A digital rock physics approach to effective and total porosity for complex carbonates: pore-typing and applications to electrical conductivity

Abstract

Recent advances in micro-CT techniques allow imaging heterogeneous carbonates at multiple scales and including voxel-wise registration of images at different resolution or in different saturation states. This enables characterising such carbonates at the pore-scale targeting the optimizing of hydrocarbon recovery in the face of structural heterogeneity, resulting in complex spatial fluid distributions. Here we determine effective and total porosity for different pore-types in a complex carbonate and apply this knowledge to improve our understanding of electrical properties by integrating experiment and simulation in a consistent manner via integrated core analysis. We consider Indiana Limestone as a surrogate for complex carbonate rock and type porosity in terms of macro- and micro-porosity using micro-CT images recorded at different resolution. Effective and total porosity fields are derived and partitioned into regions of macro-porosity, micro-porosity belonging to oolithes, and micro-porosity excluding oolithes’ rims. In a second step we use the partitioning of the micro-porosity to model the electrical conductivity of the limestone, matching experimental measurements by finding appropriate cementation exponents for the two different micro-porosity regions. We compare these calculations with calculations using a single cementation exponent for the full micro-porosity range. The comparison is extended to resistivity index at partial saturation, further testing the assignment of Archie parameters, providing insights into the regional connectivity of the different pore types.

Improvement of remeshed Lagrangian methods for the simulation of dissolution processes at pore-scale

“…The initial geometry of the rock is provided by Ferreira et al through the digital rocks portal [19,20]. Based on the methodology outlined in Arns et al [5] (the concept of backgroud porosity), Ferreira et al argue that none of the voxels are pore-free at the resolution used. They conclude in a 5% porosity for the solid matrix, which is fully compatible with the present formalism, by setting ε = 0.05 in the solid region.…”

Stacked Ensemble Machine Learning for Porosity and Absolute Permeability Prediction of Carbonate Rock Plugs

“…DRP relies on advanced imaging approaches, image processing techniques, and computational methods. The high-resolution digital images of pores and grains structures are used to conduct numerical simulations at the pore scale and infer rock properties such as porosity and directional permeability 11,12,21,22,[13][14][15][16][17][18][19][20] . However, predicting the rock properties using DRP work ow can have signi cant uncertainties when the rock structures are heterogeneous.…”