SCEC Award Number 09187 View PDF
Proposal Category Collaborative Proposal (Integration and Theory)
Proposal Title Laboratory Experiments on Fault Shear Resistance Relevant to Coseismic Earthquake Slip
Investigator(s)
Name Organization
David Goldsby Brown University Terry Tullis Brown University
Other Participants
SCEC Priorities A8, A11, B2 SCEC Groups FARM, N/A, Seismology
Report Due Date 02/28/2010 Date Report Submitted N/A
Project Abstract
 Our research efforts of the past year have focused on further understanding and quantifying the frictional behavior of crustal rocks at near-seismic slip rates. We have focused primarily on obtaining a better understanding of dynamic fault weakening due to flash heating of asperity contacts. In the Instron apparatus at Brown, we have used a sample grip that is less compliant and more massive than the grip used in previous tests. As we described in last year’s report, using this new grip results in negligible oscillations in shear stress due to machine resonance, as has plagued some of our previous tests. We also conducted an extensive series of experiments at Oak Ridge National Laboratory using an established apparatus for studying flash heating, the so-called pin-on-disk (POD) apparatus. These results compliment and corroborate the results obtained using the Instron apparatus, as described below. As in previous years, our research on high-velocity friction reported on here has been jointly supported by SCEC and the USGS.
Intellectual Merit All of the weakening mechanisms that we are studying have profound implications for the magnitude of stress-drops during earthquakes and consequently for the magnitude of strong ground shaking. The manner in which fault strength varies with displacement and rupture velocity, as well as the rate at which healing occurs as slip velocity drops behind the rupture tip, can control the mode of rupture propagation, i.e. as a crack or as a pulse. Furthermore, these data can be important for resolving questions concerning stress levels in the crust. If coseismic friction is low, and seismic data seem to constrain the magnitude of dynamic stress drops to modest values, then the tectonic stress levels must also be modest. We may have a strong crust that is nevertheless able to deform by faulting under modest tectonic stresses if the strength is overcome at earthquake nucleation sites by local stress concentrations and at other places along the fault by dynamic stress concentrations at the rupture front. Thus, understanding high speed friction is important not only for practical matters related to predicting strong ground motions and resulting damage, but also for answering major scientific questions receiving considerable attention and funding, e.g. the strength of the San Andreas fault / the heat-flow paradox, the question that ultimately is responsible for the SAFOD project.
Broader Impacts Our research has implicatons for society relating to the possible magnitudes of strong ground motions. If the large reductions in shear stress seen in our experiments are characteristic of earthquakes, it implies that dynamic stress drops may be nearly complete and that, unless the initial stress is also small, accelerations and strong ground motions should be quite large.
Exemplary Figure Figure 7. Comparison of data from POD tests with those from tests in Instron apparatus. a) Comparison of quartz POD data (blue dots) with those from CVV tests (solid black line) and VS tests (red dots) in the Instron. Dashed blue line shows inferred trend of the POD data. The solid green trace is a theoretical expression of flash weakening behavior from Beeler et al. [2008]. b) Comparison of data from Instron tests on gabbro (solid line) with POD tests on diabase (red dots).