Implications of new paleoseismological data from the Carrizo section of the San Andreas Fault

How does strain accumulated at a plate boundary get released over several thousand years? Is earthquake recurrence time-dependent or time-independent? The Carrizo section of the San Andreas fault (SAF) is potentially an ideal laboratory for addressing these questions because this section of the fault is structurally simple, it is oriented parallel to the overall Pacific plate motion and long-term slip rate of the fault (34+-3 mm/yr) agrees with geodetically determined decadal strain accumulation rate of ~36mm/yr. The Carrizo section of the SAF has been proposed to rupture relatively infrequently, with centuries long recurrence, and only during the largest earthquakes, thereby controlling the occurrence of great quakes on the southern SAF. We will present paleoseismic evidences from five closely spaced and connected fault-perpendicular trenches at the Bidart Fan site along with ~60 C14 dates which indicate an average recurrence interval of ~100 years. The event chronology obtained from the Bidart site trenches contradicts the slip per event data from 3-D excavations at the Wallace Creek site, ~5 km to the northwest, which show that four of the last six earthquakes were 1857-like large slip events. While we do not have a long-enough record of past earthquakes at the Bidart Fan site to test various fault behavior models, we propose that while some of the surface rupturing events at the Bidart Fan site are part of 1857-like, multi-segment ruptures, others are smaller offsets which could be the tail ends of Cholame ruptures extending south into the northern section of Carrizo segment. New sites for paleoseismic studies along the Cholame and southern Carrizo sections of the SAF are needed to test this hypothesis, which has significant scientific and seismic hazard implications.