A Mechanistic Laboratory Investigation of Seismic Preslip

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

This research addresses the question of what leads up to fault sliding - why today and not tomorrow? The research consists of laboratory experiments on a highly controlled frictional fault that will help us understand the formation of earthquakes on natural faults. Fault surfaces are made of of tiny contact points - asperities - and they individually slip and transfer stress to other asperities in the run-up to faulting. In the field these are known as foreshocks. In the laboratory we have control of nearly all variables so we can focus on the behaviors of interest. Glaser conical displacement sensors will be used to very accurately measure the seismic signals from the asperity dynamics and invert back to how the foreshocks lead to earthquakes. In contrast, in the field it is nearly impossible to observe the physical interaction of the fault surfaces leading up to an earthquake, so indirect observation methods such as finite source inversion use sensors at the Earth's surface to infer the precursory events taking place underground. This project will improve the understanding of earthquake source mechanisms by comparing direct observations made possible in the laboratory to the indirect methods available to seismologists, and will shed light on what leads to an earthquake. This project will elucidate the mechanisms behind three currently important questions in understanding earthquake precursors: What are the mechanistic components of preslip? What are the effects on rupture initiation from decreasing normal stress on faults? What are the mechanisms causing extended high frequency content in slow slip earthquakes and in ruptures following extended healing periods? A sequence of tests will be conducted to examine the contact mechanisms of the fault interface leading to gross rupture. This will shed light on so-called unusual phenomena, including episodic tremor, low and very-low frequency earthquakes and slow slip events. Previous work by the principal investigator has shown tantalizing indications that the asperity-level mechanisms that lead to higher frequency content due to extended healing time, and for slow-slip events might be explained. The laboratory-scale earthquakes, using PMMA as an analog of ductile rock, has provided data hinting that the same premonitory behavior appears in the experiment scaled to the much smaller size of our fault and contact asperities, e.g., preslip of minutes duration. Use of a high-speed video camera and pressure sensitive film allows for direct observation of the fault surfaces before, during, and after simulated mini-earthquakes. Seismological techniques are used to interpret data from an array of ultrasonic transducers and slip sensors adjacent to the fault. The combination of these direct and indirect observations will provide new insights as to the nucleation and propagation of slip, with a focus on the strength of asperities and interaction between them.