Articles | Volume 10, issue 1
Research article
16 Feb 2016
Research article |  | 16 Feb 2016

Tremor during ice-stream stick slip

B. P. Lipovsky and E. M. Dunham

Abstract. During the 200 km-scale stick slip of the Whillans Ice Plain (WIP), West Antarctica, seismic tremor episodes occur at the ice–bed interface. We interpret these tremor episodes as swarms of small repeating earthquakes. The earthquakes are evenly spaced in time, and this even spacing gives rise to spectral peaks at integer multiples of the recurrence frequency ∼ 10–20 Hz. We conduct numerical simulations of the tremor episodes that include the balance of forces acting on the fault, the evolution of rate- and state-dependent fault friction, and wave propagation from the fault patch to a seismometer located on the ice. The ice slides as an elastic block loaded by the push of the upstream ice, and so the simulated basal fault patch experiences a loading velocity equal to the velocity observed by GPS receivers on the surface of the WIP. By matching synthetic seismograms to observed seismograms, we infer fault patch area ∼ 10 m2, bed shear modulus ∼ 20 MPa, effective pressure ∼ 10 kPa, and frictional state evolution distance ∼ 1 μm. Large-scale slip events often occur twice daily, although skipped events have been increasing in frequency over the last decade. The amplitude of tremor (recorded by seismometers on the ice surface) is greater during the double wait time events that follow skipped events. The physical mechanism responsible for these elevated amplitudes may provide a window into near-future subglacial conditions and the processes that occur during ice-stream stagnation.

Short summary
Small repeating earthquakes occur at the ice-bed interface of the Whillans Ice Stream, West Antarctica. The earthquakes occur as rapidly as 20 earthquakes/s. We conduct numerical simulations of these earthquakes that include elastic and frictional forces as well as seismic wave propagation. We create synthetic seismograms and compare these synthetics to observed seismograms in order to constrain subglacial parameters. We comment on decadal-scale changes in these parameters.