CKI-03: Estimation of Local Site Effects on Strong Ground Motion
• Prof. Keiiti Aki
• Prof. Ta-Liang Teng
A total of 175 earthquakes recorded from 1984 to March, 1990 by the short-period seismic system of the USGS Menlo Park seismographic network were collected for this study. The magnitude of these earthquakes ranges form 1.8 to 3.5.
Part I. Relation between the weak-motion amplification factor and the site geological age for the frequency range from 1.5 Hz to 12 Hz.
Part II. Applicability of the weak-motion amplification factor to strong ground motion during the Loma Prieta earthquake of 1989.
Evaluation of Local Site Effects from Strong Motion Spectra
• Dr. Masanori Niwa
• Dr. Masayuki Takemura
• Mr. Kenichi Kato
• Mr. Tomonori Ikeura
• Mr. Kenji Urao
• Dr. Masamitu Miyamura
• Prof. Etsuzo Shima
• Dr. Tokiharu Ohta
The final goal of this study was to obtain a reliable and practical method for evaluating the local site effect for the strong ground motion. Three different methods have been applied to observed seismic data in Japan and in California, which are the linear inversion method to S-waves (S-wave method), the regression analysis to obtain the attenuation curve of strong motion spectra (Regression method) and the linear inversion method to coda-waves (Coda-method).
The coda method has been applied to the data in California by the USC team. Our study has been carried out in the four sections as follows:
(1) Revision of the constraint condition based on the numerical approach to the effect of topographic irregularity around the KDG site. The reliable constraint condition is obtained for the Japanese data analyzed by the S-wave method
(2) Application of S-wave method to the Japanese dataset for evaluating the relation between local site effect and surface geology. The site amplifications of S-waves at 19 stations are obtained and summarized with respect to the surface geology.
(3) Application of S-wave method to the California dataset for comparing with the result by the coda method for the same station. The consistency between the site amplifications of S-waves and of coda-waves is found at most stations.
(4) Application of the regression method to the Japanese dataset for comparing with the result by the S-wave method. Qs-value and site amplification factors obtained by the regression method are almost consistent with those by the S-wave method in both the cases that the S-wave portions are used in the regression analysis, and that the response spectrum is used as the conventional regression analysis.
Calibrations among the different methods has been performed in the present study. The evaluation of the non-lenear effect of the ground for young sediment site during large earthquake is a problem remained in the future.
2. Effect of surface topography in an inversion analysis for evaluation of local site effect from strong motion spectra
3. Site amplification of S-waves from strong motion records in special relation to surface geology
4. Site amplifications of S-waves and coda waves in central California
5. Evaluation of site amplifications by the regression analysis of strong motion spectra
Also listed as Report No.: CK91-01 (February 1991)