Abstract:
In this study we attempt to estimate parameters that are related to path and near-surface effects, due to earthquakes that occurred in northwestern Turkey, by drawing on data from the Istanbul Earthquake Rapid Response System (IERRS) that is operated by the Department of Earthquake Engineering at Boğaziçi University’s Kandilli Observatory and Earthquake Research Institute. Our data set includes over 4677 strong motion recordings from 115 stations of the IERRS. The records are associated with events having local magnitudes ranging within 4.0<ML<6.5 and epicentral distances between 6-444 km. Only records, that have a signal-to-noise ratio larger than 3, are kept for a more accurate selection of κ and estimation of Q. Horizontal-to-vertical spectral ratio profiles are observed, and fundamental frequencies are checked in order to prevent possible biases in the attenuation models. κ is measured from the acceleration Fourier amplitude spectra of shear-waves, as originally proposed by Anderson and Hough (1984). Several regional models are prepared as a function of site (κ0) and distance (m×R) by linearly regressing over distance. Correlations of azimuths, SNR levels and VS30 values with these models are discussed. A regional model for frequency dependent anelastic attenuation Q(f) is proposed by a regression analysis of acceleration Fourier amplitudes of shear-waves following Castro and Ruiz-Cruz (2005) and Safarshahi et al. (2013).