Wave Propagation and Source Models Compatible with Strong Motion Applications

John G. Anderson, Richard J. Brune, James N. Brune, & Glenn P. Biasi

In Preparation January 3, 2017, SCEC Contribution #6266

The goal of this project is to generate realistic, strong motion synthetic seismograms at distances of engineering significance, with emphasis on parameterizing the models with variables that are in standard use by seismologists. This approach is transferable to new regions, where strong motion data are sparse. Wave propagation is approximated with thin, flat, attenuating layers to approximate the local velocity gradient found by exploration seismology. The method produces seismograms with appropriately-polarized P-waves, S-waves, and surface waves. The attenuation model has high values of Q in the crustal waveguide (e.g. >5 km deep) to reproduce the distance dependence of the spectral amplitudes and of the spectral decay parameter (kappa), and low values of Q in shallow soil layers to reproduce the values of kappa extrapolated to zero distance (kappa-0).
The application we describe here simulates a M7.65 earthquake, approximating the southern end of where the 1857 rupture of the San Andreas fault passed Lovejoy Buttes with about 4 m of slip. Precarious rocks at Lovejoy Buttes mostly appear old enough that they survived the 1857 earthquake. Within the range of observed physical values for the input, we can generate synthetic seismograms that topple very few of these rocks, and others that topple most of them. Thus calibrating with precarious rocks, the composite source model has the potential to explore source physics parameters as well as to provide realistic, constrained ground motions from future great earthquakes.


Citation
Anderson, J. G., Brune, R. J., Brune, J. N., & Biasi, G. P. (2017, 01). Wave Propagation and Source Models Compatible with Strong Motion Applications. Oral Presentation at 16 World Conference on Earthquake Engineering.