Zisin (Journal of the Seismological Society of Japan. 2nd ser.)
Online ISSN : 1883-9029
Print ISSN : 0037-1114
ISSN-L : 0037-1114
Strong Ground Motion Simulation and Source Modeling of the Kagoshima-ken Hokuseibu Earthquakes of March 26 (MJMA6.5) and May 13(MJMA6.3), 1997, Using Empirical Green's Function Method
Hiroe MIYAKETomotaka IWATAKojiro IRIKURA
Author information
JOURNAL FREE ACCESS

1999 Volume 51 Issue 4 Pages 431-442

Details
Abstract

In 1997, two magnitude 6 class earthquakes occurred in the northwestern part of Kagoshima prefecture in Japan. One of them occurred on March 26 (MJMA6.5) and the other on May 13 (MJMA6.3). We examined the source models of these two events using the observed seismograms by K-NET at four stations surrounding the source areas. We proposed an objective estimation method to obtain the parameters N and C which are needed for the empirical Green's function method by IRIKURA (1986). This method, we call source spectral ratio fitting method, gives estimates of seismic moment ratio between a large and a small event and their corner frequencies by fitting the observed source spectral ratio with the ratio of source spectra which obeys the ω-2 model. This method has an advantage of removing site effects in evaluating the parameters. The best source model of the March mainshock was estimated by comparing the observed waveforms with synthetics using empirical Green's function method. The size of the asperity is about 7km length in the E-W direction by 6km width in the dip direction. The rupture started at the eastern-bottom of the asperity (8.2km) and extended radially to the western-upper direction. We noticed that the source model with additional small asperity to the eastern extension of major asperity improved the waveform fitting of the east-side station. In case of the May event, the aftershock distribution is ‘L’ shaped, and the mainshock waveforms on the rock site are composed of two clear pulses. The hypocenter of the mainshock is located close to the bottom (7.7km deep) of the intersection of the ‘L’ shape. Therefore, we considered two fault planes, the N-S plane and the E-W plane. The seismic moments of the two pulses were estimated nearly the same. We tested which fault plane ruptured first by comparing the initial part of the observed seismograms with synthetics. The matching was fairly good only when the N-S plane ruptured first and the E-W plane did later. The best source model of the May mainshock consists of two asperities. The first asperity has a size of about 3km length by 4km width in the N-S plane, where the rupture propagated radially from the northern-bottom. The other asperity with almost the same size but trending E-W ruptured radially from the western-bottom about two seconds after the first rupture.

Content from these authors
© The Seismological Society of Japan
Previous article Next article
feedback
Top