Tokeshi, K.

In the last decades, geophysicists and seismologists have focused their attention on the inversion of empirical surface-waves’ dispersion curves from microtremor measurements for estimating the Swaves velocity structure at a site. This procedure allows a fast and convenient investigation without strong active sources, which are difficult to deploy especially in urban areas. In this study we report on a 2D seismic noise array experiment carried out at Bevagna (Central Italy) near the station BVG of the Italian Accelerometric Network (RAN). The site was investigated within the DPC-INGV S4 Project (2007-2009). The Rayleigh- and Love- waves dispersion characteristics were estimated using different methods. The inversion of the dispersion curves was then performed independently, obtaining two estimations for the S-waves velocity profiles. The results of cross-hole logging near the seismic station are used for a comparison. The shear waves velocity profiles estimated by microtremor analyses range up to 150m depth. The two independent procedures provide consistent shear waves velocity profiles for the shallow part of the model (20-30 m in depth) in agreement with the results of the cross-hole logging. Some problems arise between 30 and 40 m in depth in the profile estimated by surface waves. In this range cross-hole logging evidences an inversion of S-waves velocity. Although the cross-hole logging stops at 40 m of depth, we are confident about the results provided by the Rayleigh-waves analysis below 40-50 m. This case study suggests that greater efforts should be devoted to exploit the potential of a coupled analysis of Rayleigh and Love waves from microtremor array measurements.