Estimating pore size distribution in carbonate reservoir rocks using joint inversion of NMR and complex conductivity data

In this study, we applied our recently developed joint inversion algorithm of nuclear magnetic resonance (NMR) and complex conductivity (or spectral induced polarization - SIP) data on complex carbonate reservoir rocks to better estimate pore size distribution. The basic principles of NMR and SIP porosimetries as well as the intrinsic petrophysical relationships among them have been discussed and reviewed. A brief introduction of the proposed Levenberg-Marquardt algorithm is also given to demonstrate the workflow of joint inversion method we used. Results of joint inversion from oomoldic grainstone datasets showed similar improvement of pore size distribution estimation as previously studied using Berea sandstone. NMR data obviously contribute more than SIP data during the inversion. The mismatch between the inverted SIP data and the experimental data indicating an alternative SIP model is needed. Here, we proposed a new SIP model considering both electrical double layer polarization as well as the membrane polarization. This new model predicts more realistic real and imaginary conductivities, especially at intermediate frequencies. Our results provide us a new perspective on understanding the SIP mechanisms in rocks with complex structure.