Reservoir Modeling in Support of Shale Gas Exploration

Abstract

Abstract Shale gas and other unconventional gas plays have become an important factor in the United States energy market, and are often referred to as statistical plays due to their high heterogeneity. They present real engineering challenges for characterization and exploitation, and their productivity depends upon an inter-related set of reservoir, completion and production characteristics. Shale gas plays are generally characterized by low geologic risk and a high commercial risk. Extensive and continuous deposits of tight, usually naturally-fractured shale provide the duality of a potentially-productive reservoir being the hydrocarbon source. Commercial production is a huge unknown in these plays, and reservoir modeling as well as production predictions involve considerable uncertainty. Because of the large number of unknowns, a merely deterministic approach is often incapable of capturing the complete impact of all interdependencies present in a shale gas resource play. Consequently, one must take into account multiple scenarios to find better exploitation plans. Tools are therefore needed to identify the most important geologic and engineering factors, and to quantify the range of variability in uncertain variables. Reservoir simulation coupled with stochastic methods, i.e., Monte Carlo and geostatistical procedures, have provided excellent means to predict production profiles with a wide variety of reservoir character and producing conditions. Defining and representing uncertainties with a quantitative understanding of their respective impacts on commercial achievability is crucial to subsequent decisions involving continued investment for commercial purposes. This paper describes a systematic process employed in the evaluation of a new prospect area (a shale gas play) with very limited available data. In order to properly model the problem with uncertainty, geological and engineering issues were framed within conventional Monte-Carlo procedures and geostatistical characterization algorithms to identify key production parameters so that relevant data can be collected. This process also allows for the investigation of how the combination of a nested natural fracture system, appropriate wellbore design and stimulation are necessary to drive productivity, and provide project results in terms of ranges of outcomes and associated probabilities. Consequently, managers can be in a better position to make informed decisions regarding the uncertainty of such projects.