Geometry Of The San Andreas Fault In The Salton Trough And Its Effect On Simulated Shaking For A Rupture Similar To That Of The Great California Shakeout Of 2008

The southernmost San Andreas fault (SAF) zone, in the northern Salton Trough, is considered likely to produce a large-magnitude, damaging earthquake in the near future (Jones et al., 2008, USGS OFR). The geometry of the SAF and adjacent sedimentary basins will strongly influence energy radiation and strong ground motion during a future rupture. The Salton Seismic Imaging Project (SSIP) was undertaken, in part, to provide more accurate information on SAF and basin geometry in this region. We report interpretations of seismic profiles in the Salton Trough (Lines 4-7). Lines 4 through 6 are SW-NE fault-perpendicular profiles that cross the Coachella Valley and extend into the mountain ranges on either side. Line 7 crosses the Salton Sea and sedimentary basin deposits similar to those of the Coachella Valley to the north. On three lines (4, 6, 7), seismic imaging, potential-field studies, and (or) earthquake hypocentral relocations provide evidence that active strands of the SAF dip moderately NE. Importantly, on Line 4, we have obtained a reflection image of the SAF zone, in the depth range of 5-10 km, that coincides with the microearthquake pattern here (Hauksson et al., 2012, BSSA). We interpret a moderate northeast dip (~60 deg.) for the SAF, as previously reported by Fuis et al. (2012, BSSA). We used a 3D finite-difference wave propagation method to model shaking in southern California expected from rupture on the SAF with the propeller-shaped geometry reported by Fuis et al. (2012), and we have compared this shaking to that modeled from the generally vertical geometry used in the Great California ShakeOut (Jones et al., 2008). Our results were obtained by projecting the kinematic 2008 ShakeOut rupture onto the newly characterized, dipping SAF geometry. We estimate higher levels of shaking on the hanging wall of the SAF (to the NE) and lower levels on the footwall (to the SW), as compared to the 2008 estimates. The change in ground motion level for most shaking quantities (PGA, PGV, 0.3 to 3 sec SA [spectral acceleration]) averages about +/- 20%. Southeast of Cajon Pass, where the change in fault dip between the propeller and 2008 ShakeOut geometries ranges from 10 to 50 deg., new modeled ground motions are systematically increased on the NE side by as much as a factor of 2 and decreased on the SW side by a similar amount compared to the shaking estimated in the 2008 ShakeOut rupture scenario. For the region northwest of Cajon Pass, this pattern is reversed despite the fact that the fault geometry is basically the same for the two models (vertical dip). We interpret this pattern to result from radiation pattern and directivity effects being carried northward from the southern portions of the fault.