Determination of Site amplification in the Campania-Lucania Region (southern Italy) by comparison of different Site-Response Estimation Techniques

Abstract

We have applied a non-linear inversion technique to estimate both the earthquake source parameters and site response from a set of microearthquake records acquired by a dense seismic network, in order to provide a simple uniform site classification for the strong-motion stations installed in Campania-Lucania region (Southern Italy) by ISNet network. The main assumption of this inversion method, is the linear response of the sites with respect to the input earthquakes. The technique is applied to horizontal acceleration and velocity records of 101 seismic events occurred in the Campania-Lucania region. We assumed the -2 source model, a R-1 geometric attenuation, and a frequency dependent anelastic attenuation factor. All the records are corrected by propagation path such that only the source and site effects remain in the records. The site effect at each station is thus obtained from the spectral ratio between the corrected amplitude displacement spectrum and the theoretical spectrum retrieved from the inversion. The estimated transfer functions show that stations located on stiff and soft soils attenuate the ground motion in the high frequency range, while the site effects for rock site present significant amplification at higher frequencies than 2.0 Hz. Next, at each station, the retrieved site-amplification spectra are averaged within four frequency bands: (Ib) 1-5 Hz, (IIb) 5-10 Hz, 10-15 Hz and 15-20 Hz. These bands are chosen based on the size of the analyzed earthquakes and to retrieve the amplification coefficient for different input ground motion parameters, such as PGA, PGV or Sa(T) that are moderate-to-high frequency parameters. Finally, we constructed four maps, one for each frequency band in order to correlate the average site amplifications with superficial geology. The accuracy of the site response maps will evolve as more data became available in order to highlight significant variations in site response for different geologic unit.