Abstract

The rock-pore textural properties in unconventional fine-grained reservoirs are much more complicated than in conventional reservoirs. In fine-grained rocks, this complexity derives from very small component grain assemblages (clay and silt fractions) and very small pores (nano- to microsizes) overprinted by diagenesis. It is further compounded by the pores being developed in intergranular, grain dissolution, and organic-associated-intragranular voids. Additionally, because locally sourced hydrocarbons in unconventional reservoirs, in contrast to migrated hydrocarbons that are present in conventional reservoirs, exist within similar pore sizes as the original saturating water phase, they create geometrically complex fluid distributions.

These imply that, in the context of nuclear magnetic resonance (NMR) logging and petrophysical interpretation of tight oil unconventional reservoirs, pore size can no longer be considered a proxy for fluid type and vice versa. This means using T2 or T1 cutoffs may give inaccurate predictions. Current industry applications of NMR T1-T2 logging have demonstrated reliable interpretation of fluid types and water saturation. However, because petrophysical controls of advanced relaxation effects (such as surface and bulk fluid relaxation properties, pore-size distribution, and wettability) are not properly understood, these applications have been limited to estimating fluid saturations and have not been applied to estimating pore sizes.

We extend the application of NMR T1-T2 measurements in tight oil unconventional reservoirs to model pore-size distributions by using apparent surface relaxivities and bulk relaxation times that have been jointly estimated from poro-fluid relaxations. To test these predictions, modeled pore-size distributions are compared to rock-pore textural properties from petrographic images of core samples. Such comparison allows us to derive insights into controls of mudstone reservoir quality (RQ) in conjunction with the impact of rock fabric endmembers (e.g., matrix-supported clays, grain-supported framework, diagenetic cements, and solid organic matter) on pore-size distributions. With these assumptions in mind, it follows, based on the Kozeny-Carman formulation, that unconventional rock permeability is reliably predicted from the NMR-based pore-size distributions. Furthermore, we deduce the impact of mineral assemblages in the Herron permeability model to infer influence on rock textural properties and ultimately predict permeability in limited data environments.

This paper introduces a novel approach to estimate pore-size distributions per fluid type (water and hydrocarbons) from NMR T1-T2 measurements in unconventional reservoirs. Additionally, petrophysical controls of RQ, storage, and transport properties are derived from NMR-based pore-size distributions and mineral assemblages. This enables more reliable RQ and permeability assessment than typical T2- or T1-cutoff methods.

This content is only available via PDF.
You can access this article if you purchase or spend a download.