Small-scale Ambient Noise Tomography and rapid Vs estimation for near-surface seismic characterization in complex environments: the case of Nirano Mud Volcano system

This thesis presents the development, implementation, and application of a novel methodology for near-surface seismic characterization based exclusively on the analysis of ambient seismic noise. The work focuses on the extraction of dispersive information about surface waves (Rayleigh waves) using minimal acquisition configurations, with particular attention to two-station approaches. The main objective is to evaluate the reliability of such method in complex and heterogeneous contexts, where the use of extended arrays or active seismic sources can be impractical. After a discussion of the theoretical foundations of surface wave-based tomographic strategies and the main assumptions and limitations of ambient noise-based methods, the thesis describes the main codes and algorithms developed during the work. These include a partially automatic program for extracting phase dispersion curves from pairs of stations, which includes an innovative picking algorithm based on the joint use of group and phase velocity curves, capable of overcoming the ambiguities typical of two-station methods. A linear 2D travel-time tomography inversion approach, implemented in a single step while simultaneously considering all analysed frequencies, is also illustrated, along with a novel dispersion curve inversion algorithm designed to efficiently and quickly handle a large number of curves, based on a frequency-depth link to produce the subsoil profile. The different strategies of the methodology were validated through synthetic tests and subsequently applied to a real case study represented by the Salse di Nirano area (Northern Apennines, Italy), a site characterized by the presence of active sedimentary volcanism. The area, already the subject of previous geological and geophysical investigations, is characterized by the presence of lateral heterogeneities even over small distances, thus it constitutes a significant testbed in order to evaluate the reliability of simplified acquisition configurations and station-pair approaches. The results show that significant phase dispersion curves can be obtained even from short-duration records and closely spaced station pairs. The final 2D and 3D shear wave (Vs) velocity models show significant lateral and vertical variations even at small scales, consistent with a complex structural arrangement dominated by the presence of preferential pathways for fluid and gas migration within the investigated site. Overall, despite the limitations of basic two-station methods, the approximation of the adopted tomographic inversion schemes, and the extreme simplicity of the implemented algorithm for dispersion curve inversion, the proposed approach proved robust and effective, returning physically plausible and interpretable results. The developed procedure represents a flexible and rapid tool for passive seismic investigations in constrained operational contexts and is potentially applicable to numerous scenarios, both in urban and natural contexts.