Study of the Parkfield section of the San Andreas Fault, California, using a new microearthquake catalog

We present a new catalog for the Parkfield section of the San Andreas Fault, California, for the 6 months before and after the 2004 M6 Parkfield earthquake. Parkfield is one of the best seismically instrumented regions and lies in a transition zone between a creeping section to the northwest and a locked section to the southeast. Recent studies have shown that earthquake catalogs are incomplete, including in Parkfield, due to missing lower magnitude events, particularly in the aftershock period of large mainshocks. The matched filter technique is a well-established method to improve catalogs using known events to search for similar waveforms. Building upon previous studies in Parkfield, we perform a long-term matched filter detection of microearthquakes using waveforms of the High-Resolution Seismic Network (HRSN) and 4052 events as templates to improve current catalogs. Detection was performed filtering data between 2-8 Hz. Taking advantage of an additional 250-sample data recorded by the HRSN, we performed additional detections in the 10-40 and 70-90 Hz frequency bands in the 7 days after the mainshock. The high-frequency detections reduce the interference between overlapping arrivals, resulting in a more complete catalog with detections in the first hundred seconds after the mainshock, and four times more detections than the 2-8 Hz filtered detections. The new catalog was relocated using waveform cross-correlation location package XCORLOC. This new relocated catalog will be used to further investigate and understand the migration patterns of the aftershock sequence, how stress changes imposed from regional and remote earthquakes affect local seismicity, such as the 2003 M6.5 San Simeon earthquake and the 2004 M9.2 Sumatra earthquake. In addition, we plan to use the new catalog to better understand the different behaviors of the locked and creeping sections. Preliminary results of the template catalog show a variable Vp/Vs ratio along the creeping section and a uniform ratio in the locked section. Updated results will be presented at the meeting.