Improving the spatial resolution of ground motion variability using earthquake and seismic noise data: the example of Bishkek (Kyrgyzstan)

Abstract

Site response analysis plays an important role in seismic hazard and risk assessment, and in defining the optimal engineering design for civil structures. However, due to increasing urbanization, target areas are often too vast to be covered by standard approaches, resulting in large uncertainties in the spatial variability of the expected ground motion. Here, we propose a method to improve the spatial resolution of ground motion variability in terms of Standard Spectral Ratios (SSRs), using earthquakes recorded at a few selected sites for a relatively short amount of time, and seismic noise data collected over a denser grid, taking advantage of clustering and correlation analysis. The method is applied to Bishkek, Kyrgyzstan. Using the K-means clustering algorithm, three clusters of site response types have been identified, based on their similarity of SSRs. The cluster's site responses were adopted for sites where only single station noise measurements were carried out, based on the results of correlation analysis. The spatial variability of the site response correlates well with the main geological features in the area. In particular, variability is noted from south to north, consistent with both the changes in the thickness of the sedimentary cover over the basin and in the Quaternary material outcropping at the surface. This method has therefore the potential to improve the estimation of site effects at the local scale in the future.

A new approach to site classification: Mixed-effects Ground Motion Prediction Equation with spectral clustering of site amplification functions

“…(Left panel) Median response spectra. (Right panel) between-event (τ), between-site (φS2S), eventand-site corrected standard deviations (φ0) and the total aleatory variability (σ) of the GMPE in natural-log scale5 Spectral clustering analysisWith increasing amount of strong motion data, ground motion modelers are adapting advanced statistical tools to analyze large amounts of ground motion data to identify repeatable effects, evaluate parametrization, and quantify better the uncertainties in prediction (e.g Derras et al (2012),Ullah et al (2013),Luzi et al (2011)…”

An overview on the seismic microzonation and site effect studies in Central Asia

“…Using earthquakes recorded at a few selected sites for a relatively short amount of time and seismic noise data collected over a denser grid, Ullah et al [2013] showed that the spatial resolution of ground motion variability in terms of SSRs can be significantly improved. Using a K-means algorithm, clusters of sites that display a similar response can be identified based on their similarity of the SSRs.…”

Toward a Loss-Driven Earthquake Early Warning and Rapid Response System for Kyrgyzstan (Central Asia)

“…In the case of Bishkek, local amplification effects have been estimated at 19 different locations within the town, where seismic stations have been installed (Parolai et al, 2010;Ullah et al, 2012;see Fig. 3).…”

“…The site response is obtained by means of standard spectral ratio (Borcherdt, 1970) between the Fourier spectra of the recordings at the stations located in Bishkek and those of a reference station (Parolai et al, 2010). The site amplifications estimated in this way are affecting a wide frequency band (from nearly 0.2 Hz to a few hertz) and generally increase from south to north (Parolai et al, 2010;Ullah et al, 2012).…”