← Back to Sessions

Evaluation of the High-Frequency Attenuation Parameter in Italy for Site Effect Characterization

Session: How Well Can We Assess Site Effects So Far? I

Type: Oral

Date: 4/20/2021

Presentation Time: 10:00 AM Pacific

Description: 

The parameter κ₀ describes the deviation at high frequencies between the amplitude of the acceleration Fourier spectrum (FAS) computed from a seismic record and a source model proportional to ω^-2 (Ktenidou et al., 2014). κ₀ is broadly used in the classical framework of ground-motion models (GMMs), both as an input parameter for the host-to-target correction of GMMs (Cotton et al., 2006; Douglas et al., 2006) and as explanatory variables in the functional forms of predictive models.

In this work, we develop a semi-automatic procedure to compute the values of κ₀ for several recording sites, mainly located in Central and Southern Italy. The estimated values of κ₀ for the reference rock sites in Central Italy (Lanzano et al., 2020) are compared with high-frequency spectral amplitudes of the systematic site error (δS2S_ref) obtained from the non-ergodic model of the horizontal component for the FAS. The statistical coefficient between δS2S_ref values in FAS and κ₀ shows a significant level of anti-correlation in the 5-25 Hz frequency range (i.e., as κ₀ increases, we observe an overall reduction in δS2S_ref). Moreover, we study the relation between κ₀ and other site-effect proxies, such as the average shear-wave velocity V_S,30, usually adopted to represent the site response in GMMs and seismic codes. κ₀ and V_S,30 are only weakly correlated, confirming that they can be considered as independent parameters for site effects characterization. We also observe a dependency of κ₀ on the lithological characteristics of the recording sites: lower values of the attenuation parameters correspond to rigid formations composed of limestone, while higher values are mainly located on flyschoid units. As a final step, a preliminary map of κ₀ in Central Italy has been built by means of spatial correlation analysis.

Presenting Author: Giovanni Lanzano

Student Presenter: No

Authors