Description: Heinrich events are among the dominant modes of glacial climate variability. During these events, massive iceberg armadas were released by the Laurentide Ice Sheet, sailed across the Atlantic, and caused large-scale climate changes. We study these events in a fully coupled complex ice sheet–climate model with synchronous coupling between ice sheets and oceans. The ice discharges occur as internal variability of the model with a recurrence period of 5kyr, an event duration of 1–1.5kyr, and a peak discharge rate of about 50mSv, roughly consistent with reconstructions. The climate response shows a two-stage behavior, with freshwater release effects dominating the surge phase and ice-sheet elevation effects dominating in the post-surge phase. As a direct response to the freshwater discharge during the surge phase, the deepwater formation in the North Atlantic decreases and the North Atlantic deepwater cell weakens by 3.5Sv. With the reduced oceanic heat transport, the surface temperatures across the North Atlantic decrease, and the associated reduction in evaporation causes a drying in Europe. The ice discharge lowers the surface elevation in the Hudson Bay area and thus leads to increased precipitation and accelerated ice sheet regrowth in the post-surge phase. Furthermore, the jet stream widens to the north and becomes more zonal. This contributes to a weakening of the subpolar gyre, and a continued cooling over Europe even after the ice discharge. This two-stage behavior can explain previously contradicting model results and understandings of Heinrich Events.

Description: This study analyzes the response of the Atlantic meridional overturning circulation (AMOC) to different CO2 concentrations and two ice sheet configurations in simulations with the coupled climate model MPI-ESM. With preindustrial (PI) ice sheets, there are two different AMOC states within the studied CO2 range: one state with a strong and deep upper overturning cell at high CO2 concentrations and one state with a weak and shallow upper cell at low CO2 concentrations. Changes in AMOC variability with decreasing CO2 indicate two stability thresholds. The strong state is stable above the first threshold near 217 ppm, and the weak state is stable below the second threshold near 190 ppm. Between the two thresholds, both states are marginally unstable, and the AMOC oscillates between them on millennial time scales. The weak AMOC state is stable when Antarctic Bottom Water becomes dense and salty enough to replace North Atlantic Deep Water (NADW) in the deep North Atlantic and when the density gain over the North Atlantic becomes too weak to sustain continuous NADW formation. With Last Glacial Maximum (LGM) ice sheets, the density gain over the North Atlantic and the northward salt transport are enhanced with respect to the PI ice sheet case. This enables active NADW formation and a strong AMOC for the entire range of studied CO2 concentrations. The AMOC variability indicates that the simulated AMOC is far away from a stability threshold with LGM ice sheets. The nonlinear relationship among AMOC, CO2, and prescribed ice sheets provides an explanation for the large intermodel spread of AMOC states found in previous coupled LGM simulations.

Description: The Last Glacial Maximum (LGM, 21 000 years ago) is one of the suite of paleoclimate simulations included in the current phase of the Coupled Model Intercomparison Project (CMIP6). It is an interval when insolation was similar to the present, but global ice volume was at a maximum, eustatic sea level was at or close to a minimum, greenhouse gas concentrations were lower, atmospheric aerosol loadings were higher than today, and vegetation and land-surface characteristics were different from today. The LGM has been a focus for the Paleoclimate Modelling Intercomparison Project (PMIP) since its inception, and thus many of the problems that might be associated with simulating such a radically different climate are well documented. The LGM state provides an ideal case study for evaluating climate model performance because the changes in forcing and temperature between the LGM and pre-industrial are of the same order of magnitude as those projected for the end of the 21st century. Thus, the CMIP6 LGM experiment could provide additional information that can be used to constrain estimates of climate sensitivity. The design of the Tier 1 LGM experiment (lgm) includes an assessment of uncertainties in boundary conditions, in particular through the use of different reconstructions of the ice sheets and of the change in dust forcing. Additional (Tier 2) sensitivity experiments have been designed to quantify feedbacks associated with land-surface changes and aerosol loadings, and to isolate the role of individual forcings. Model analysis and evaluation will capitalize on the relative abundance of paleoenvironmental observations and quantitative climate reconstructions already available for the LGM.

