Bayesian Inversion of Surface-Based ERT Data Using a Structure-based Prior
G. de Pasquale, J. Doetsch and N. Linde
Event name: 23rd European Meeting of Environmental and Engineering Geophysics
Session: Modelling and Inversion in Near-surface Geophysics I
Publication date: 03 September 2017
Info: Extended abstract, PDF ( 1.26Mb )
Price: € 20
Inversion results obtained by surface-based electrical resistivity tomography (ERT) are strongly dependent on model regularisation (deterministic inversion) or the prior model (Bayesian inversion). Here, we present the first results of using a structure-based prior in Bayesian inversion of ERT data using a Markov chain Monte Carlo method. The method can handle unstructured meshes, which implies that topography and internal boundaries can be accounted for. The results obtained are greatly improved compared to those obtained by a prior based on uncorrelated model parameters. In the future, we will consider the problem of inferring the sediment-bedrock interface depth (and the associated uncertainty) under strong geological heterogeneity below and above the bedrock.