GNGTS 2024 - Atti del 42° Convegno Nazionale

Session 3.2 GNGTS 2024 Combining actve and passive methods to understand the seismic velocity distributon in a thick Quaternary succession of the Po Plain. (Terre del Reno, Ferrara, Italy) G. Di Giulio 1 , L. Minarelli 1 , M. Stefani 2 , G. Milana 1 , G. Tarabusi 1 , M. Vassallo 1 , S. Amoroso 1,3 , A. Afatato 4 , L. Baradello 4 , L. Petronio 4 1 Isttuto Nazionale di Geofsica e Vulcanologia (Rome, Italy) 2 Università degli Studi di Ferrara (Ferrara, Italy) 3 Università degli Studi di Chiet-Pescara (Pescara, Italy) 4 Isttuto Nazionale di Oceanografa e Geofsica Sperimentale – OGS (Trieste, Italy) Introducton To properly evaluate the local efects of earthquakes and improves the seismic profles inversion, a good knowledge of the seismic velocity depth distributon is needed. Acquiring such a distributon is however a challenging issue in fast-subsiding alluvial sedimentary basins, with thick unlithifed, low-velocity successions, and reduced seismic impedance contrasts, as in the Po Plain (Mascandola et al ., 2021). Passive acquisiton techniques, based on stand-alone statons and focused on the estmaton of resonance frequency, can provide only raw informaton on the local velocity profle and are not partcularly efectve, by themselves, in areas as the lower Po Plain. Their proper interpretaton needs independent stratgraphic data on the subsurface. Other passive techniques, based on array confguratons of seismic statons, with varying geometries and apertures, increase the frequency band of analysis. Recent technology improvements and the advances in the array processing of three-component ambient-vibraton recording further constrains the inversion process, generatng a reliable shear-wave velocity profle, and supportng the extracton of the Rayleigh-wave elliptcity (Whatelet et al ., 2018). HVSR and array methods however produce the most reliable results only in presence of sharp seismic contrasts, developed at a relatvely shallow depth (approximately < 200 m), which is not the case of the study area. An efectve approach to the reconstructon of deep velocity profles, in similar sedimentary basins, consists in the combinaton of passive and actve methods. Downhole and seismic refecton prospectng can provide reliable S and P velocity profles. Petronio et al . (2023) shows an example of a successful applicaton of such actve methods in a case study from the Po Plain area, about 25 km west to our investgaton site. The authors derived shear-wave (Vs) and compressional (Vp) velocity tomography models down to a depth of about 300 and 700 m respectvely; Vs never exceed 600 m/s. Our study combines multple geophysical techniques to defne the seismic wave propagaton model in a site at the south-west of Ferrara, near San Carlo (Fig. 1A), afected by the 2012 earthquakes. We acquired one of the deepest down-hole measurements so far available in the Po Plain, providing a rare opportunity to calibrate the velocity models produced by surface geophysical investgatons. The Quaternary succession was drilled to the depth of almost 400 m,