[S07-6] Site effects I

[S07-6-02] HVSR site classification method for Chinese seismic code based on Japanese strong motion data

^○Ruizhi Wen^{1, 2}, Yefei Ren^{1, 2}, Kun Ji^1,2, Haiying Yu^1,2 (1.Institute of Engineering Mechanics, China Earthquake Administration,Harbin,China, 2.Key Laboratory of Earthquake Engineering and Engineering Vibration,Harbin,China)

The local site condition is a very important factor for strong motion data. We performed the site classification of the strong motion stations for China National Strong Motion Observation Network System (NSMONS), by using an empirical Horizontal-vertical spectral ratio (hereafter, HVSR) method. Due to not enough Chinese strong motion data and borehole, we firstly selected all needed information from Japanese KiK-net and statistically assigned the mean HVSR curve of the site classification (CL-I, CL-II, or CL-III) defined in the Chinese seismic code. The mean the HVSR curve for each site class was computed using Chinese strong motion recordings captured during the period 1996–2012 . For each strong motion stations, the station HVSR curves were compared with those proposed by Zhao et al. (2006a) for four types of site classes (SC-I, SC-II, SC-III, and SC-IV) defined in the Japanese seismic code (JRA, 1980). It was found that an approximate range of the natural period could be identified by the predominant peak of the HVSR curve for the CL-I and SC-I sites, CL-II and SC-II sites, and CL-III and SC-III + SC-IV sites. Meanwhile, an empirical method of Chinese site classification was proposed based on comprehensive consideration of peak period, amplitude, and shape of the HVSR curves. 178 NSMONS stations were computed based on recordings from 2007 to 2015 and the sites classified using the proposed method. The mean HVSR curves were re-calculated for site classes and compared with those from KiK-net data. It was found that both the peak period and the amplitude were similar for the HVSR curves derived from NSMONS and KiK-net data, implying the effectiveness of the proposed method in identifying different site classes. It is also found that this classification has a good agreement with site classes based on 81 borehole data in China, which indicates that this site classification results are acceptable.