A multiphysics approach to electromagnetic inversion with application to enhanced oil recovery

Traditionally 4D seismic methods are used to image the flow of an injected fluid within a reservoir for the purpose of enhanced oil recovery or CO2 sequestration. However, this method does not provide direct information about changes in material properties and is typically difficult and expensive to repeat accurately. Since the electrical properties of a porous media depend on saturation, it is possible to use electromagnetic (EM) geophysical methods for time-lapse monitoring as the saturation will change during fluid injection. As a consequence of the ill-posedness of the nonlinear EM inverse problem there is no unique solution, and so other information must be incorporated through regularization in order to recover a reasonable solution. As the changes in electrical conductivity depend on changes in saturation, a two-phase flow simulation is used to compute reference saturation models for the regularization, where Archie's law provides a relation between conductivity and saturation. We find the use of multiphysics to constrain the EM inversion produces model estimates that better represent the physics of the flow.