Numerical Stabilisation of Anisotropic Ice-Sheet Models

Funding

This project has received funding from the European Union’s Horizon Europe research and innovation programme under grant agreement No 101284480.

Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Executive Agency (REA). Neither the European Union nor the granting authority can be held responsible for them.

About

The largest source of uncertainty in sea-level projections is the Antarctic Ice Sheet and can partly be attributed to the computational restrictions of modelling such large bodies of ice. One aspect of ice-sheet modelling that is particularly restrictive is the small time-step size needed to ensure numerical stability. This in turn limits the feasible range in spatial resolution and the physical complexity of ice-sheet models. However, the recent development of numerical stabilisation schemes for ice-sheet models that allow larger time step sizes has enabled a re-allocation of computational resources into improving the physical complexity of ice-sheet models.

What is ice fabric?

Ice is typically modelled as a shear-thinning, non-Newtonian fluid with an isotropic viscosity, meaning that the internal friction of ice is modelled as being direction-independent. However, the viscosity of ice is, in reality, direction-dependent and ranges by a factor of ~100. This intrinsic physical property of ice is not typically incorporated in continental-scale ice-sheet models due to numerical stability issues in fast-flowing ice and limited geophysical data for model validation.

What is the aim of NumAniso?

The aim of this project is to develop a numerically stable anisotropic ice-sheet model capable of large-scale simulations including fast-flowing ice. This coupled system of equations is difficult to solve because of the need to couple a multidimensional advection equation that models the evolution of fabric to the momentum and mass conservation equations. This exacerbates the numerical issues that typically restrict the time-step size of ice-sheet simulations. However, the implementation of novel methods including numerical stabilisation schemes that treat certain terms in the equations implicitly will allow for greater numerical stability and applicability of anisotropic ice-flow models in fast-flowing ice. NumAniso will build on existing open-source codes in the finite element software Elmer/Ice, FEniCS and Firedrake to provide an ice-sheet model that is physically more accurate and numerically stable.

Who is involved?

This project brings together ice-sheet modellers and numerical analysis experts in the UK and internationally.