-
Surface elevation errors in finite element Stokes models for glacier evolution
Authors:
Ed Bueler
Abstract:
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 functi…
▽ More
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.
△ Less
Submitted 19 September, 2024; v1 submitted 12 August, 2024;
originally announced August 2024.
-
Performance analysis of high-resolution ice sheet simulations
Authors:
Ed Bueler
Abstract:
Numerical ice sheet models compute evolving ice geometry and velocity fields using various stress-balance approximations and boundary conditions. At high spatial resolution, with horizontal mesh/grid resolutions of a few kilometers or smaller, these models usually require time steps shorter than climate-coupling time scales because they update ice thickness after each velocity solution. High-resol…
▽ More
Numerical ice sheet models compute evolving ice geometry and velocity fields using various stress-balance approximations and boundary conditions. At high spatial resolution, with horizontal mesh/grid resolutions of a few kilometers or smaller, these models usually require time steps shorter than climate-coupling time scales because they update ice thickness after each velocity solution. High-resolution performance is degraded by the stability restrictions of such explicit time-stepping. This short note, which considers the shallow ice approximation and Stokes models as stress-balance end members, attempts to clarify numerical model performance by quantifying simulation cost per model year in terms of mesh resolution and the number of degrees of freedom. The performance of current-generation explicit time-stepping models is assessed, and then compared to the prospective performance of implicit schemes. The main result, Table 2, highlights the key roles played by the algorithmic scaling of stress-balance and implicit-step solvers.
△ Less
Submitted 4 July, 2022; v1 submitted 28 June, 2022;
originally announced June 2022.
-
The shallow shelf approximation as a "sliding law" in a thermomechanically coupled ice sheet model
Authors:
Ed Bueler,
Jed Brown
Abstract:
The shallow shelf approximation is a better ``sliding law'' for ice sheet modeling than those sliding laws in which basal velocity is a function of driving stress. The shallow shelf approximation as formulated by \emph{Schoof} [2006a] is well-suited to this use. Our new thermomechanically coupled sliding scheme is based on a plasticity assumption about the strength of the saturated till underlyi…
▽ More
The shallow shelf approximation is a better ``sliding law'' for ice sheet modeling than those sliding laws in which basal velocity is a function of driving stress. The shallow shelf approximation as formulated by \emph{Schoof} [2006a] is well-suited to this use. Our new thermomechanically coupled sliding scheme is based on a plasticity assumption about the strength of the saturated till underlying the ice sheet in which the till yield stress is given by a Mohr-Coulomb formula using a modeled pore water pressure. Using this scheme, our prognostic whole ice sheet model has convincing ice streams. Driving stress is balanced in part by membrane stresses, the model is computable at high spatial resolution in parallel, it is stable with respect to parameter changes, and it produces surface velocities seen in actual ice streams.
△ Less
Submitted 19 October, 2008;
originally announced October 2008.
-
An exact solution to the temperature equation in a column of ice and bedrock
Authors:
Ed Bueler
Abstract:
We construct a solution of a pure conduction problem in ice and bedrock. This solution is suitable for, and is used in this paper for, verifying a numerical scheme for conservation of energy. This exact solution forms one of a suite of verification tests for PISM (Parallel Ice Sheet Model; https://gna.org/projects/pism/).
We construct a solution of a pure conduction problem in ice and bedrock. This solution is suitable for, and is used in this paper for, verifying a numerical scheme for conservation of energy. This exact solution forms one of a suite of verification tests for PISM (Parallel Ice Sheet Model; https://gna.org/projects/pism/).
△ Less
Submitted 5 October, 2007;
originally announced October 2007.
-
On exact solutions and numerics for cold, shallow, and thermocoupled ice sheets
Authors:
Ed Bueler,
Jed Brown
Abstract:
This three section report can be regarded as an extended appendix to (Bueler, Brown, and Lingle 2006). First we give the detailed construction of an exact solution to a standard continuum model of a cold, shallow, and thermocoupled ice sheet. The construction is by calculation of compensatory accumulation and heat source functions which make a chosen pair of functions for thickness and temperatu…
▽ More
This three section report can be regarded as an extended appendix to (Bueler, Brown, and Lingle 2006). First we give the detailed construction of an exact solution to a standard continuum model of a cold, shallow, and thermocoupled ice sheet. The construction is by calculation of compensatory accumulation and heat source functions which make a chosen pair of functions for thickness and temperature into exact solutions of the coupled system. The solution we construct here is ``TestG'' in (Bueler and others, 2006) and the steady state solution ``Test F'' is a special case. In the second section we give a reference C implementation of these exact solutions. In the last section we give an error analysis of a finite difference scheme for the temperature equation in the thermocoupled model. The error analysis gives three results, first the correct form of the Courant-Friedrichs-Lewy (CFL) condition for stability of the advection scheme, second an equation for error growth which contributes to understanding the famous ``spokes'' of (Payne and others, 2000), and third a convergence theorem under stringent fixed geometry and smoothness assumptions.
△ Less
Submitted 16 October, 2006;
originally announced October 2006.
-
Computation of a combined spherical-elastic and viscous-half-space earth model for ice sheet simulation
Authors:
Ed Bueler,
Craig S. Lingle,
Jed A. Kallen-Brown
Abstract:
This report starts by describing the continuum model used by Lingle & Clark (1985) to approximate the deformation of the earth under changing ice sheet and ocean loads. That source considers a single ice stream, but we apply their underlying model to continent-scale ice sheet simulation. Their model combines Farrell's (1972) elastic spherical earth with a viscous half-space overlain by an elasti…
▽ More
This report starts by describing the continuum model used by Lingle & Clark (1985) to approximate the deformation of the earth under changing ice sheet and ocean loads. That source considers a single ice stream, but we apply their underlying model to continent-scale ice sheet simulation. Their model combines Farrell's (1972) elastic spherical earth with a viscous half-space overlain by an elastic plate lithosphere. The latter half-space model is derivable from calculations by Cathles (1975). For the elastic spherical earth we use Farrell's tabulated Green's function, as do Lingle & Clark. For the half-space model, however, we propose and implement a significantly faster numerical strategy, a spectral collocation method (Trefethen 2000) based directly on the Fast Fourier Transform. To verify this method we compare to an integral formula for a disc load. To compare earth models we build an accumulation history from a growing similarity solution from (Bueler, et al.~2005) and and simulate the coupled (ice flow)-(earth deformation) system. In the case of simple isostasy the exact solution to this system is known. We demonstrate that the magnitudes of numerical errors made in approximating the ice-earth system are significantly smaller than pairwise differences between several earth models, namely, simple isostasy, the current standard model used in ice sheet simulation (Greve 2001, Hagdorn 2003, Zweck & Huybrechts 2005), and the Lingle & Clark model. Therefore further efforts to validate different earth models used in ice sheet simulations are, not surprisingly, worthwhile.
△ Less
Submitted 7 June, 2006;
originally announced June 2006.
