This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5)
Large faults that generate earthquakes exhibit a wide range of slip behaviors including stick-slip motion resulting in episodic earthquakes, stable and steady sliding without fast earthquake slip, and oscillatory or non-steady slip sliding without earthquakes. This range in slip behaviors can be understood and modeled through the application of friction laws derived from laboratory sliding experiments with natural or synthetic rock samples. The friction parameters obtained from these experiments are implemented in numerical models of faults imbedded in the earth in order to predict the slip behavior of faults. Small changes in the friction parameters implemented in these models can lead to dramatically different sliding behaviors. For example, small changes in parameters can change from stick-slip, earthquake producing slip behavior to steady, stable sliding behavior without earthquakes. Knowledge of the distribution of these parameters on faults is therefore essential for earthquake hazard assessment.
The object of this work is to use measurements of motions of the Earth?s surface near faults to infer friction parameters on faults. Measurements from the Hayward fault in the eastern San Francisco Bay Area and from the San Andreas fault near San Juan Bautista will be compared with numerical models of fault slip. Data from these locations recorded transient slip episodes on the faults triggered by distant earthquakes. Also, various existing methodologies for modeling fault creep will be systematically examined to assess the influence of simplifying assumptions on model inferences and therefore provider a clearer understanding of how well the frictional properties of faults can be resolved using such approaches.
Models incorporating rate-state friction have been remarkably successful in explaining a wide range of observations reflecting the nature of fault slip including afterslip, decay rates of aftershocks, episodic slip at subduction zones, spontaneous silent earthquakes, as well as general features of the earthquake cycle. Numerical studies invoking rate-state friction employ laboratory-derived friction parameters which are often assumed, for simplicity, to vary only with depth. It remains unclear to what extent the laboratory friction values are relevant to real faults.
