Timo Lähivaara – Estimation of aquifer dimensions from seismic signals using the Bayesian inversion approach Quand 7 janvier 2014 Ajouter au Calendrier Télécharger ICS Calendrier Google iCalendar Office 365 Outlook Live In this study, the goal is on developing computational tools to estimate aquifer dimensions from seismic measurements. The poroelastic signature from an aquifer is simulated and methods to use this signature to estimate the level of the water table and aquifer thickness are investigated. As the numerical forward model, we employ the spectral-element (SE) method. The SE method has several features that make it an attractive candidate for large-scale wave simulations since it combines the accuracy of the global pseudospectral method with the flexibility of the FE method. We adopt the Bayesian approximation error approach in which a statistical model is constructed for the errors that are induced by using model approximations such as sparse meshing and simplified physical models. As the computing infrastructure, we use a Cray XT4/XT5 massively parallel processor supercomputers Louhi and Sisu, which are a part of the Finnish IT Center for Science (CSC) computing environment.