HR: 1330h
AN: S52A-0110    [PDF]
TI: Three-Dimensional Finite-Difference Simulations of Strong Ground Motions during the 1720 Shacheng Earthquake($M_{w}7.0$) of Yanhuai Area, Beijing, China using a Stochastic Finite-Fault Model
AU: * Wang, G
EM: wang@geophysik.uni-muenchen.de
AF: Institute of Geophysics, University of Munich, Theresien str. 41, Munich, 80333 Germany
AU: Igel, H
EM: igel@geophysik.uni-muenchen.de
AF: Institute of Geophysics, University of Munich, Theresien str. 41, Munich, 80333 Germany
AU: Wang, H
EM: haijiang@geophysik.uni-muenchen.de
AF: Institute of Geophysics, University of Munich, Theresien str. 41, Munich, 80333 Germany
AB: Three-dimensional finite-difference (3D-FD) simulations of elastic waves in the Yanhuai Basin are performed for the 1720 Shacheng earthquake ($M_{w}$7.0) with a stochastic finite-fault model. The goal of this study is to incorporate the stochastic finite-fault model used widely in earthquake engineering for simulation of strong ground motion (Beresnev and Atkinson, 1997, 1998, etc.) into the 3D-FD program. The basic idea of the stochastic finite-fault model is that the causative fault plane can be subdivided into several subfaults (elements), and radiation from a large earthquake is the sum of contributions from all subfaults with proper time delays, each of which acts as a small independent source. The rupture starts at a given point on the fault and propagates with constant velocity, triggering subsources as soon as it reaches them.The causative fault of the 1720 Shacheng earthquake ($42km\times 18km$) is divided into $6\times 4$ subfaults with dimensions of $7km\times 4.5km$ in this study. An important aspect of the problem of characterizing the earthquake source is the degree of fault heterogeneity or roughness. Heterogeneity of the fault rupture process are modeled by randomizing the location of the initial rupture, and by randomizing the subevent rise times and slip distributions in this study, from which the model was named. This study emphasizes particularly on the applications of the theoretical simulation in the engineering practice. The simulated results include the distribution maps of peak ground velocities (PGV), duration of strong ground motions, 5%-damped response spectra (pseduo-velocity response spectra) for different oscillator periods (e.g., 0.7sec, 1.0sec, 1.5sec, 2.0sec, 5.0sec), and seismic intensity. We study the effects of the randomness implicated in our source model on the simulated results with a homgeneous 3D model systematically. The general finding is that the randomness produce slight effects on the distribution of PGV (less than 5 cm/s), response spectra, durations, as well as seismic intensity, and furthermore, the effects are mostly limited in the range of the horizontal projection of the fault surface on ground surface. A 3D velocity model of the Beijing Area is constructed from studies that analyzed available geological information, seismic-reflection suveys, borehole logs, and gavity data. The smallest shear-wave velocity in the model is 1.0 km/sec. A grid increment of 100m in three directions is used in this simulation, which made it possible to caputure the short period information with a resolution of 0.5 sec. It is the most interesting period range in earthquake engineering. This study highlights the possibility of using the finite-difference and stochastic finite-fault combined method into the simulation of strong earthquake ground motions,and domentrates the important effects of the basins occured in the Beijing area on the strong ground motions.
DE: 0604 Antenna arrays
DE: 0689 Wave propagation (4275)
DE: 0902 Computational methods, seismic
DE: 0903 Computational methods, potential fields
SC: Seismology [S]
MN: 2003 Fall Meeting