High-Frequency Site Effect of Hard Rocks at Ashio, Central Japan.

Abstract

Site effects of seismic S waves have been extensively studied for the last decade (Rogers et al., 1984;Phillips and Aki, 1986;Archuleta et al., 1992;Borcherdt and Glassmoyer, 1992). For sediment sites in particular, many efforts have been paid from the engineering point of view to clarify their amplification behaviors, and large site amplification at low frequencies (around a few Mertz and below) were revealed as compared with that at hard-rock sites (e.g., Aki, 1989;Boatwright et al., 1991;Takemura et al., 1991). On the other hand, there have been few studies on the hard-rock site effect (e.g., Tucker et al., 1984;Cranswick, 1988), and it is generally considered that hard-rock site effects are much smaller than those of sediment sites. However, quantitative estimation of high-frequency source characteristics such as the corner frequency and spectral fall-off is much affected by small distortion of seismograms due to the site effect as well as propagation path effect (e.g., Yoshimoto et al., 1993). This paper, presenting frequency-dependent site amplification at hard-rock sites, emphasizes that site effects are not negligible at high frequencies even for hard-rock sites.A three-component accelerograph array was deployed at Ashio, central Japan (see Figs. 1 and 2) for the period from August 1991 to April 1992. The array consisted of six stations: three temporal stations and three permanent stations of the National Research Institute for Earth Science and Disaster Prevention (NIED). The six stations constructed on hard rocks are plotted in Fig. 2 with surface geology. AS2. and AS2' were installed on the surface and at 15 m depth, respectively, at the same place. The overall frequency characteristics of the accelerograph response are nearly flat from 0.1 to 30 Hz. Seismic signals, digitally recorded with 12 or 16 bits, show high SN ratio up to 30 Hz. The sampling rate is 100 samples/s for AS2 and AS2', and 200 samples/s for all the other stations. In this study, we used seismograms of ten moderate local earthquakes which are plotted in Fig. 1 by shaded circles. These earthquakes were recorded at three or more stations including AS1. As are listed in Table 1, those earthquakes range from 4.1 to 6.1 in local magnitude, from 9.8 to 85.3 km in focal depth, respectively, and the hypocentral distances are from 67 to 180 km.We calculate the relative site amplification factors (RSAF) for direct S and P waves.

Q and Site Amplification Factors of Hard-Rock Region in the Kitakami Massif, Northeastern Japan.

“…A similar feature was found for site responses of granite hard-rock sites (Phillips and Aki, 1986;Yoshimoto et al, 1993). Yoshimoto et al (1993) showed that near-surface reverberation within a thin weathering layer causes large site amplification at high frequencies.…”

“…A similar feature was found for site responses of granite hard-rock sites (Phillips and Aki, 1986;Yoshimoto et al, 1993). Yoshimoto et al (1993) showed that near-surface reverberation within a thin weathering layer causes large site amplification at high frequencies. In the Kitakami massif, Paleozoic and early Mesozoic rocks have fractured and weathered through tectonic processes such as the intrusion of plutonic rocks or uplifting of the HTB.…”

“…Even if uncertainties in our results are taken into consideration, Q below the E-W profile seems closer to the northern part of the N-S profile than the southern part. Tucker et al (1984) and Yoshimoto et al (1993) investigated differences in amplification factors among hard-rock sites 5-8 km apart. They found differences within a factor of 2-3 for P-waves at frequencies lower than 15 Hz.…”

“…On the other hand, amplification factors of hard-rock sites have been generally assumed to be a half-space response in the study of source spectra and seismic hazard estimation. As data on hard rock site amplification increased recently in number and quality, the limit of this assumption became realized and studies on hard-rock site amplification behavior started by using deep boreholes or employing small-aperture arrays (e.g., Kato et al, 1992;Yoshimoto et al, 1993;Steidl et al, 1996). These studies confirmed that hard-rock sites have amplification factors of their own and that the factors are variable among sites only a few kilometers apart.…”

Site Effects on Large Ground Motions at KiK-net Iwase Station IBRH11 during the 2011 Tohoku Earthquake

“…Parameter a for the eighth layer in equations (1) and (6) is fixed to be zero, as in the other layers. We compare the theoretical H/V ratios for c 1, c 2, and c 3 in equation (6), because it has been shown that Q S is approximately one to two times as much as Q P in the frequency range higher than approximately 1 Hz for sedimentary soils (Kinoshita, 2008) and rocks (Yoshimoto et al, 1993). We found that parameter c is not sensitive to the theoretical H/V ratios.…”