Changes in sea-level, regional climate, and future coastlines due to the progressive disintegration of the Greenland and West Antarctic ice sheets

Current estimates of sea-level rise suggest nearly half a billion people could live on land vulnerable to temporary flooding by the end of this century, with potentially larger global populations, in addition to ecological and climatological threats, becoming more at risk in the far future when sea-level rise becomes increasingly dominated by melt from the Greenland (GrIS) and Antarctic (AIS) ice sheets. Long-term projections remain uncertain due to differences in models, process understanding and their parameterization with many simulations limited to only to 2100 or 2500 CE. Few studies have examined the multi-millennial response; those that do typically consider the GrIS or AIS alone and the focus is limited to global mean sea-level change, whereas spatial variations in sea-level and its implications for regional climate change are neglected.

Using the fast Earth system model CLIMBER-X, we perform idealized 50 kyr long simulations under pre-industrial CO₂ concentrations, in which the GrIS, West Antarctic (WAIS), and combined GrIS+WAIS are progressively disintegrated following a realistic pattern of melt derived from previous studies. We repeat these disintegration experiments for different prescribed constant atmospheric CO₂ concentrations, and target at changes in global mean near-surface temperature (ΔGMST) of 0–3 °C. Simulations start from a non-equilibrium state based on a dedicated transient simulation of the last glacial cycle with prescribed greenhouse gases and ice-load history, resulting in a present-day disequilibrium in bedrock elevation. These idealized experiments are able to quantify global and regional climate and sea-level response across varying levels of ice sheet melt and GMST change, as well as ocean thermal expansion.

First, we assess the individual impact of a GrIS+WAIS disintegration. Progressively disintegrated ice sheets leads to further warming on top of the targeted ΔGMST due to the albedo effect. However, we find that the added-up response from the individual ice sheet experiments do not reproduce the results from the GrIS+WAIS experiment, indicating the presence of nonlinear feedbacks when combined. There are significant interhemispheric differences, with regional temperatures in the added-up response from the individual ice sheet experiments differing by -13–3 °C when compared to the GrIS+WAIS experiment under pre-industrial CO₂ concentrations. These numbers tend to grow as ΔGMST increases.

Second, we compare our experiments to those initialized from a pre-industrial equilibrium to assess the effect of bedrock uplift, differing spatial variations in sea-level, and coastline migration. Whereas bedrock uplift has little effect on ΔGMST, it partially compensates long-term inundation in areas like Northern Europe and Hudson Bay. Hazard maps of progressive inundation are shown for the different simulations, illustrating plausible future coastlines. A dedicated experiment with the complete disintegration of the GrIS+AIS (all ~65m) is also shown. The presented results highlight the sensitivity of regional climate and sea-level to ongoing cryospheric change, and provide a framework to assess the long-term effect of ice sheet melt in the Earth system.