Rupture process and energy budget of the 29 July 2008 M_w 5.4 Chino Hills, California, earthquake

Abstract

[1] The source model of the 2008 M w 5.4 Chino Hills, California, earthquake is constrained using near-field seismic body waves recorded by the California Integrated Seismic Network (CISN). Finite fault inversions are preformed for the two fault models based on the nodal planes derived from the CISN moment tensor solution. The northeast dipping plane (strike = 289 ; dip = 62 ), which has a similar strike as the nearby Whittier fault, is chosen as the causative fault because it fits the data significantly better. Our inversion result indicates that the majority of the Chino Hills earthquake rupture occurred in a compact area. In particular, 48% of the total seismic moment (1.6 Â 10 17 Nm) was released by the failure of a 1.8 km 2 asperity located east of the hypocenter in a short time window from 0.4 to 0.8 s after the rupture initiation. The average slip is approximately 0.5 m but the maximum slip is 1.8 m. The average rupture velocity is 1.9 km/s. The static stress drop calculated using the slip model is up to 80 MPa and the average stress drop changes from 19 to 38 MPa, depending on the average schemes. The weighted average slip velocity is 6.5 m/s for entire rupture and is 11 m/s for the east asperity. The inferred available energy and radiated energy are 8 Â 10 13 J and 2.5 Â 10 13 J, respectively. Radiation efficiency is then 0.31, which is moderately low compared with previous earthquakes but consistent with the inferred high average fracture energy density, ranging from 6.5 to 14.8 MJ/m 2 .

Ground-Motion Simulation and Validation of the 2008 Chino Hills, California, Earthquake Using Different Velocity Models

“…We adopt model I of the source inversion done by Shao et al (2012), which corresponds to a plane striking at an angle of 289°and dipping to the north at an angle of 62°. This model describes the rupture as occurring on a deep patch at the bottom of the Whittier fault.…”

“…They carried out waveform inversions in the wavelet domain using a simulated annealing method to compute the slip, rake, rupture time, and slip-rate functions. Shao et al (2012) also applied corrections for the amplitude of P-and SH-wave arrivals using amplification factors calculated with the ratios between 3D and 1D synthetics. The 3D synthetics were generated with a spectral element code (Komatitsch et al, 2004) for a model with a numerical accuracy up to 1 Hz and using an earlier version of the CVM-H velocity model used here (see Material Models section).…”

“…This model describes the rupture as occurring on a deep patch at the bottom of the Whittier fault. Shao et al (2012) relied on the strong-motion records of 26 stations within a 50 km range of the epicenter. They carried out waveform inversions in the wavelet domain using a simulated annealing method to compute the slip, rake, rupture time, and slip-rate functions.…”

“…Subfault rake angles vary between 115°and 160°, with an average rake angle of 133.5°. According to Shao et al (2012), the size of the subfaults was chosen to represent seismic signals up to 2.5 Hz, but the slip functions carry some energy beyond this frequency. In our previous simulation and validation of the Chino Hills earthquake , we observed the source model seemed to lack sufficient energy between 3 and 4 Hz.…”

“…The FE mesh used in the simulation was tailored to reproduce the ground motion up to a maximum frequency, f max 4 Hz, and using a minimum shear-wave velocity, V S min 200 m=s. The source model corresponded to that of an independent inversion study (Shao et al, 2012). Similar physics-based, deterministic earthquake simulations have become commonplace in recent years thanks to advances in high-performance computing and numerical modeling techniques (e.g., Komatitsch et al, 2004;Aagaard et al, 2008;Graves, 2008;Olsen et al, 2009;Bielak et al, 2010).…”