Determination of shallowS-wave attenuation by down-hole waveform deconvolution: a case study in Istanbul (Turkey)

Abstract

S U M M A R YThe estimation of shear wave velocity and attenuation in near-surface geology is of primary importance in engineering seismology. In fact, their knowledge is essential for site response studies when preparing improved seismic hazard scenarios. In this study, we propose two approaches for estimating the average shear wave quality factor Qs by using recordings of a vertical array of accelerometers. The methods are mainly based on the deconvolution of the wavefield recorded in a borehole with that recorded at the surface.The first method requires the Fourier transform of the deconvolved wavefield to be fitted with a theoretical transfer function valid for the vertical or nearly vertical (in the case at hand up to 30 • incidence angle) propagation of S waves. The second method is based on the spectral fitting of the Fourier transform of only the acausal part of the deconvolved wavefield with a theoretical transfer function.Both methods can be applied without any prior knowledge of the subsoil structure (since they are based on empirical data analysis) and do not require a precise knowledge of the azimuthal orientation of the sensors in the boreholes (which is seldom available). First, we describe the theoretical framework of the proposed methodologies for Qs estimation, which are based on the assumption that the structure in the borehole is weakly heterogeneous in the vertical direction (i.e. no large impedance contrast exists between the borehole sensor and the surface). Second, by using synthetic accelerograms, we verify that in a realistic subsoil structure, the assumption of vertical homogeneity can hold and we investigate the robustness and the suitability of the proposed methods. Finally, only the method that was shown to provide the more stable results, based on fitting the borehole-to-surface spectral ratio with a theoretical function, is applied to earthquakes signals recorded by a vertical array of accelerometers installed in Ataköy (western Istanbul). Results show that using borehole data provides a fair and robust estimate of an average Qs (of about 30, 46 and 99 for the 0-50, 0-70, 0-140 m depth ranges, respectively) that can be used for numerical simulations of ground motion.

Seismic‐Wave Propagation in Shallow Layers at the GONAF‐Tuzla Site, Istanbul, Turkey

“…Despite this bias, which causes large errors especially in Q P , our apparent quality factors are similar to the ones found in other studies in the Marmara Sea region. For example, Parolai et al (2010) estimated Q S values of 30, 46, and 99 for the depth ranges 0-50, 0-70, and 0-140 m in Ataköy (western Istanbul). They also used the recordings of a vertical borehole array and fitted the borehole-to-surface spectral ratios with a theoretical transfer function to derive Q S .…”

“…The deconvolution was also checked in the reverse sense, using the deepest sensor as reference. Instead of performing an inversion as proposed by Trampert et al (1993) or analyzing the deconvolved wavefield in the frequency domain as proposed by Parolai et al (2010Parolai et al ( , 2012, we forward modeled the deconvolved seismograms in the time domain. With a grid search we found the seismic velocities for the P and S waves, V P and V S , and quality factors Q P and Q S that best fit the data in a least-squares sense.…”

“…Deconvolution interferometry has also found applications in estimating site-specific Q values. Parolai et al (2010) determined site-related Q values by fitting the amplitude spectrum of the deconvolved wavefield using analytical models, and Parolai et al (2012) derived Q by performing a full inversion of the spectrum.…”

k0: The role of Intrinsic and Scattering Attenuation

“…Second, the role of the intrinsic attenuation alone is studied by setting Q S equal to 10 in the uppermost 100 m. A frequency-independent quality factor is adopted, consistent with standard engineering practice (Parolai et al, 2010). The frequency dependence of the quality factor is still an open issue (Morozov, 2008(Morozov, , 2010Cantore et al, 2012); and, in any case, its introduction would not modify the main outcomes of this study regarding the respective roles of intrinsic and scattering attenuation.…”

Joint Deconvolution of Building and Downhole Strong‐Motion Recordings: Evidence for the Seismic Wavefield Being Radiated Back into the Shallow Geological Layers

“…When coupled with standard engineering approaches such as modal analysis using frequency domain decomposition (Brincker et al, 2001) or Fourier spectral analysis of an earthquake, active and/or passive source measurements, this approach allows the separation of the building's dynamic behavior from that arising from the soil-structure interaction. Deconvolution interferometry has also been extensively applied to borehole strong-motion data (e.g., Mehta et al, 2007a,b;Parolai et al, 2009Parolai et al, , 2010Oth et al, 2011) to gain information about wave propagation in the shallow geological layers. However, a full picture of the wavefield propagation from the subsurface through the structure and back to it, and an identification and quantification of the wavefield radiated back from a structure to the soil, can only be obtained if simultaneous recordings from boreholes and instrumented buildings located nearby are available and jointly analyzed.…”