The 2014 Mw6.1 South Napa Valley earthquake, an energetic event with shallow asperities and rapid postseismic slip

The 2014 South Napa Earthquake is the largest inland event in northern California since the 1989 Mw6.9 Loma Prieta earthquake. The earthquake was well recorded by the seismic, geodetic (i.e., InSAR and GPS) instruments as well as field observations which provide excellent data sets to investigate the kinematic slip distribution and postseismic deformations. The two InSAR images, with date pairs of 2014/07/26-2014/08/27 and 2014/08/07-2014/08/31, show very consistent deformation patterns that can be well explained by a slip model with two major asperities of similar peak slip amplitude. The static inversion prefers eastward dipping fault geometry with dip angles of about 82 degree. The hypocenter location of the earthquake was relocated by using the P-wave arrivals on the nearest strong motion stations. The refined epicenter is located at about 1km to the east of the major surface rupture, indicating a high angle east dipping fault geometry which is consistent with the static inversion results. Despite its large hypocenter depth (11km), finite fault inversion using strong motion waveform data displays two major asperities centroid at the depth of 5km (first) and 3 km (second), and is about 7km and 11km north of the hypocenter along the strike, respectively. The aftershock locations from the 1D catalog, however, are mainly distributed 1 to 3 km to the west of the surface rupture, at the depth of 8 to 12 km. The risetimes in the kinematic slip model are mostly less than 1s with peak slip of about 1m, indicating a relatively fast slip rate (~1m/s), given its shallow location. The slip distribution is similar as in the static inversion, but has much smaller peak amplitude for the first asperity. Considering the rapid postseismic deformation observed in field, the larger slip amplitude in the static inversion might be contributed from the rapid postseismic slip at the depth of about 5km around the first asperity. The 1D velocity models used in the inversion have been verified with the M3.9 aftershock, which is also used to estimate the effect of 3D velocity structure near the source by 3D FD calculation. The comparison between 1D and 3D synthetics indicate that some basin paths can be better explained by the 3D velocity structure, which also produces better waveform fits for the mainshock with a kinematic slip model as input.