C010-0012
Spontaneous Formation of an Internal Shear Band Facilitates the Flow-to-sliding Transition on Rough Beds Suggesting an Alternative Sliding Law

Tuesday, 8 December 2020
Poster
Emma Liu, Stanford University, Stanford, CA, United States, Ludovic Räss, ETH Zurich, Zurich, Switzerland, Frederic Herman, Institute of Earth Surface Dynamics (IDYST), Faculty of Geosciences and Environment (FGSE), University of Lausanne, Lausanne, Switzerland, Yuri Podladchikov, University of Lausanne, Lausanne, Switzerland and Jenny Suckale, Stanford University, Department of Geophysics, Stanford, CA, United States
Abstract:
The dramatic acceleration of ice surface velocity is a noticeable feature on most ice sheets margins. However, the physical processes triggering this transition remain unclear, limiting the ability of ice flow models to capture this transition accurately. Here, we investigate the role of basal topography in the onset of the flow-to-sliding transition.

We model the thermal-mechanical coupled ice flow over rough basal topography starting from a frozen bed using an iterative nonlinear Stokes solver. We implement the Immersed Boundary Method, a fictitious domain method, to describe the interactions between ice and bed. This method permits the computational grid not to conform to the physical bed shape, releasing researchers from laborious meshing. We allow the ice surface to evolve spatially to account for the thinning process using the level set approach. By resolving the ice-bed interactions and surface elevation changes, we can capture the dynamic development of ice flow with high accuracy.

Our simulation results suggest that one possible cause for the flow-to-sliding transition is the spontaneous formation of an internal shear band due to localized shear heating triggered by roughness in the basal topography. The shear band forms near the bed with an orientation parallel to the ice flow direction. It gradually accommodates the majority of the deformation as ice is flowing downstream and induces a sliding motion of the upper part of the ice domain. The consequence is a gradual transition from ice flow with distributed deformation to localized ice sliding. We compare our simulation results with the field measurements at the West Greenland Ice Sheet and find satisfactory agreement in the shape of computed and measured depth profiles of strain rates. To derive a simple, parametrized representation of the spontaneous creation of an internal sliding interface on top of the basal topography, we propose a new sliding law that captures this internal sliding.