SITE AMPLIFICATION IN TAIWAN: COMPARISON OF VARIOUS ESTIMATION METHODS AND EVIDENCE OF NONLINEARITY
ZHANG, F. and PAPAGEORGIOU, A.S.; Department of Civil, Structural and Environmental Engineering, State University of New York at Buffalo, Buffalo, NY 14260; firstname.lastname@example.org and email@example.com.
This study analyzes two data sets: (1) Strong motion data recorded by the seismic network of Taiwan Strong Motion Instrumentation Program (TSMIP) during the main event and 33 aftershocks of 1999 Chi-Chi, Taiwan, earthquake (Mw=7.6). This data set consists of 3161 motions recorded at 410 stations, (ML4.11 to ML 7.3); (2) short period data recorded by the network of Central Weather Bureau Seismic Network equipped with 3-component Teledyne-Geotech S-13 seismometers. This data set consists of 5499 records generated by 108 events (ML 2.90 to ML 4.97). The strong motion data were grouped according to peak ground acceleration (PGA). Site amplification is inferred using three techniques: (1) Generalized Inversion of S-waves; (2) H/V method (i.e., the ratio of the spectral amplitudes of the horizontal and vertical components of motion); (3) Coda wave inversion. Coda waves from both short period and strong motion data are used. As reference sites for the generalized inversion we selected stations that have been classified as belonging to site class B. The site amplification estimates obtained using the abovementioned three different techniques are reasonably close to each other for weak motions (<0.1g). The presence of nonlinearity, due to the intensity of ground motion, is clearly identified at three stations that experienced both strong (>0.2g) as well as weak (<0.1g) motions. Finally, we correlate site amplification with (i) geologic formation, and (ii) NEHRP classification, and we conclude that the latter provides the smaller scatter.