A method for generating synthetic ground motions. In this method, the
characteristics of seismic source, path attenuation, and local soil condition are taken into account,
when generating synthetic ground motions. Given a moment magnitude and an epicentral
distance, we use a stochastic model to generate an acceleration time history at the rock outcrop.
Then, we perform a nonlinear site response analysis to generate an acceleration time history at
the ground surface. Variability of ground motion resulting from uncertainties in modeling of
seismic source, path attenuation, and local site condition is not included in this report and is
addressed in the paper by Hwang (2000).
The method has been applied to generate synthetic ground motions resulting from large New
Madrid earthquakes. In this study, a deep soil profile overlaying the bedrock was established
based on a boring log in the Memphis area. In addition, 12 pairs of moment magnitudes and
epicentral distances were selected, and for each pair of moment magnitude and epicentral
distance, two samples of ground motion at the rock outcrop and at the ground surface were
simulated; thus, a total of 24 synthetic acceleration time histories were generated at the rock
outcrop and at the ground surface. For different combinations of moment magnitude and
epicentral distance, the synthetic ground motions have different amplitude and duration. It is
noted that the seismic source is modeled as a point source; thus, the ground motions simulated in
this study are appropriate for far- field condition. These ground motions may be used to perform
seismic response analysis of buildings and bridges located on the top of a deep soil profile in the
central United States.


See More: Generation of Seismic Synthetic Ground Motion