Atmospheric drivers of surface melting on the Larsen C ice shelf, Antarctic Peninsula

Observational data and high resolution (<4 km grid spacing) Met Office Unified Model (MetUM) output is used to investigate the dominant causes of surface melting on the Larsen C ice shelf. In the first two parts of the thesis, a case study approach is used to examine the role of wintertime foehn winds and summertime cloud phase on the surface energy balance (SEB) of Larsen C, and therefore surface melting. Firstly, wintertime foehn events are shown for the first time to drive significant and unseasonal surface melting by greatly enhancing surface sensible heat fluxes. Secondly, it is demonstrated that cloud phase, and particularly liquid water content, strongly influences the SEB and surface melting. More accurate model representations of cloud phase are shown to reduce biases in SEB terms and melt. As part of this work, an optimised MetUM configuration is developed for the Antarctic Peninsula. Thirdly, the final part of the thesis presents and analyses a novel, multi-decadal (1998- 2017) model hindcast for Larsen C. The hindcast reproduces observed patterns of foehn-driven melt, making it one of the first long model simulations to do so. Solar radiation is the dominant driver of melting, but cloud phase is shown to determine its extent and duration via feedbacks on temperature and energy fluxes, and foehn winds are especially important for producing melt in non-summer seasons. Large-scale patterns of climate variability like the Southern Annular Mode (SAM) establish conditions for foehn- and cloud-mediated melting to occur. This advanced understanding of processes contributing to surface melting on Larsen C establishes a baseline for future projections. If recent trends towards a more positive SAM and higher temperatures continue in future, surface melting could increase enough to destabilise the ice shelf, potentially contributing to sea level rise.

Details