A suite of models for studying the dynamics of ice sheets, ice caps and glaciers

A variety of numerical models are needed to study the terrestrial ice masses of the world. The large ice sheets of Greenland and Antarctica are most efficiently modelled using fairly simple models of their internal mechanics (often called zero-order models), which are appropriate at spatial resolutions of 20 km and (in comparison to the models discussed below) are (relatively) computationally undemanding. These models are inappropriate at finer spatial scales (smaller ice caps and glaciers) and where basal slip is important (ice shelves and ice streams). The large ice shelves which fringe Antarctica are modelled using first-order models which incorporate longitudinal stresses. In this case, a further simplifying assumption is made that vertical shear is unimportant (because the underlying ocean does not exert a traction). Similar models are required in the study of ice streams and smaller ice masses, however in these cases vertical shearing cannot be ignored and the vertical variation of all stress components must be considered.

Numerical models of terrestrial ice masses comprise the following components:

1	The prognostic equation for ice thickness evolution incorporating net addition and loss to the ice mass, as well as internal flow.
2	The prognostic evolution for internal temperature evolution incorporating diffusion, advection, dissipation and geothermal heat input.
3	One of a series of increasingly complex models of the internal stress distribution within the ice mass (see above).
4	Rheological models relating the strain rate of ice to the internal stress regime of the ice mass, which are typically based on the Glen flow law, as well as relationships between the viscosity of ice and its temperature.
5	One of a series of increasingly complex models predicting mass gain via the accumulation of snow, and mass loss via the ablation of ice and snow.
6	Models concerning the underlying substrate, its hydrology and the way in which it interacts with the overlying ice mass.

Large-scale numerical models of the Greenland and Antarctic Ice Sheets (including fringing ice shelves) are currently available to the group. Higher-order models of ice flow are actively being developed (a transect model already exists). The objectives of this project are to:

1	Implement parallel versions of large-scale numerical models of the Greenland and Antarctic Ice Sheets.
2	Develop a higher-order model of ice flow suitable for the study of ice caps and glaciers, as well as regional studies of the Greenland and Antarctic Ice Sheets.
3	Trial the application of this model using the Pine Island basin, West Antarctica.
4	Develop a methodology for the use of satellite data (in particular, interferometric SAR ice-surface velocities) to constrain higher-orders of ice flow.