Lawrence Berkeley National Laboratory

A unique data set of seismograms for 720 source-receiver paths has been collected as part of a controlled source Vibroseis experiment San Andreas Fault (SAF) at Parkfield. In the experiment, seismic waves repeatedly illuminated the epicentral region of the expected M6 event at Parkfield from June 1987 until November 1996. For this effort, a large shear-wave vibrator was interfaced with the 3-component (3-C) borehole High-Resolution Seismic Network (HRSN), providing precisely timed collection of data for detailed studies of changes in wave propagation associated with stress and strain accumulation in the fault zone (FZ). Data collected by the borehole network were examined for evidence of changes associated with the nucleation process of the anticipated M6 earthquake at Parkfield. These investigations reported significant traveltime changes in the S coda for paths crossing the fault zone southeast of the epicenter and above the rupture zone of the 1966 M6 earthquake. Analysis and modeling of these data and comparison with observed changes in creep, water level, microseismicity, slip-at-depth and propagation from characteristic repeating microearthquakes showed temporal variations in a variety of wave propagation attributes that were synchronous with changes in deformation and local seismicity patterns. Numerical modeling suggests 200 meters as an effective thickness of SAF. The observed variations can be explained by velocity 6 percent velocity variation within SAF core. Numerical modeling studies and a growing number of observations have argued for the propagation of fault-zone guided waves (FZGW) within a SAF zone that is 100 to 200 m wide at seismogenic depths and with 20 to 40 percent lower shear-wave velocity than the adjacent unfaulted rock. Guided wave amplitude tomographic inversion for SAF using microearthquakes, shows clearly that FZGW are significantly less attenuated in a well-defined region of the FZ. This region plunges to the northwest along the northwest boundary of the region of highest moment release and separates locked and slipping sections of the SAF at depth, as determined independently from geodesy, seismicity and the recurrence rates of characteristically repeating microearthquakes. The mechanism for low FZGW attenuation in the zone is possibly due to dewatering by fracture closure and/or fault-normal compression, or changes in fracture orientation due to a complex stress or strain field at the boundary between creeping and locked zones of the San Andreas Fault. Temporal changes of FZGW correlates with changes in overall seismicity. Active monitoring of changes in FZGW has a potential for imaging and detecting of changes in strees within FZ cores. Since FZGW primarily propagate in the low-velocity core region of fault zones, they sample the most active zone of fault deformation and provide greater structural detail of the inner fault core than body waves which propagate primarily outside of the central core region. FZGW also can be used for FZ continuity studies.