Key points are not available for this paper at this time.
The Atlantic Meridional Overturning Circulation (AMOC) is projected to weaken due to the increase in buoyancy caused by anthropogenic warming and consequent freshening during the 21st century and beyond. Atmosphere-ocean general circulation models simulate this AMOC weakening due to warming of the surface ocean, changes in the hydrological cycle that shift the North Atlantic salt budget, melting sea-ice, and changes in the atmospheric circulation. However, the freshwater contribution from the melting Greenland ice sheet is often either only considered in idealized scenarios or entirely omitted due to computational constraints. This simplification contributes to the large uncertainty surrounding the possibility of the AMOC crossing a tipping point in the forthcoming centuries. Here we employ the fully coupled Earth system model of intermediate complexity Bern3D v3, which dynamically simulates all ice-ocean-atmosphere interactions. We conduct a set of simulations driven by idealized CO2 concentration paths to investigate the impact of the melting Greenland ice sheet on the stability of the AMOC over the next 3000 years. We find that for a slow CO2 increase of 0.5%/yr up to twice pre-industrial levels, the general trends of the AMOC evolution are independent of whether Greenland meltwater is taken into account, with an initial weakening, but long-term recovery. Yet, the additional meltwater results in a further weakening of about 3 Sv after 100 years, but without leading to a full collapse of the circulation. This effect is due to melt rates remaining relatively low for the initial 100 years and only reaching their peak after 500 years. In the long-term, the curtailed AMOC and hence northward heat transport substantially slows down the disintegration of the Greenland ice sheet. Only in scenarios where the melt rates are kept artificially high, the AMOC does not recover. This highlights that the meltwater-induced AMOC weakening stabilizes the Greenland ice sheet, which in turn limits further AMOC weakening. This suggests that the potential for cascading interactions may be limited.
Pöppelmeier et al. (Fri,) studied this question.