Gas-Reserves Estimation in Resource Plays

Abstract

This article, written by Senior Technology Editor Dennis Denney, contains highlights of paper SPE 130102, ’Gas-Reserves Estimation in Resource Plays,’ by W.J. Lee, SPE, Texas A&M University, and R.E. Sidle, SPE, Consultant, prepared for the 2010 SPE Unconventional Gas Conference, Pittsburgh, Pennsylvania, 23-25 February. See SPE Econ & Mgmt, October 2010, page 86. The petroleum industry’s lack of understanding of physical properties and the physics that control production from many important resource plays limits the ability to model and forecast production and reserves from these resources with confidence. While measurement technology and modeling accuracy are improving, the industry often is forced to resort to empirical methods that lack the validation usually required for high-confidence results. Some of the procedures in common use for forecasting are reviewed, and key strengths and limitations of the techniques are identified. Introduction Preferably, proper characterization of a resource play begins with production. When attempting to characterize a resource and to forecast production, it is important to know what can and cannot be determined. Thus, pilot tests, resource characterization, and analogs are favored over curve fitting or highly complex reservoir simulation. Analytical models not based on analogs are useful, but only to investigate the generalized incremental effects of different production conditions, such as horizontal vs. vertical wells or completion and well-operation methods. Analogs (usually based on older in-field wells) for decline analysis or type-curve techniques generally would be the preferred methods for high-confidence forecasts and reserves estimates. This approach has proved valuable in examples as diverse as the Antrim shale, Pinedale tight gas, and California diatomite. As a result, high-confidence reserves estimates begin with older wells—wells that have experienced any early-life transient-production effects and have reached a stage in which only stable, long-term trends are the dominant influence on production rate. The considerable complexities of production from resource plays make it important to isolate early effects from those that are fundamental to ultimate recovery. An incomplete understanding of the basic physics of fluid flow and stimulation results in complex-resource plays currently of interest has limited the ability to develop models that provide high confidence, particularly in shales. Because of this lack of understanding, empirical methods take on an unusually important role, but production histories from the most-modern completion and stimulation techniques are limited in duration, causing problems with validation of a forecasting methodology of any kind. The full-length paper provides a detailed critique of the methods used to forecast production and estimate reserves in poorly understood resource plays.