Keywords: Bam earthquakes, empirical Green's functions, Kostrov slip function, synthesized strong motion, genetic algorithm, source parameters.

The main purpose of this article is to estimate the seismological source parameters of the December 26, 2003, Bam earthquake Mw6.5 (Iran). The selected station is far away from the causative fault so that the synthesized ground motion would not be influenced by near source problems such as directivity effects. The first successful attempt for theoretical calculation of strong motions was made by Aki [1] and Haskell [2]. One of the best approaches to overcome difficulties in synthesizing ground motion is to sum the recordings of small earthquakes, such as realistic aftershocks, delayed between each other so as to reproduce the rupture propagation effect. Each of the small-earthquake recordings represents all the propagation effects between the source and the receiver and is regarded as an empirical Green's function [3].

Hutchings [4] used empirical Green's functions to constrain the propagation path and site response information and proposed a range of simple kinematic rupture models to describe the source in predicting strong ground motion for the full time history. The method used in this study relies on only linear-soil-response. However, it is possible that the recorded ground motions have been influenced by the soil nonlinearity during the main earthquake. A FORTRAN computer code EMPSYN, originally written by Hutchings [4,5], is used to estimate synthetic seismograms by numerically computing the discredited representation relation, permitting the use of empirical Green's functions (EGF). Sources of uncertainty in engineering safety problems such as strong motion simulation methods are classified into two major groups known: aleatory uncertainty (due to inherent randomness of seismological model parameters) and epistemic uncertainty (due to the model formulation). It is desirable to reduce both types of uncertainties.

We used the genetic algorithm GA technique to reduce the uncertainty existing in the model parameters as much as possible by repeating, measuring and reducing the differences between the elastic response spectra corresponding to the synthesized time-series with those of the observed data.

The paper contains a description on the EGFs, the GA and the sources of uncertainties in synthesizing the earthquake using EGF method. The comparative good agreement of the elastic response spectrum with a 5% damping ratio corresponding to the simulated ground motions and those of the recorded data confirms the applicability of the proposed GA in minimizing the aleatoric uncertainty inherently existing in the model parameters.

1

K. Aki, "Seismic Displacements near a Fault", Geophys.Rer., 73, 5359-5376, 1968.

2

N.A. Haskell, "Elastic Displacements in the Near-Field of a Propagating Fault", Bull. Seism. Soc. Am., 59, 865-908, 1969.

3

S. Hartzell, "Earthquake Aftershocks as Green's Functions", Geophys. Res. Letters, 5, l-4, 1978. doi:10.1029/GL005i001p00001

4

L. Hutchings, "Prediction of strong ground motion for the 1989 Loma Prieta earthquake using empirical Green's functions", Bull. Seism. Soc. Am., 81, 88-121, 1991.

5

L. Hutchings, F. Wu, "Empirical Green's functions from small earthquakes A waveform study of locally recorded aftershocks of the San Fernando earthquake", J. Geophys. Res., 95, 1187-1214, 1990. doi:10.1029/JB095iB02p01187

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £145 +P&P)