Pore-Scale Network Modeling of Microporosity in LowResistivity Pay Zones of Carbonates

Abstract

Carbonates present an important research area as classical empirical methods primarily developed for sandstone characterizations give erroneous results when being applied to carbonates. Specifically, the question of applicability of the Archie equation for these rocks is of great interest in the industry since it is directly related to reserves estimation. Archie equation often calculates high water saturation for intervals dominated by carbonates which would yield zero or little water-cut when being put on production. This phenomenon is known as “Low Resistivity Pay” (LRP). The problem arises from the inherent multi-scale heterogeneity of these rock types where different scales of pore populations constitute the pore space. As a result, the traditional techniques employed in digital rock physics has to be modified for accurate descriptions of the pore space in carbonates. Here in this work, we develop a state-of-the-art pore-scale network model (PSNM) that populates pore space for the microporous zones of carbonates and consequently, simulates displacement sequences during drainage and imbibition. To generate representative pore networks, the digital 3D images of the pore space should be larger than the representative elementary volume (REV) of the sample where all the effective pore space (i.e., contributing to the flow) are fully resolved. Existence of mixed-wet states in carbonates reduces the minimum pore sizes that contribute to the flow of the wetting phase. At the same time, inherent heterogeneity of carbonates increases their REV. This means that high-resolution images of large sample volumes have to be collected and processed. This, however, is not possible due to the physical and computational power limitations of the current imaging tools and computers. Therefore, it is often the case that the 3D image of a REV does not have high enough resolution, thus, overlooking a large number of micropores, which usually account for a significant portion of the pore space in carbonates. In this work, we develop a PSNM to investigate microporosity impact on the electrical and transport properties of 2D (lattice-based) and 3D pore networks. We have addressed the resolution problem by generating stochastic pore networks spatially located within the domains of unresolved zones (i.e., microporosity). The model stochastically reconstructs the pore space at higher resolutions based on the given input parameters of local porosity and pore size distribution for the microporous zones. We use two carbonate samples: one outcrop of Estaillades limestone and one reservoir limestone. The latter is taken from the Mississippian formation of the Osagean age in the STACK play in Oklahoma. For this interval, traditional techniques suggest high water saturation, however, core analysis reveals a significant oil saturation throughout the zone. The results indicate a strong effect of the pore size distribution on the electrical and transport properties of these carbonate samples. The model allows to simulate and explain non-Archie behavior of the Resistivity Index curve. It also allows investigating the extent to which microporosity can have an impact on the petrophysical properties of carbonates by identifying key parameters with the biggest impact on the simulated properties.