Exciting news! We're transitioning to the Statewide California Earthquake Center. Our new website is under construction, but we'll continue using this website for SCEC business in the meantime. We're also archiving the Southern Center site to preserve its rich history. A new and improved platform is coming soon!

Poster #025, San Andreas Fault System (SAFS)

Detecting creep event propagation along the San Andreas Fault

Dan B. Gittins, & Jessica C. Hawthorne
Poster Image: 

Poster Presentation

2020 SCEC Annual Meeting, Poster #025, SCEC Contribution #10519 VIEW PDF
The San Andreas Fault creeps at the surface along a 150 km-long section between San Juan Bautista and Parkfield. This creep occurs in bursts known as creep events. The creep events are well-recorded using decades-long USGS creepmeter records, but they are typically examined only at the point locations of these instruments. Thus, we do not know how large an area of the fault creeps during these events. Further, we do not know which the physical process causes creep events. Several models have been considered, including distributed shear (Wesson, 1988), size-limited frictionally weakening patches (Scholz, 1998; Wei et al., 2013), and shear-induced dilatancy (Segall et al., 2010).

> Here we seek to better constrain and understand creep events along the San Andreas by (1) identifying the events in the creepmeter record, (2) determining the along-strike length of creep events, and (3) examining the slip rate evolution during the creep event onsets.

To detect creep events, we use a cross-correlation approach. We pick a template creep event for each creepmeter which has a well-defined shape. Then we identify portions of the creep record that are similar to the template creep event. Finally, we check that these creep intervals have slip that is significantly larger than the instrumental noise. By isolating intervals with high similarity and significant slip, we successfully detect at least 99% of the creep events identified in a visual inspection.

Using identified creep events from neighbouring creepmeters XSJ, XHR and CWN, at the northern end of the creeping section, we estimate the length of creep events. Some creep events that are detected at both XHR and CWN, creepmeters located 4 km away from each other. These creep events, which represent at least 10-17% of the events at XHR, must have an along-strike length of at least 4 km. This along-strike length is a significant fraction of the along-depth extent of the seismogenic zone. It suggests that shallow creep events may play an important role in the creeping region’s slip dynamics.

Currently, we are now examining the shape of the onset of creep event. We are investigating how slip rate decays with time after the event begins: whether it decays as a power-law, as expected for velocity-strengthening friction, or as an exponential, as expected for distributed viscous shear.