Geostatistical population-mixture approach to unconventional-resource assessment with an application to the Woodford Gas Shale, Arkoma Basin, eastern Oklahoma

Abstract

Evaluation of resources such as tight sands and gas shales requires the formulation of assessment models that are different from those used for the inference of conventional resources. Formulations in present use are based in classical statistics that ignore the partly organized and partly random geographical variation of attributes related to the occurrence of hydrocarbons. This paper is the third in a series of methodological tests aimed at enhancing the assessment of unconventional resources through more-effective use of implicit and explicit information contained in the data, more-accurate evaluation of resources, and more-informative display of results. Reprocessing of estimated-ultimate-recovery (EUR) data at the Woodford gas shale in Oklahoma shows that subdivision of the play into areas as homogeneous as possible can produce results comparable to those obtained using several variables correlated to local productivity.