Description: As ice sheets grow or decay, the net flux of freshwater into the ocean changes and the bedrock adjusts due to isostatic adjustments, leading to variations in the bottom topography and the oceanic boundaries. This process was particularly intense during the last deglaciation due to the high rates of ice-sheet melting. It is, therefore, necessary to consider transient ocean bathymetry and coastlines when attempting to simulate the last deglaciation with Earth system models (ESMs). However, in most standard ESMs the land-sea mask is fixed throughout simulations because the generation of a new ocean model bathymetry implies several levels of manual corrections, a procedure that is hardly doable very often for long runs. This is one of the main technical problems towards simulating a complete glacial cycle with general circulation models. For the first time, we present a tool allowing for an automatic computation of bathymetry and land-sea mask changes in the Max Planck Institute Earth System Model (MPI-ESM). The algorithms developed in this paper can easily be adapted to any free-surface ocean model that uses the Arakawa-C grid in the horizontal and z-grid in the vertical including partial bottom cells. The strategy applied is described in detail and the algorithms are tested in a long-term simulation demonstrating the reliable behaviour. Our approach guarantees the conservation of mass and tracers at global and regional scales; that is, changes in a single grid point are only propagated regionally. The procedures presented here are an important contribution to the development of a fully coupled ice sheet–solid Earth–climate model system with time-varying topography and will allow for transient simulations of the last deglaciation considering interactive bathymetry and land-sea mask.

Description: The continually evolving large ice sheets present in the Northern Hemisphere during the last glacial cycle caused significant changes to river pathways both through directly blocking rivers and through glacial isostatic adjustment. Studies have shown these river pathway changes had a significant impact on the ocean circulation through changing the pattern of freshwater discharge into the oceans. A coupled Earth system model (ESM) simulation of the last glacial cycle thus requires a hydrological discharge model that uses a set of river pathways that evolve with Earth's changing orography while being able to reproduce the known present-day river network given the present-day orography. Here, we present a method for dynamically modelling river pathways that meets such requirements by applying predefined corrections to an evolving fine-scale orography (accounting for the changing ice sheets and isostatic rebound) each time the river directions are recalculated. The corrected orography thus produced is then used to create a set of fine-scale river pathways and these are then upscaled to a coarser scale on which an existing present-day hydrological discharge model within the JSBACH land surface model simulates the river flow. Tests show that this procedure reproduces the known present-day river network to a sufficient degree of accuracy and is able to simulate plausible paleo-river networks. It has also been shown this procedure can be run successfully multiple times as part of a transient coupled climate model simulation.

Description: The recent increase in the rate of the Greenland ice sheet melting has raised with urgency the question of the impact of such a melting on the climate. As former model projections, based on a coarse representation of the melting, show very different sensitivity to this melting, it seems necessary to consider a multi-model ensemble to tackle this question. Here we use five coupled climate models and one ocean-only model to evaluate the impact of 0.1 Sv (1 Sv = 106 m3/s) of freshwater equally distributed around the coast of Greenland during the historical era 1965–2004. The ocean-only model helps to discriminate between oceanic and coupled responses. In this idealized framework, we find similar fingerprints in the fourth decade of hosing among the models, with a general weakening of the Atlantic Meridional Overturning Circulation (AMOC). Initially, the additional freshwater spreads along the main currents of the subpolar gyre. Part of the anomaly crosses the Atlantic eastward and enters into the Canary Current constituting a freshwater leakage tapping the subpolar gyre system. As a consequence, we show that the AMOC weakening is smaller if the leakage is larger. We argue that the magnitude of the freshwater leakage is related to the asymmetry between the subpolar-subtropical gyres in the control simulations, which may ultimately be a primary cause for the diversity of AMOC responses to the hosing in the multi-model ensemble. Another important fingerprint concerns a warming in the Nordic Seas in response to the re-emergence of Atlantic subsurface waters capped by the freshwater in the subpolar gyre. This subsurface heat anomaly reaches the Arctic where it emerges and induces a positive upper ocean salinity anomaly by introducing more Atlantic waters. We found similar climatic impacts in all the coupled ocean–atmosphere models with an atmospheric cooling of the North Atlantic except in the region around the Nordic Seas and a slight warming south of the equator in the Atlantic. This meridional gradient of temperature is associated with a southward shift of the tropical rains. The free surface models also show similar sea-level fingerprints notably with a comma-shape of high sea-level rise following the Canary Current.

Description: This study analyzes the response of the Atlantic meridional overturning circulation (AMOC) to different CO2 concentrations and two ice sheet configurations in simulations with the coupled climate model MPI-ESM. With preindustrial (PI) ice sheets, there are two different AMOC states within the studied CO2 range: one state with a strong and deep upper overturning cell at high CO2 concentrations and one state with a weak and shallow upper cell at low CO2 concentrations. Changes in AMOC variability with decreasing CO2 indicate two stability thresholds. The strong state is stable above the first threshold near 217 ppm, and the weak state is stable below the second threshold near 190 ppm. Between the two thresholds, both states are marginally unstable, and the AMOC oscillates between them on millennial time scales. The weak AMOC state is stable when Antarctic Bottom Water becomes dense and salty enough to replace North Atlantic Deep Water (NADW) in the deep North Atlantic and when the density gain over the North Atlantic becomes too weak to sustain continuous NADW formation. With Last Glacial Maximum (LGM) ice sheets, the density gain over the North Atlantic and the northward salt transport are enhanced with respect to the PI ice sheet case. This enables active NADW formation and a strong AMOC for the entire range of studied CO2 concentrations. The AMOC variability indicates that the simulated AMOC is far away from a stability threshold with LGM ice sheets. The nonlinear relationship among AMOC, CO2, and prescribed ice sheets provides an explanation for the large intermodel spread of AMOC states found in previous coupled LGM simulations.

Description: We investigate the changes in terrestrial natural methane emissions between the Last Glacial Maximum (LGM) and preindustrial (PI) periods by performing time-slice experiments with a methane-enabled version of MPI-ESM, the Max Planck Institute Earth System Model. We consider all natural sources of methane except for emissions from wild animals and geological sources, i.e. emissions from wetlands, fires, and termites. Changes are dominated by changes in tropical wetland emissions, with mid-to-high-latitude wetlands playing a secondary role, and all other natural sources being of minor importance. The emissions are determined by the interplay of vegetation productivity, a function of CO2 and temperature; source area size, affected by sea level and ice sheet extent; and the state of the West African monsoon, with increased emissions from northern Africa during strong monsoon phases. We show that it is possible to explain the difference in atmospheric methane between LGM and PI purely by changes in emissions. As emissions more than double between LGM and PI, changes in the atmospheric lifetime of CH4, as proposed in other studies, are not required.

Description: Publications on temperate deciduous tree refugia in Europe are abundant, but little is known about the patterns of temperate tree refugia in eastern Asia, an area where biodiversity survived Quaternary glaciations and which has the world's most diverse temperate flora. Our goal is to compare climate model simulations with pollen data in order to establish the location of glacial refugia during the Last Glacial Maximum (LGM). Limits in which temperate deciduous trees can survive are taken from the literature. The model outputs are first tested for the present by comparing climate models with published modern pollen data. As this method turned out to be satisfactory for the present, the same approach was used for the LGM. Climate model simulations (ECHAM5 T106), statistically further downscaled, are used to infer the temperate deciduous tree distribution during the LGM. These were compared with available fossil temperate tree pollen occurrences. The impact of the LGM on the eastern Asian climate was much weaker than on the European climate. The area of possible tree growth shifts only by about 2∘ to the south between the present and the LGM. This contributes to explaining the greater biodiversity of forests in eastern Asia compared to Europe. Climate simulations and the available, although fractional, fossil pollen data agree. Therefore, climate estimations can safely be used to fill areas without pollen data by mapping potential refugia distributions. The results show two important areas with population connectivity: the Yellow Sea emerged shelf and the southern Himalayas. These two areas were suitable for temperate deciduous tree growth, providing corridors for population migration and connectivity (i.e. less population fragmentation) in glacial periods. Many tree populations live in interglacial refugia, not glacial ones. The fact that the model simulation for the LGM fits so well with observed pollen distribution is another indication that the model used is good enough to also simulate the LGM period.

Description: In this paper we introduce a Bayesian framework, which is explicit about prior assumptions, for using model ensembles and observations together to constrain future climate change. The emergent constraint approach has seen broad application in recent years, including studies constraining the equilibrium climate sensitivity (ECS) using the Last Glacial Maximum (LGM) and the mid-Pliocene Warm Period (mPWP). Most of these studies were based on ordinary least squares (OLS) fits between a variable of the climate state, such as tropical temperature, and climate sensitivity. Using our Bayesian method, and considering the LGM and mPWP separately, we obtain values of ECS of 2.7 K (0.6–5.2, 5th–95th percentiles) using the PMIP2, PMIP3, and PMIP4 datasets for the LGM and 2.3 K (0.5–4.4) with the PlioMIP1 and PlioMIP2 datasets for the mPWP. Restricting the ensembles to include only the most recent version of each model, we obtain 2.7 K (0.7–5.2) using the LGM and 2.3 K (0.4–4.5) using the mPWP. An advantage of the Bayesian framework is that it is possible to combine the two periods assuming they are independent, whereby we obtain a tighter constraint of 2.5 K (0.8–4.0) using the restricted ensemble. We have explored the sensitivity to our assumptions in the method, including considering structural uncertainty, and in the choice of models, and this leads to 95 % probability of climate sensitivity mostly below 5 K and only exceeding 6 K in a single and most uncertain case assuming a large structural uncertainty. The approach is compared with other approaches based on OLS, a Kalman filter method, and an alternative Bayesian method. An interesting implication of this work is that OLS-based emergent constraints on ECS generate tighter uncertainty estimates, in particular at the lower end, an artefact due to a flatter regression line in the case of lack of correlation. Although some fundamental challenges related to the use of emergent constraints remain, this paper provides a step towards a better foundation for their potential use in future probabilistic estimations of climate sensitivity.