GNGTS 2018 - 37° Convegno Nazionale

GNGTS 2018 S essione 3.2 663 The environmental noise records were performed at the 156 grid nodes (Fig. 1d), using 6 portables 3-component digital ‘tromographs’. At each node, the environmental noise was recorded for 16 minutes with a sampling frequency of 128 Hz. The records were elaborated, with the H/V spectral ratios technique (Nakamura 1989). The single component spectra, related to almost all the acquisitions carried out, suggest the presence of two areas characterized by velocity inversion, indeed it is possible to observe the decrease of the horizontal components below the vertical one (Fig. 2b). Moreover, even if the H/V>2 SESAME (2005) criterion fails, so no significant resonance peaks are observed (Fig. 2a), however a stratigraphic interpretation can be make (Castellaro and Mulargia, 2009). It is possible to observe, indeed, local minima in the vertical component, which produce the typical eye-shape (Fig. 2b), indicating the presence of a stratigraphic transition in the subsoil. Based on the geological knowledge of the investigated area, which does not indicate lithological variations within the site, one MASWsurvey (Park et al., 1999) has been undertaken to obtain the shear waves velocity profile using a multichannel digital system (SoilSpy Rosina). The survey was performed along an alignment oriented SSW-NNE and located on the central portion of the seismic ambient noise acquisition grid (Fig. 1d). The array was arranged using 25 geophones with natural frequency of 4.5 Hz and spacing of 2 m, with an overall length of 48 m. To obtain the Vs depth profile for the studied area, a joint fit was made between the dispersion spectrum of Rayleigh waves (Fig. 2c) and one of the H/V spectra acquired near the alignment (Fig. 2a). Starting from a subsoil model, characterized by values of thickness, Vs, Vp, density and Poisson’s ratio, assigned on the basis of geological knowledge of the site, the program calculates a dispersion curve and a theoretical H/V spectrum, that are displayed in superposition Fig. 2 Joint fit between HVSR spectrum (a), relating to one seismic ambient noise record (G6) acquired near the MASW survey (see fig. 1d for location) and Rayleigh waves dispersion curve (c), obtained from the MASW survey; b) example of single component spectra, related to one seismic ambient noise sampling (G6) (the area highlighted in yellow show the typical eye-shape, the two blu area highlight the possible presence of areas characterized by velocity inversion) d) velocity-depth profile obtained from MASW-HVSR joint fit.