Improved Core Analysis Measurements in Low Permeability Tight Gas Formations

ABSTRACT

Recovery of hydrocarbon gas in micro-Darcy rock is now becoming more common-place for gas producers. As part of the reservoir characterization associated with these systems, it is required to understand parameters such as the total porosity and permeability of the system, but also the initial water saturation and the effective gas permeability at this connate water saturation. This is still under the realm of what is termed “routine core analysis”, but in these tight formations it is imperative to understand that these tests are actually very non-routine by nature. Current work presents a case study of core analysis that was run on plugs from a tight gas producing formation, and illustrates some of the pitfalls and corrections associated with measurements of this low permeability rock. The case study combines low field NMR measurements with core flooding measurements of porosity and permeability. Cores are initially tested at their connate water saturations: NMR measures the initial water present in each core, and effective porosity and permeability to gas are measured through different core analysis techniques. By considering the time scales required for pressure to propagate within the cores, it is shown that “routine” measurement of permeability through multi-rate Darcy flow of gas is simply not adequate to properly characterize the core. Instead, core properties are measured through analysis of transient pressure data. Cores are then run through a DeanStark cleaning process, and NMR measurements are used to verify the ability of DeanStark to remove all water out of the cores. In this manner, post-cleaning measurements of porosity and permeability can be understood as effective values rather than their assumed absolute properties. This case study illustrates the challenges in making accurate measurements in micro-Darcy rock systems, and helps end-users to understand the dangers in using routine data without knowing how the data was collected.

A full version of this paper is available on JGMAAS Online.