none

• Main Authors: Chi-Hsuan Chung, 鍾奇軒
• Other Authors: 溫國樑
• Format: Others
• Language: zh-TW
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/984uuu

碩士 === 國立中央大學 === 地球科學學系 === 105 === 　　Lots of infrastructures are under construction in metropolises in Taiwan in recent years and thus lead to increasement of population density and urbanization in those area. Taiwan island is located in plate boundaries where the high seismicity is caused by active tectonic plates. The Chi-Chi earthquake (Mw 7.6) in 1999 caused a fatality of more than 2000, and the Meinong earthquake (Mw 6.5) in 2016 caused a fatality of 117 in Tainan city as well as damages on hundreds of buildings. The cases imply seismic vulnerability of urban area. The severity of the earthquake hazard is often associated with the site effect, so the study of different site conditions of seismic wave amplification characteristics, in improving the seismic hazard analysis and seismic design is a very important issue. 　　This study used the strong motion records observed by the TSMIP network. The site conditions based on Vs30 used in this study were investigated at most stations (Kuo et al. 2012; Kuo et al. 2017). Since strong motion records and site conditions are both available, we are able to use the data to analyze site amplifications of seismic waves at different periods. The result may be a reference for future modification of seismic design codes to decrease potential seismic hazards and losses. 　　We adopted the strong motion and site database of the SSHAC Level 3 project in Taiwan. The significant crustal and subduction events of magnitude larger than six are selected for analysis. The amplification factors of PGA, PGV, PGD, and spectra acceleration at 0.3, 1.0, and 3.0 seconds were evaluated. According to the recommendation of SSHAC Level 3 project, the site condition of Vs30 = 760 m/s is considered as the reference rock site. The stations with Vs30 between 600 m/s and 900 m/s were used as the reference rock sites in this study. For each event, we find a reference rock site and a target site within a certain distance (regional dependent) to calculate site amplifications of ground motions. Relationships of site amplification factors and Vs30 are therefore derived by regression analysis. Soil nonlinearity (decrease of amplifications) has to be considered at soft soil sites during a strong shaking. We also discuss amplification factors in terms of different intensities if data is available. 　　In the current code, the relationship between amplification factor and Vs30 is a ladder-like form. Similar Vs30 values near the site classification boundaries may have quite different amplification factors. This study intends to improve this problem by providing a smooth curve to represent the relationship between amplification factor and Vs30.