The Elizabeth Lake paleoseismic site: Rupture pattern constraints for the past ~800 years for the Mojave section of the south-central San Andreas Fault

The southern San Andreas Fault in California has hosted two historic surface-rupturing earthquakes, the ~M7 1812 Wrightwood earthquake and the ~M7.9 1857 Fort Tejon earthquake (e.g., Sieh, 1978; Jacoby et al., 1988). Numerous paleoseismic studies have established chronologies of historic and prehistoric earthquakes at sites along the full length of the 1857 rupture (e.g., Sieh, 1978; Scharer et al., 2014). These studies provide an unparalleled opportunity to examine patterns of recent ruptures; however, at least two significant spatial gaps in high-quality paleoseismic sites remain. At ~100 km long each, these gaps contribute up to 100 km of uncertainty to paleo-rupture lengths and could also permit a surface rupture from an earthquake up to ~M7.2 to go undetected [using scaling relationships of Wells and Coppersmith (1994)]. Given the known occurrence of an ~M7 earthquake on this portion of the SAF (1812), it is critical to fill these gaps in order to better constrain paleo-rupture lengths and to increase the probability of capturing the full spatial record of surface rupturing earthquakes.  

In this study, we target a new site within the 100 km long stretch of the San Andreas Fault between the Frazier Mountain and Pallett Creek paleoseismic sites (Figure 1), near Elizabeth Lake, California. Prior excavations at the site during 1998-1999 encountered promising stratigraphy but these studies were hindered by shallow groundwater throughout the site. We began our current phase of investigations in 2012, targeting the northwestern end of a 40 x 350 m fault-parallel depression that defines the site (Figure 2). Subsequent investigations in 2013 and 2014 focused on the southeastern end of the depression where the fault trace is constrained between topographic highs and is proximal to an active drainage. In total, our paleoseismic investigations consist of 10 fault-perpendicular trenches that cross the depression (Figure 2) and expose a >2000 year depositional record. These trenches reveal that the thickest section of young stratigraphy occurs at the southeastern end of the site where the fault zone projects through an area of relatively continuous sediment accumulation from a northeast-flowing drainage. This portion of the site contains a 3-m-wide pop-up structure within the fault zone that separates alternating alluvial and paludal deposits south of the fault zone from a thick organic-rich loam on the north side of the fault zone. Faults, fissures, and tilted blocks provide evidence for 4 to 5 paleoearthquakes since ca. 1250 A.D. Radiocarbon dating established that the site has a significant component of detrital charcoal producing an age spread of up to 500 years. To supplement our age chronology we incorporated ages from collections of micro-scale organic fractions and post-IR infrared stimulated luminescence dating in order to better estimate true layer ages.