When a shale gas reservoir is being developed, two fundamental questions need to be answered: (1) how much gas is in place and (2) how much gas can be produced when reservoir is depleted to a specific pressure. This paper examines the pore size distribution impact on gas volume in place during reservoir depletion. A calculation procedure for a multiple component system will be presented with an illustration using Barnett core mercury injection data.

Literature suggests that a monolayer adsorption model is not sufficient for shale gas reservoirs with multiple components in volumetric calculations. In this paper, we propose to use the cylindrical Simplified Local Density model with Peng-Robinson Equation of State (SLD-PR EOS) to solve the local hydrocarbon density distribution for gas in micro-and meso-pores. The integration of the local density over the pore width yields an average density. Pore size distribution (PSD) data such as mercury injection provide the pore volume contribution of different pore radii. Coupling the pore volume distribution with the calculated average density, we calculate the gas in place due to pressurization and adsorption under different pressures. The recovery from the shale reservoir is determined by repeating the process at different pressures. Because of the high non-linearity of the SLD-PR EOS, a trust-region optimization algorithm has been used to solve the local density profile.

Pore size distribution has a tremendous impact on the gas storage capacity of a shale formation. Results from this study reveal that neglecting PSD can yield more than 40% errors for original gas in place (OGIP) calculation. Incremental pore volume is modeled with a Log-Gamma distribution for the Barnett field. OGIP sensitivities to the distribution parameters are also investigated. For the same porosity under the same ambient pressure and temperature, more small pores indicate more gas in place. Therfore under the same depletion process, more recovery can be expected from small pores. This paper utilizes MICP data for a Barnett gas shale core sample to demonstrate the calculation procedure for implementing core data, fluid characterization, and SLD -PR EOS modeling to calculate original gas in place.

This paper has the following two novel points: (1) An improved OGIP calculation procedure for multicomponent gas shales is proposed using the cylindrical form of the SLD-PR EOS model and MICP data from core samples; (2) PSD has a tremendous influence on OGIP values in shale formations because of the presence of micropores and neglecting PSD can yield significant errors in OGIP values.

You can access this article if you purchase or spend a download.