Long-term simulations of Antarctic ice sheet evolution can help us better understand its climatic drivers and constrain past southern hemisphere ice volume. So far, such long simulations have used proxy-based parameterizations of climatic drivers, presuming that external forcings are synchronous and spatially uniform. To improve on this, we use a transient, three-dimensional climate simulation over the last eight glacial cycles to drive an Antarctic ice sheet model. For the climate simulation, we use the intermediate complexity Earth system model LOVECLIM, which was forced with time-evolving orbital parameters, greenhouse gas concentrations, and northern hemisphere ice sheet volume. The ice sheet simulation was performed with the Penn State University Ice Sheet Model (PSU-ISM).

Penn State Ice Sheet Model

Ice sheet model – PSU-ISM has been used extensively for long-term simulations of past and future evolution of the Antarctic ice sheet. It uses a hybrid combination of the scaled shallow-ice and shallow-shelf approximations, and calculates the position of the grounding line using an ice flux parameterization. In the model, melting at the ice-ocean interface is parameterized using a direct dependence on 400-m ocean temperatures; for melting at the ice surface, a Positive Degree Day scheme is applied to the seasonal cycle of temperature. The model also includes recently developed treatments of hydrofracturing and ice cliff failure, which lead to increased calving rates in areas of high surface melt. Our simulation is performed on a polar stereographic grid at 20-km resolution.

For present-day simulations, PSU-ISM uses observed annual mean climatologies of atmospheric temperature and accumulation with a parameterized sinusoidal seasonal cycle, and observed annual mean ocean temperatures at 400-m depth. To these present-day climatologies, we add monthly atmospheric temperature and annual mean ocean temperature anomalies from the LOVECLIM simulation. For accumulation, the ratio of past to present-day precipitation is used to avoid negative values. Lapse-rate corrections are applied to the atmospheric variables to account for changes in ice-sheet elevation.

Model description Pollard, D., & DeConto, R. M. (2012). Description of a hybrid ice sheet-shelf model, and application to Antarctica. Geoscientific Model Development, 5(5), 1273.

Penn State Ice Sheet Model documentation


Climate model – LOVECLIM has been used for simulations of both present-day and paleo-climate. It consists of a quasi-geostrophic three-level atmospheric model at T21 spectral resolution, coupled to an ocean general circulation model with 3° horizontal resolution and twenty vertical levels. The model includes a thermodynamic-dynamic sea ice model and terrestrial vegetation component. It was forced with time-evolving orbital parameters; ice core reconstructions of CO2, CH4, and N2O; and time-evolving northern hemisphere ice sheets, using an acceleration factor of five. In the LOVECLIM simulation, the Antarctic ice sheet remains set to its present-day size.

Model description Goosse, H., Brovkin, V., Fichefet, T., Haarsma, R., Huybrechts, P., Jongma, J., et al. (2010). Description of the Earth system model of intermediate complexity LOVECLIM version 1.2. Geoscientific Model Development, 3, 603-633.

LOVECLIM model documentation