Calculation of broadband time histories of ground motion: Comparison of methods and validation using strong-ground motion from the 1994 Northridge earthquake
S. Hartzell et al., Calculation of broadband time histories of ground motion: Comparison of methods and validation using strong-ground motion from the 1994 Northridge earthquake, B SEIS S AM, 89(6), 1999, pp. 1484-1504
This article compares techniques for calculating broadband time histories o
f ground motion in the near field of a finite fault by comparing synthetics
with the strong-motion data set for the 1994 Northridge earthquake. Based
on this comparison, a preferred methodology is presented. Ground-motion-sim
ulation techniques are divided into two general methods: kinematic- and com
posite-fault models. Green's functions of three types are evaluated: stocha
stic, empirical, and theoretical. A hybrid scheme is found to give the best
fit to the Northridge data. Low frequencies (< 1 Hz) are calculated using
a kinematic-fault model and a 3D finite-difference code to propagate energy
through a realistic 3D velocity structure. High frequencies (> 1 Hz) are c
alculated using a composite-fault model with a fractal subevent size distri
bution and stochastic, bandlimited, white-noise Green's functions, At frequ
encies below 1 Hz, theoretical elastic-wave-propagation synthetics introduc
e proper seismic-phase arrivals of body waves and surface waves. The 3D vel
ocity structure more accurately reproduces record durations for the deep se
dimentary basin structures found in the Los Angeles region. At frequencies
above I Hz, scattering effects become important and wave propagation is mor
e accurately represented by stochastic Green's functions. A fractal subeven
t size distribution for the composite fault model ensures an omega(-2) spec
tral shape over the entire frequency band considered (0.1-20 Hz).