High-resolution climate modelling of Antarctica and the Antarctic Peninsula

In this thesis we have used a high-resolution regional atmospheric climate model (RACMO2.3) to simulate the present-day climate (1979-2014) of Antarctica and the Antarctic Peninsula. We have evaluated the model results with several observations, such as in situ surface energy balance (SEB) observations by automatic weather stations and stake measurements of surface mass balance (SMB). After a recent physics update within the model, we show that the model is now better capable of realistically simulating the complex climate of the Antarctic ice sheet than a previous version of the model. The most important update is the inclusion of a scheme that allows for super-saturation of ice clouds, leading to the transportation of more clouds into the interior of the continent. As a result, modelled snowfall rates and downwelling longwave radiation have increased, and the model better represents the SEB and the SMB in East Antarctica. We have specifically applied RACMO2.3 to the Antarctic Peninsula (AP). For this purpose, the model is for the first time used at a high spatial resolution of 5.5 km in order to accurately represent the complex AP topography. RACMO2.3 is coupled to a sophisticated firn densification model that calculates processes in the snowpack such as meltwater percolation, refreezing and runoff into the ocean. By a comparison with weather stations and weather balloons, we first show that the model is capable of simulating AP wind and temperature. Secondly, we present a high-resolution estimate of the climatological (1979-2014) SMB of the AP. We show that the model is capable of resolving the distinct differences in snowfall rates between the western AP (WAP) and the eastern AP. For instance, the WAP receives nearly 80% of all AP snowfall, including snowfall rates of up to 40 meter per year. When integrated over the AP including ice shelves (an area of 410000 km2), the model calculates an average SMB of 351 gigaton per year (with an interannual variability of 58 gigaton per year) from 1979 to 2014, which mostly consists of snowfall (363 ± 56 gigaton per year). The other SMB components, sublimation, drifting snow erosion and meltwater runoff, are small in comparison (11, 0.5 and 4 gigaton per year, respectively). Finally, we applied the model to the northwestern AP, one of the most rapidly changing regions on Earth. In order to provide improved insight in the effects of these changes, we present the characteristics of the WAP meteoric freshwater budget. This budget consists of snowfall, rainfall, meltwater runoff and glacial discharge into the ocean. Changes in these processes have implications for the marine ecosystem, the surrounding sea-ice, ocean circulation and, ultimately, sea-level rise. Here, we present modelled estimates of the spatial and temporal variability of the freshwater budget, including a first estimate of glacial discharge into the ocean, assuming balance between snow accumulation and glacial discharge. We find that due to this process, locally between 2 and 11 gigaton of ice is lost to the ocean per year