GNGTS 2015 - Atti del 34° Convegno Nazionale

GNGTS 2015 S essione 3.2 85 the wavelength, in consideration of the fact that the investigation depth of a (Rayleigh) surface wave is linked to its wavelength. Here we assumed β = 1. For the assembly process we consider the 2-D subsoil as discretized into squared blocks where the value of the velocity for each block is calculated by a weighted average procedure. Averaging has recently proved very useful for immediate (or rough) Vs estimation and allows retrieving very satisfactory results even without inversion (Socco and Comina, 2015). Indeed, we built our weighting strategy based on some phenomenological observations, but better strategies may exist. (3) where (4a) otherwise; (4b) ij represent the receiver pair, p k,ij = ( λ k,ij , V S k,ij ) is the corresponding COP, k is an index running on the accepted points in the cloud and 1 = βλ is a suitable fraction of the wave-length λ . The 1.1 factor is thought to empirically translate V R into V S . It is well known that SASW does not allow resolving all the different propagation modes but only one whole apparent mode is retrieved. Note that the weighting average naturally takes into account the presence of multiple modes because multiple values of velocity associated to the same frequency affect different portions of the pseudo-section once expressed in the wavelength-velocity domain. An example of 2-D pseudo–section is shown in Fig. 2a. To judge Fig. 2 – a) Obtained 2-D Vs pseudo-section for a MASW-like survey. b) Example of the influence of different receiver pairs on the interpreted profile(a) and sensitivity of the proposed method with depth. The maximum depth showing a reliable resolution is roughly λ max /3, where d <λ< L, d is the inter-geophonic distance and L is the length of the array. Velocities in the poor resolution zone are evaluated using only distant receivers so that deeper velocity values are correct but horizontally stretched. c) Computed normalized sensitivity corresponding to the Vs profile in (a).The white curve in a and b, highlights the portion with “good resolution” (i.e. in this case, the shallower 6 m, with a number of contributions roughly higher than 30%, in this case).