The Parkfield segment of the San Andreas fault has been intensively monitored with a broad array of geophysical instrumentation in anticipation of recording the next in a series of earthquakes that repeat on average approximately every 22 years. Although the official prediction for the occurrence of the event in 1988 +- 5 years (Bakun and Lindh, 1985) failed, fortunately the monitoring efforts continued. On September 28, 2004 a magnitude 6 earthquake occurred, and it has produced a wealth of new data to study the mechanics of the faulting process both coseismically (at the time of the mainshock) and postseismically. The 2004 Parkfield earthquake is arguably the best recorded to date with observations of near-fault strong ground motion, continuous GPS, surface creep, borehole strain, ground water and EM signals, as well as regional broadband waveform data. In the proposed research a high-resolution model of the rupture process, namely the distribution of fault slip in both space and time will be determined from local and more distant seismic instruments as well as surface deformation obtained from GPS. Additionally a postseismic slip model will be obtained from the continuous GPS data. These high-resolution models will then be used to calculate the stress change on the fault and relate it to the behavior of aftershocks, characteristically repeating earthquakes (whose behavior was changed by the occurrence of the mainshock), and postseismic slip giving insight into the fault zone rheology. Regional and teleseismic records of the previous Parkfield earthquakes in 1922, 1934, and 1966 are available and will be compared with similar records for the 2004 event to evaluate the characteristic earthquake hypothesis for large events in the region. This project will address fundamental questions about the relationships between mainshocks, aftershocks, pre-event seismicity, and post-seismic deformation. It will also address the question of source parameter resolution of fault slip inversions with the goal of better understanding underlying rupture dynamics and the earthquake hazard application of near-realtime ground motion reporting.
