Numerical modeling and evaluation of lab-scale CO2-injection experiments based on electrical resistivity measurements

Abstract

This paper presents numerical models for validating electrical resistivity (ER)-derived carbon dioxide (CO2) saturation obtained from lab-scale CO2 flooding experiments. ER assessments have generally been implemented using Archie's empirical equation, but it is only applicable to non-saline water, homogeneous medium, and non-conductive minerals. Therefore, it is necessary to develop a reliable numerical model to scale up experimental results by calibrating uncertain properties, such as capillary pressure, relative permeability, and aquifer heterogeneity. Cylindrical sandstone samples from the homogeneous Berea sandstone and heterogeneous Tako sandstone were prepared to evaluate CO2 transport within brine-saturated media. The reliability of the numerical method was verified by the trajectories of ER-derived CO2 saturation (SCO2) in the homogeneous Berea sandstone, which showed similar profiles and a root mean square error (RMSE) of 0.0503. The aquifer heterogeneity of the Tako sandstone influenced CO2 transport, i.e., its frontal velocity, by producing retardation at the less porous layers, whereas the homogeneous Berea sandstone showed consistent movement of the CO2 front. The numerical simulation of the heterogeneous Tako sandstone confirmed the effects of the less porous layers and matched the experimental profiles of SCO2, with an RMSE of 0.0421. The developed models enable validation of ER-derived SCO2 at the early stage and forecasting of the CO2 distribution in heterogeneous saline aquifers.