Point-source modeling of moderate-tolarge magnitude earthquakes and associated ground-motion saturation effects

Abstract We modeled the source and attenuation attributes of well-recordedM 6 earthquakes based on the equivalent point-source approach, with the goal of determin-ing how to treat ground-motion saturation effects within this context. We consider ground motions as originating from an equivalent point source and mimic finite-fault effects by treating the motion as emanating from a virtual point (not a real point on the fault rupture), such that ground motions are correctly predicted at close distances. This is achieved by using an effective distance metric R D2rup h20:5, in which Drup is the closest distance to the rupture and h is a pseudodepth term that accounts for sat-uration effects. We found that the distance-saturation effect is magnitude dependent, extending to further distances with increasing magnitude. For earthquakes of M ≥6, we model the saturation term as logh −1:72 0:43 M with a standard deviation of 0.19 in log 10 units, based on the values obtained from the study earthquakes. The apparent source spectra of most M 6 earthquakes can be modeled using a simple Brune point-source model. For a few of the M 6 earthquakes, notably those in California, we observed a spectral sag at intermediate frequencies. For such earth-quakes, a two-corner point-source model provides a better match than the Brune model. We conclude that an equivalent point-source model based on the effective distance concept can successfully predict the average ground motions from M6 earthquakes over a wide distance range, including close distances (<20 km)