Effects of surface geology on seismic ground motion deduced from ambient-noise measurements, in the town of Avellino, Irpinia Region (Italy)

The effects of surface geology on ground motion provide an important tool in seismic hazard studies. It is well known that the presence of soft sediments can cause amplification of the ground motion at the surface, particularly when there is a sharp impedance contrast at shallow depth. The town of Avellino is located in an area characterised by high seismicity in Italy, about 30 km from the epicentre of the 23 November 1980, Irpinia earthquake (M = 6.9). No earthquake recordings are available in the area. The local geology is characterised by strong heterogeneity, with impedance contrasts at depth. We present the results from seismic noise measurements carried out in the urban area of Avellino to evaluate the effects of local geology on the seismic ground motion. We computed the horizontal-to-vertical (H/V) noise spectral ratios at 16 selected sites in this urban area for which drilling data are available within the first 40 m of depth. A Rayleigh wave inversion technique using the peak frequencies of the noise H/V spectral ratios is then presented for estimating Vs models, assuming that the thicknesses of the shallow soil layers are known. The results show a good correspondence between experimental and theoretical peak frequencies, which are interpreted in terms of sediment resonance. For one site, which is characterised by a broad peak in the horizontal-to-vertical spectral-ratio curve, simple one-dimensional modelling is not representative of the resonance effects. Consistent variations in peak amplitudes are seen among the sites. A site classification based on shear-wave velocity characteristics, in terms of Vs30, cannot explain these data. The differences observed are better correlated to the impedance contrast between the sediments and basement. A more detailed investigation of the physical parameters of the subsoil structure, together with earthquake data, are desirable for future research, to confirm these data in terms of site response.