The PSU/UofC finite-element thermomechanical flowline model of ice-sheet evolution

Abstract

Ice-sheet modeling to understand past changes and project future ones remains limited by uncertainties in key parameters as well as by shortage of computational resources. A fast two-dimensional dynamic/thermodynamic flowline ice-sheet model has been developed and benchmarked to complement three-dimensional models by allowing a more-thorough exploration of parameter space. A nonlinear Glen flow law with an exponent equal to 3 is assumed for ice. A diffusion formulation of the continuity equation for mass balance of ice sheets is employed. The ice-sheet model is coupled to an elastic lithosphere/relaxed asthenosphere isostatic bedrock model. Snow and superimposed ice thicknesses are determined using advective continuity equations. These thicknesses are then used in the parameterization of the surface accumulation and ablation rates under specified meteorological conditions constrained by ice- and ocean-core data. Heat-flow continuity is maintained by the time-dependent advection/diffusion equation in the ice and time-dependent diffusion equation in the underlying rock. We conduct the first benchmarking of which we are aware of a 2-D model against the European Ice-Sheet Modeling Initiative's (EISMINT) Level 2 and Level 3 intercomparison “Greenland” experiments. Appropriate behavior is simulated, with deviations from results of a three-dimensional model that are fully explainable as arising only from the change in dimensionality. Basal temperatures are quite accurate when compared to data and results from 3-D models.