Seismograms from 55 intermediate-depth Vrancea earthquakes (M 4.0-7.1) recorded at 43 stations of an accelerometric network in Romania are used to derive source spectra and site amplification functions from S waves in the frequency range 0.5-20 Hz with the generalized inversion technique (GIT) (Castro et al., 1990). Attenuation is taken into account using the nonparametric attenuation functions derived by Oth et al. (2008) from the same dataset, and the attenuation-corrected data are then split into source and site contributions. The source spectra follow the omega(-2) model (Brune, 1970, 1971) with high corner frequencies and a related Brune stress drop of the order of 100 MPa. The site amplification functions are separately determined for both horizontal and vertical components. Contrary to widespread expectation, the vertical component shows significant amplification effects at high frequencies. The H/Z ratios determined from the GIT results compare well with H/V ratios computed directly from the S-wave window of the accelerograms (Lermo and Chavez-Garcia, 1993). The basic assumption for the determination of site effects from H/V ratios is that the vertical component is not or only little affected by site effects. For Vrancea earthquakes, this assumption is incorrect; consequently, site effects should not be estimated from H/V ratios. The reason for this peculiar fact is the geometry of intermediate-depth seismicity that leads to almost vertical ray paths beneath the stations.
Source Spectra and Site Response from S Waves of Intermediate-Depth Vrancea, Romania, Earthquakes
Parolai S.;
2009-01-01
Abstract
Seismograms from 55 intermediate-depth Vrancea earthquakes (M 4.0-7.1) recorded at 43 stations of an accelerometric network in Romania are used to derive source spectra and site amplification functions from S waves in the frequency range 0.5-20 Hz with the generalized inversion technique (GIT) (Castro et al., 1990). Attenuation is taken into account using the nonparametric attenuation functions derived by Oth et al. (2008) from the same dataset, and the attenuation-corrected data are then split into source and site contributions. The source spectra follow the omega(-2) model (Brune, 1970, 1971) with high corner frequencies and a related Brune stress drop of the order of 100 MPa. The site amplification functions are separately determined for both horizontal and vertical components. Contrary to widespread expectation, the vertical component shows significant amplification effects at high frequencies. The H/Z ratios determined from the GIT results compare well with H/V ratios computed directly from the S-wave window of the accelerograms (Lermo and Chavez-Garcia, 1993). The basic assumption for the determination of site effects from H/V ratios is that the vertical component is not or only little affected by site effects. For Vrancea earthquakes, this assumption is incorrect; consequently, site effects should not be estimated from H/V ratios. The reason for this peculiar fact is the geometry of intermediate-depth seismicity that leads to almost vertical ray paths beneath the stations.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.