Pinpointing transient aseismic slip at depth with seismological observations

Wednesday, 24 February 2016: 3:00 PM
William Frank1,2, Nikolai Shapiro2,3, Allen L Husker4, Vladimir Kostoglodov4, Mathilde Radiguet5, Baptiste Rousset5, Alexander A Gusev3, Nathalie Cotte5 and Michel Campillo5, (1)Massachusetts Institute of Technology, Department of Earth, Atmospheric, and Planetary Sciences, Cambridge, MA, United States, (2)Institut de Physique du Globe de Paris, Paris, France, (3)Institute of Volcanology and Seismology, Petropavlovsk Kamchatsky, Russia, (4)UNAM National Autonomous University of Mexico, Mexico City, Mexico, (5)ISTerre Institute of Earth Sciences, Saint Martin d'Hères, France
Abstract:
With the recent discovery of slow slip events, tectonic tremor, and low-frequency earthquakes, there are many new fundamental questions about how these slow slip phenomena modify our understanding of the seismic cycle and hazard of megathrust subduction zones. Geodetic observations have been able to constrain the largest slow slip events but are blind to the smaller aseismic slip events that are beneath the noise level. On the other hand, seismology provides indirect observations of slow slip in the form of tectonic tremor and low-frequency earthquakes. We show here that by exploiting the repeating nature of low-frequency earthquakes to investigate the evolution in time and space of these small seismic asperities, seismological data can be used to track aseismic slip. We will first describe how characteristic space-time patterns of seismicity can be used as a guide to align and stack GPS displacement time series to increase the signal to noise ratio and reveal slip events hidden under the noise. We will then show that the distribution of repeating low-frequency earthquakes changes radically in time with the onset of even the smallest slow slip events due to interaction between asperities. Our results demonstrate that bursts of low-frequency earthquakes result from the collective behavior of asperities whose interaction depends on the state of the fault interface.