CLAY SEPARATION AND X-RAY DIFFRACTION FOR REFINED K/AR DATING OF FAULT MOTION ON THE CREEPING CENTRAL SECTION OF THE SAN ANDREAS FAULT

While the northern and southern locked sections of the San Andreas Fault are associated with large earthquakes, its central creeping section has not historically hosted rupture. However, new dating and biomarker maturity data on core from the San Andreas Fault Observatory at Depth (SAFOD) provide evidence that large paleoearthquakes have occurred geologically recently along the central section (Coffey et al., 2022). Diffusion experiments on these samples suggest that the ages are only partially reset. Thus, actual event ages must be younger than bulk apparent ages.

To refine earthquake ages, we dated size separates of eight of the original samples using K/Ar geochronometry. Applying hydrodynamic settling, we split each sample into five fine size fractions less than 2 µm. Chemical composition and illite crystallinity were determined via benchtop and synchrotron x-ray diffraction. Ar concentrations from signal standardization and ⁴⁰Ar/³⁶Ar ratios were measured on a noble gas mass spectrometer to obtain the concentration of radiogenic ⁴⁰Ar (⁴⁰Ar*). The same aliquots were spiked with ⁴¹K, then put through column chromatography before isotope ratio measurement for precise determination of K concentration.

Characterization of the 10-angstrom 001 peak of illite confirms that in samples with high thermal maturity and young bulk ages, the finest size fractions host the most authigenic illite, interpreted to form during fault slip (Schleicher et al., 2010). Consistent with this inverse relationship between size and authigenic fraction, size is inversely related to K content and directly related to age in these samples. Calculating a York regression between age and detrital illite abundance, we extrapolate the age at which there was 0% detrital and 100% authigenic illite — the formation age. Samples with high thermal maturity record age intercepts as young as 0.67 ± 1.8 Ma (SEM).

Not only are the new ages of authigenic illite measured here younger than bulk sample ages, they are young enough to suggest that translation of this material from the locked southern San Andreas Fault into the creeping section did not occur. Therefore, our results imply that the evidence of earthquake heating identified in Coffey et al. (2022) was a consequence of large earthquakes propagating into the creeping section within the last million years.