Estimation of S-Wave Velocity Profiles and Site Amplification Around the K-NET Tsukidate Station, Miyagi Prefecture, with Reference to Large PGA During the 2011 off Pacific Coast of Tohoku Earthquake, Japan

Abstract

In this study we investigated site amplifications in shallow soil around Tsukidate station (MYG004) of KNET in Miyagi prefecture with a seismic intensity of 7 during the 2011 off the Pacific Coast of Tohoku Earthquake from aftershock observation and microtremor explorations. We conducted aftershock observations by installing 8 temporary stations. We estimated site amplifications of S-wave from the observed data. We found that the predominant period of amplification at a site near MYG004 was 0.2 seconds. The amplification factor at a site on cut ground with a distance of 20 m from MYG004 was significantly different. Furthermore, the amplifications were also different with changing predominant periods from 0.1 to 0.5 seconds in an area of 1 km from MYG004. Shallow Swave velocity profiles were deduced from microtremor array explorations at the stations. Results indicated that shallow low-velocity layers with a thickness of less than 30 meters are responsible for variation of the amplifications. Predominant periods in calculated 1D amplifications were similar to those observed in the aftershock observations. The observed amplification factor at the site nearMYG004, however, was two times larger than the calculated one, suggesting the inappropriateness of the 1D assumption at the site possibly due to the effects of a cliff near MYG004. We also found from horizontal-vertical ratios of microtremors that the site effects on the cliff were also different at frequency higher than 10 Hz within 30 meters around MYG004. This also suggests lateral variation in the site effects at the high frequency near MYG004.

Evaluation of a novel application of earthquake HVSR in site-specific amplification estimation

“…2c clearly shows that SBSR01~04 significantly underestimate SBSR05 (ETF), suggesting that neglecting deep velocity variations causes a considerable underestimation of site responses at both low and, more interestingly, high frequencies although downward waves may also be somewhat responsible. Our empirical evidence is consistent with the numerical results by Yamanaka et al [22]. Therefore, only when there is no abrupt impedance contrast (IC) below the downhole sensor at a site, can SBSR be used as its ETF.…”

“…It is expected that deep velocity structures affect the amplification spectra in the low-frequency range. However, using a numerical approach, Yamanaka et al [22] showed that deep velocity configurations could influence the amplifications at high frequencies as well. When the underlying layer having higher 𝑉 is removed, the levels of amplification at high frequencies are lower than those of the complete model.…”