Present and future sensitivity of the Antarctic surface mass balance to oceanic and atmospheric forcings: insights with the regional climate model MAR

Recent Global Warming has caused widespread ice losses from the Antarctic Ice Sheet (AIS) leading to an increase in mean sea level. By influencing the ice dynamics and the mass of water that accumulates on the continent, the surface mass balance (SMB, i.e, the difference between snow accumulation and ablation at the surface of the ice sheet) contributes to sea-level variations. A better knowledge of how present and future SMB will change is therefore needed to refine sea-level-rise estimates.

With the aim of identifying the driving processes from different components of the climate system (from the surface of the ocean to high-elevation clouds), we reconstruct and project the Antarctic SMB using the regional climate model MAR developed at ULiège, over 1980--2100. The results of MAR have been first compared to diverse observations to evaluate its performance. We gathered observations of several types (near-surface climate and snow accumulation) to guarantee the robustness of our results and conclusions based on our climate modeling. A first objective of this thesis was to determine to what extent the recent changes at the ocean surface can exert a direct feedback on the atmosphere and SMB. Our simulations with perturbed sea-ice concentration and sea-surface temperature around Antarctica reveal that strong and persistent katabatic winds prevent most atmospheric changes induced by the ocean to penetrate inland. This suggests a limited influence of the ocean surface on the Antarctic SMB. We focused afterwards on the sensitivity of the SMB to atmospheric warmings projected by global models using high-emission scenarios (RCP8.5 and ssp585). Higher temperatures are projected to increase SMB on the grounded ice as a result of stronger snowfall while the future SMB over the ice shelves should be dominated by higher meltwater-runoff values (compromising the stability of ice shelves) and is consequently projected to decrease. Leaving aside the role of the ocean on the thinning of ice shelves, increasing surface melt should however remain weak under the Paris Agreement limiting potential ice-shelf collapses and accelerated Antarctic ice losses. However, our results suggest a large spread in melt increase over the ice shelves during the 21st century resulting in large uncertainties in their potential disappearance. Given the important role of ice shelves in limiting the acceleration of Antarctic ice losses (as they restrain the grounded ice to flow into the ocean), the third subject of this thesis has been devoted to the physical drivers explaining differences in increased summer melt over the Antarctic ice shelves. Although the melt increase results from higher greenhouse-gas concentrations, differences in projected melt increases arises from liquid-containing clouds. These clouds re-emits more longwave energy towards the surface, increasing melt over the ice shelves and later favouring absorption of solar energy again strengthening melt.

In conclusion, we investigate the sensitivity of the Antarctic SMB to different components of the
climate system over 1981--2100. Uncertainties linked with the grounded Antarctic SMB essentially
depend on the projected increased rates in snowfall associated with higher temperatures while uncertainties in the ice-shelf SMB decrease are related to cloud properties with more liquid-containing clouds leading to a stronger decrease of the ice-shelf SMB.