GNGTS 2019 - Atti del 38° Convegno Nazionale

704 GNGTS 2019 S essione 3.2 Dossi M., Forte E. and Pipan M.; 2015: Auto-picking and phase assessment by means of attribute analysis applied to GPR pavement inspection , proceedings of 8 th international workshop on advanced Ground Penetrating Radar, IWAGPR 2015, 7-10 July 2015, Florence, Italy. Grasmueck M., Weger R.J. and Horstmeyer H.; 2005: Full-resolution GPR imaging . Geophysics, 70 (1), K12-K19. Forte E., Pipan M., Casabianca D., Di Cuia R. and Riva A.; 2012: Imaging and characterization of a carbonate hydrocarbon reservoir analogue using GPR attributes . Journal of Applied Geophysics, Special Issue: Recent, Relevant and advanced GPR studies in Applied Geophysics, 81, 76-87. Young R.A., Deng Z., Marfurt K.J. and Nissen S.E.; 1997: 3-D dip filtering and coherence applied to GPR data: a study. The Leading Edge, 16, 1011–1018. McClymont, A.F. et al. ; 2008: Visualization of active faults using geometric attributes of 3D GPR data: an example from the Alpine Fault Zone, New Zealand . Geophysics, 73(2), B11–B23. Zhao W., Forte E., Pipan M. and Tian G.; 2013: Ground penetrating radar (GPR) attribute analysis for archeological prospection. Geophysics, 97, 107-117. Zhao W., Forte E. and Pipan M.; 2016a: Texture Attribute Analysis of GPR Data for Archaeological Prospection . Pure and Applied Geophysics, 173, 2737–2751. Zhao W., Forte E., Colucci R.R. and Pipan M.; 2016b: High-resolution glacier imaging and characterization by means of GPR attribute analysis . Geophys. J. Int., 206(2), 1366–1374. Zhao W., Forte E., Fontolan G. and Pipan M.; 2018: Advanced GPR imaging of sedimentary features: integrated attribute analysis applied to sand dunes . Geophysical Journal International, 213, 1, 147–156. HAMILTONIAN MONTE CARLO INVERSION OF SURFACE WAVE DISPERSION CURVES: PRELIMINARY RESULTS A. Salusti 1 , M. Aleardi 2 1 University of Florence, Earth Sciences Dept., Florence, Italy 2 University of Pisa, Earth Sciences Dept., Pisa, Italy Introduction. Rayleigh wave measurements are highly sensitive to the S-wave velocity ( Vs ) and for this reason they are attractive for geotechnical characterization or seismic site response studies (Socco and Strobbia 2004). Over the last years, the full-waveform inversion of surface waves is getting growing attention thanks to the increased computational power of modern parallel architectures (Gross et al. 2017). Well-established methods rely on dispersion curve inversion under the assumption of a 1D subsurface structure. The dispersion curve inversion is a highly non-linear and ill-conditioned problem. For these reasons, it is crucial adopting inversion approaches that efficiently converge toward the global minimum. In this context, local inversion methods exhibit fast convergence rates but limited capability to explore the model parameter space, resulting in a final solution highly dependent on the initial model. Global search algorithms (genetic algorithms, simulated annealing) exhaustively explore the model space but they usually require a considerable computational effort ( Cercato, 2011). Markov Chain Monte Carlo (MCMC) algorithms exhibit global convergence capabilities and honour the importance sampling principle, but they usually rely on specific MCMC recipes in order to maintain the computational cost affordable. More specifically, MCMC methods are primarily affected by low acceptance rates and show strong correlations between the sampled models. Hamiltonian Monte Carlo (MC) algorithm was designed to circumvent these two critical issues of MCMC algorithms. HMC treats a model as the mechanical analogue of a particle that moves from its current position (current model) to a new position (proposed model) along a given trajectory. The geometry of the trajectory is controlled by the misfit function, which is interpreted as potential energy ( U ), and by the kinetic energy ( K ) and the mass of the particle. After the so-called burn-in period, the ensemble of HMC sampled models can be used to numerically derive the so-called posterior probability density (PPD) function in the model