S53A-2754
The Rule of Dynamic Strain to Near Source Aftershock Distribution of the 2014, Mw 6.0, Napa (California) Earthquake

Friday, 18 December 2015
Poster Hall (Moscone South)
Raffaella De Matteis, University of Sannio, Benevento, Italy, Vincenzo Convertito, Istituto Nazionale di Geofisica e Vulcanologia, Napoli, Italy and Antonio Emolo, The University of Naples Federico II, Naples, Italy
Abstract:
The 2014 Napa was recognized as a right‐lateral strike‐slip fault. About 400 aftershocks occurred, mainly in the near‐source range, in the two months after the earthquake. They mostly occurred between 8 and 11 km depth interesting an area of about 10 km2 north‐northwest‐trending with respect to the mainshock hypocenter. However, the aftershock distribution was not able to constrain the mainshock fault plane.

Since Parsons et al. (2014) have shown that Coulomb static stress change does not completely explain near‐source aftershock distribution, we explore whether dynamic strain transfer, enhanced by source directivity, contributed to trigger the aftershock sequence. Indeed, dynamic strain transfer triggering attributes enhanced failure probabilities to increased shear stresses or strains, to permeability changes and maybe to fault weakening. In this respect, we observe that a single inverse power law fits the decay of aftershock density as function of distance from the fault plane, suggesting that dynamic stress/strain might have played a role in the aftershocks triggering. To test this hypothesis, we used Peak‐Ground Velocities (PGVs) as a proxy for peak‐dynamic strain/stress field, accounting for both fault finiteness and source directivity. We first use a point source to retrieve the best parameters of the directivity function from the inversion of the PGVs. Next, the same PGVs are used to jointly infer the surface fault projection and the dominant horizontal rupture direction. Finally, we map the peak‐dynamic strain/stress, modified by source geometry and directivity, to resolve the relationship between the aftershocks location and the

areas of large dynamic strain values. Thus, we believe that dynamic strain/stress actually contributed to the Napa aftershock distribution. Our results may help to better constrain the Napa causative fault and complement Coulomb static stress change to identify areas that will be more likely affected by aftershocks.