Hybrid Broadband Ground-Motion Simulation Using Scattering Green's Functions: Application to Large-Magnitude Events

Abstract

We calculate near-source broadband (0-10 Hz) seismograms by combining low-frequency three-dimensional (3D) finite-difference seismograms (0-0.5 Hz) computed in a 3D velocity model using site-specific scattering Green's functions for random, isotropic scattering media. The scattering Green's functions are convolved with a slip-rate function to form local scattering operators (scatterograms), which constitute the high-frequency scattered wave field. The low-frequency and highfrequency scatterograms are then combined in the frequency domain to generate broadband waveforms. Our broadband method extends the Mai et al. (2010) approach by incorporating dynamically consistent source-time functions and accounting for finite-fault effects in the computation of the high-frequency waveforms. We used the proposed method to generate broadband ground motions at 44 sites located 5-100 km from the fault, for M w 7.7 earthquake scenarios (TeraShake) on the southern San Andreas fault, which include north-to-south, south-to-north, and bilateral rupture propagation from kinematic and spontaneous dynamic rupture models. The broadband ground motions computed with the new method are validated by comparing peak ground acceleration, peak ground velocity, and spectral acceleration with recently proposed ground-motion prediction equations (GMPEs). Our simulated ground motions are consistent with the median ground motions predicted by the GMPEs. In addition, we examine overturning probabilities for 18 precariously balanced rock sites (PBR). Our broadband synthetics for the M w 7.7 TeraShake scenarios show no preferred rupture direction on the southern San Andreas fault but are inconsistent with the existence of PBRs at several of the sites analyzed.

Improving seismotectonics and seismic hazard assessment along the San Ramón Fault at the eastern border of Santiago city, Chile

“…As initially stated, we focused on some specific source effects, then avoiding to explore additional effects on ground motions when including heterogeneities on other rupture source parameters. Certainly, it may be tested and properly included in further applications, and some work has been done under this perspective (e.g., Pulido et al 2004;Dalguer 2009 Graves andPitarka 2010;Mena et al 2010). Under the same perspective, in a recent work based on the same fractal kinematic k -2 source model, where the constant stress-drop hypothesis at all scales was released, Ruiz et al (2013) shown that the high frequency is greatly enhanced by combining both: (1) a larger average stress-drop value at small scales compared to the average values of larger subevents and (2) empirical Green's functions.…”

“…Indeed, further improvements of broadband strong ground motions should include local site/soil effects and basin/topographic effects. The latter can be incorporated through the propagation of low-frequency three-dimensional seismic waves, as done for instance in the Santiago basin (Pilz et al 2011), or similarly in other areas, like Los Angeles basin in southern California, highly exposed to potential seismic hazard too (e.g., Olsen 2000; Mena et al 2010;Graves and Pitarka 2010).…”

A Study on the Variability of Kappa ( ) in a Borehole: Implications of the Computation Process

“…Thus, it is also studied in the context of site characterization and site effect studies (e.g., Drouet et al, 2010). Recently it has been correlated also to other engineering parameters such as the Arias intensity and PGA (Mena et al, 2010).…”

Hybrid Broadband Ground-Motion Simulations: Combining Long-Period Deterministic Synthetics with High-Frequency Multiple S-to-S Backscattering

“…For large extended-fault ruptures and their associated near-field radiation due to localized slip patches, this approximation may not hold. The scattering contributions from such extended-source models and their effects on the broadband wave field are the subject of Mena et al (2010).…”

“…For the purpose of our methodological study, we fix the scattering parameters to η i 0:028 and η s 0:01; however, we note that 20% variation in these values does not change the peak motions noticeably. A detailed parameter study on the effects of these scattering parameters on several strong-motion parameters is reported in the companion paper by Mena et al (2010).…”