Surface elevation errors in finite element Stokes models for glacier evolution

• Published in: arXiv.org
• Main Author: Bueler, Ed
• Format: Paper
• Language: English
• Published: Ithaca Cornell University Library, arXiv.org 19.09.2024
• Subjects: Banach spaces; Errors; Glaciers; Glaciology; Ice cover; Ice formation; Inequality; Mass balance; Stokes flow
• ISSN: 2331-8422

The primary data which determine the evolution of glaciation are the bedrock elevation and the surface mass balance. From this data, which we assume is defined over a fixed land region, the glacier's geometry solves a free boundary problem which balances the time derivative of the surface elevation, the surface velocity from the Stokes flow, and the surface mass balance. A surface elevation function for this problem is admissible if it is above the bedrock topography, equivalently if the ice thickness is nonnegative. This free boundary problem can be posed in weak form as a variational inequality. After some preparatory theory for the glaciological Stokes problem, we conjecture that the continuous space, implicit time step variational inequality problem for the surface elevation is well-posed. This conjecture is supported both by physical arguments and numerical evidence. We then prove a general theorem which bounds the numerical error made by a finite element approximation of a nonlinear variational inequality in a Banach space. The bound is a sum of error terms of different types, essentially special to variational inequalities. In the case of the implicit step glacier problem these terms are of three types: errors from discretizing the bed elevation, errors from numerically solving for the Stokes velocity, and finally an expected quasi-optimal finite element error in the surface elevation itself.