- © 2010 by the Seismological Society of America
Broadband synthetics obtained from scenario simulations of earthquakes with a frequency content between 0 and 10 Hz, referred to hereafter as “BBSs,” are playing an increasingly important role in seismic hazard analysis. An example is the Great Southern California ShakeOut, the largest disaster response exercise in U.S. history and an annual event since 2008 (Jones et al. 2008). The drill was the first to be based on BBSs, in this case for an M 7.8 scenario earthquake on the southern San Andreas fault. Another example of the important role of synthetic ground motions is the increasing awareness of the advantages of using site-specific ground-motion time series, rather than empirical intensity measures or scaled time series from different sources or locations, for more realistic non-linear dynamic analysis of buildings and performance-based earthquake engineering. BBSs appear to be one of the only viable alternatives to the very limited amount of strong-motion time series, particularly in the near-field from large earthquakes.
Effectively meeting demands of this sort for realistic BBSs requires careful validation against recorded data. BBSs are currently achieved by combining deterministic low-frequency (LF) synthetics up to a maximum frequency (fmax) of typically 1–2 Hz with high-frequency (HF) stochastic synthetics above this upper cutoff frequency (see, for example, Graves and Pitarka 2004; Liu et al. 2006; Mai et al. forthcoming). Visual inspection has been used for decades to claim success or failure of the ability of simulations to match observations (or synthetics derived from an alternative numerical method). However, at shorter periods such visual waveform fits are not practical, likely due to chaotic source and path variability. For example, specific intensity measures tend to be more practical and relevant than actual waveform fits at higher frequencies.
Candidates for metrics to measure the misfit for BBSs include commonly used ground-motion intensity …