanced deep ocean ventilation in the Atlantic sector of the Southern Ocean leads to reduced (but not reversed) deep north–south salinity gradients in the Atlantic Ocean. The Atlantic displays pronounced cooling below 3000 m in both hemispheres and near the base of the thermocline, the latter due to stronger upwelling of deep water as a result of enhanced deep water formation in both hemispheres. Reduced Mediterranean outflow during MIS3 contributes to the notable cooling and freshening observed around 30 Sea ice is notably thicker and greater in extent during MIS3 in both hemispheres and seasons. Arctic sea ice is about 2 m thicker and extends further equatorward in the Pacific during winter. The Nordic Seas are partly ice-covered in boreal summer; in winter, sea ice extent is greater but includes an opening in the south due to the intrusion of warm Atlantic water. In the Southern Hemisphere, Antarctic sea ice is thicker (mainly in the western Indian Ocean sector) and extends further north. A sensitivity experiment with boundary conditions typical of MIS3 stadial conditions does not reproduce the cold temperatures observed in Greenland, indicating that the interstadial climate in NorESM is relatively stable, and forms the baseline climate during MIS3. Further sensitivity experiments including large changes in atmospheric
The model code can be obtained upon request. Instructions on how to obtain a copy are given at
The supplement related to this article is available online at:
CG, KHN, and MB designed the study; CG set up and performed the simulation with help from IB, ZZ, and MB; CG led the analysis and writing of the manuscript; all authors contributed to the discussion and final writing of the manuscript.
The authors declare that they have no conflict of interest.
We acknowledge Peter Langen, Christian Rodehacke, and Will Roberts for fruitful discussions on configuring MIS3 boundary conditions. We thank Lev Tarasov for providing the ice sheet data to force the model. We also thank Anne-Katrine Faber, Marlene Klockmann (as a referee), and three other anonymous referees for the constructive comments which have helped significantly in improving the manuscript. The simulations were performed on resources provided by UNINETT Sigma2 – the National Infrastructure for High Performance Computing and Data Storage in Norway (nn4659k, ns4659k).
This research has been supported by the European Research Council under the European Community's Seventh Framework Programme (FP7/2007–2013)/ERC (ICE2ICE project, grant no. 610055).
This paper was edited by Laurie Menviel and reviewed by Marlene Klockmann and three anonymous referees.