Managing the Impact of Characterization and Modeling Uncertainties on EUR Assessment in Unconventional Plays

摘要

Estimated Ultimate Recovery (EUR) assessments methods have in common a history matching period followed by a forecast. In unconventional reservoirs, the forecast period can be derived from a decline curve or from an analytical or numerical reservoir model. The latter have the advantage of considering not only production rates but also producing pressures used to constrain the formation and fracture parameters inputs in the model. Combining methods such as Rate Transient Analysis (RTA) and Diagnostic Fracture Injection Tests (DFIT) has proven efficient in improving fracture and formation characterization. Yet, the duration of the transient linear flow periods, due to the ultra-low permeability of unconventional formations, makes it impossible to uniquely characterize formation permeability and fracture dimensions from these methods. Further improvements were reached by integrating well microseismic surveys, petrophysical analysis, and neighboring well results, but all these methods have inherent uncertainties and upscaling methodology issues that makes forecast still largely uncertain. As a consequence, the uncertainties in the input parameters of various models, as well as the intrinsic limitations of the models themselves, still generate widely different Estimated Ultimate Recovery (EUR) values. This paper presents a workflow that takes into account, in a systematic manner, uncertainties in formation characteristics, fracture parameters, and modeling assumptions, in order to determine the impact of these parameters on the EUR ranges. The paper describes how the methodology has been applied to existing fields, and which conclusions were drawn. The purpose of this work is to derive reliable EUR estimations ranges than can be trusted for development decisions such as optimum fracture design and land value estimates.