← Back to Sessions

Using Proxies Obtained From Horizontal-to-Vertical Spectral Ratio to Reduce the Epistemic Uncertainty in Ground Motion Models

Description: 

The effects of site characteristics in ground motion models (GMMs) are usually predicted by shear-wave velocity of top 30 m (V_S30) or depth to 1.0 km/s shear-wave isosurface (Z_1.0). However, these parameters are limited to observations of the top 30m of the soil stratigraphy or inferred from models, respectively. Previous research showed that the horizontal-to-vertical spectral ratio (HVSR) technique can be used to estimate site fundamental frequency and its corresponding amplitude, which can, in turn, reduce uncertainties in GMMs. This study uses a large data set from the KiK-net strong motion network and presents a maximum-likelihood estimate of site fundamental frequency (f_ml) and its corresponding amplitude (A_ml) for 699 sites across Japan. First, ground motion recordings are selected and processed. The Fourier Amplitude Spectra and Response Spectra of the recordings are computed in this step. Second, the HVSR is calculated for the recordings. Then, fml and Aml are estimated for each site using previously developed methodology that is based on maximum-likelihood approach. The process of estimation is automated to diminish bias and quantify the uncertainty in the estimations. As a result, the estimated parameters (site fundamental frequency and the corresponding amplitude in the HVSR curve) are compared to an independent study in order to validate the automated approach and the maps for these parameters are presented. These additional parameters reduce the uncertainty (i.e., site-to-site variability) in estimates of response spectra by 2% to 19%, depending on the spectral period and the GMM used as a reference.

Session: From Earthquake Recordings to Empirical Ground-Motion Modelling - III

Type: Oral

Date: 5/2/2024

Presentation Time: 02:30 PM (local time)

Presenting Author: Mohammad

Student Presenter: No

Invited Presentation: 

Authors