Acoustic Full Waveform Inversion with Hamiltonian Monte Carlo Method

Full-Waveform Inversion (FWI) is a high-resolution technique used in geophysics to evaluate the physical parameters and construct subsurface models noisy limited data scenario. The ill-posed nature of FWI turns this challenging problem since more than one model can match observations. In probabilistic way, solving demands efficient sampling techniques infer information on estimate uncertainties high-dimensional spaces. We investigate feasibility applying Hamiltonian Monte Carlo (HMC) method acoustic by reflection setup containing different noise level data. propose new strategy for tuning mass matrix based acquisition geometry seismic survey. Our methodology significantly improves ability HMC reconstructing reasonable with affordable computational efforts.