Abstract
Geophysical inverse problems are non-unique. Through regularization and the use of a priori information we can derive stable and geologically reasonable inversion models. Providing an analysis of the model uncertainty is necessary for the critical task of separating inversion artifacts from robust geological features. Bayesian inference is a widely used approach but is not tractable for large threedimensional electromagnetic problems. However, another approach based on extremizing model parameters, which is called most squares, shows great promise for quantifying model uncertainty and is computationally feasible. Presentation Date: Wednesday, October 17, 2018 Start Time: 1:50:00 PM Location: 213A (Anaheim Convention Center) Presentation Type: Oral

This publication has 15 references indexed in Scilit: