GNGTS 2024 - Atti del 42° Convegno Nazionale

Session 1.2 GNGTS 2024 Trans-dimensional Mt. Etna P-wave anisotropic imaging G. Del Piccolo 1 , R. Lo Bue 1 , B.P. VanderBeek 1 , M. Faccenda 1 , O. Cocina 2 , M. Fireto Carlino 2 , E. Giampiccolo 2 , A. Morelli 3 , J.S. Byrnes 4 1 Dipartmento di Geoscienze, Università degli Studi di Padova 2 Isttuto Nazionale di Geofsica e Vulcanologia, osservatorio Etneo 3 Isttuto Nazionale di Geofsica e Vulcanologia, sezione Bologna 4 School of Earth and Sustainability, Northern Arizona University Trans-dimensional inference identfes a class of methods for inverse problems where the number of free parameters is not fxed. In seismic imaging these methods are applied to let the data, and any prior informaton, decide the complexity of the models and how the inferred felds partton the inversion domains. Monte Carlo trans-dimensional inference is performed implementng the reversible-jump Markov chain Monte Carlo (rjMcMC) algorithm; the nature of Monte Carlo exploraton allows the algorithm to be completely non-linear, to explore multple possibilites among models with diferent dimensions and meshes and to extensively investgate the under- determined nature of the tomographic problems, showing quanttatve evidence for the limitatons in the data-sets used. Implementatons of this method overcome the main limitatons of traditonal linearized solvers: the arbitrariness in the selecton of the regularizaton parameters, the linearized iteratve approach and in general the collapse of the informaton behind the soluton into a unique inferred model. We present applicatons of the rjMcMC algorithm to anisotropic seismic imaging of Mt. Etna with P-waves. Mt. Etna is one of the most actve and monitored volcanoes in the world, typically investgated under the assumpton of isotropic seismic speeds. However, since body waves manifest strong sensitvity to seismic anisotropy, we parametrize a mult-felds inversion to account for the directonal dependence in the seismic velocites. Anisotropy increases the ill-conditon of the tomographic problem and the consequences of the under-determinaton become more relevant. When multple seismic felds are investgated, such as seismic speeds and anisotropy, the data-sets used may not be able to independently resolve them, resultng in non-independent estmates and corresponding trade-ofs. Monte Carlo exploraton allows for the evaluaton of the robustness of seismic anomalies and anisotropic paterns, as well as the trade-ofs between isotropic and anisotropic perturbatons, key features for the interpretaton of tomographic models in volcanic environments. The approach is completely non-linear, free of any explicit regularizaton and it keeps the computatonal tme feasible, even for large data-sets. Corresponding author: gianmarco.delpiccolo@phd.unipd.it