How Incorporating More Data Reduces Uncertainty in Recovery Predictions
Fernando P. Campozana, Larry W. Lake, Kamy Sepehrnoori, 1999. "How Incorporating More Data Reduces Uncertainty in Recovery Predictions", Reservoir Characterization—Recent Advances, Richard A. Schatzinger, John F. Jordan
Download citation file:
From the discovery to the abandonment of a petroleum reservoir, there are many decisions that involve economic risks because of uncertainty in the production forecast. This uncertainty may be quantified by performing stochastic reservoir modeling (SRM); however, it is not practical to apply SRM to account for new data every time the model is updated.
This paper suggests a novel procedure to estimate reservoir uncertainty (and its reduction) as a function of the amount and type of data used in the reservoir modeling. Two types of data are analyzed: conditioning data and well-test data; however, the same procedure can be applied to other data types.
SRM is typically performed for the following stages: discovery, primary production, secondary production, and infill drilling. From those results, a set of curves is generated that can be used to estimate (1) the uncertainty for any other situation and (2) the uncertainty reduction caused by the introduction of new wells (with and without well-test data) into the description.
Figures & Tables
Optimum reservoir recovery and profitability result from guidance by an effective reservoir management plan. Success in developing the most appropriate reservoir management plan requires knowledge and consideration of (1) the reservoir system, including rocks, fluids, and rock-fluid interactions, as well as wellbores and associated equipment and surface facilities; (2) the technologies available to describe, analyze, and exploit the reservoir; and (3) the business environment under which the plan will be developed and implemented. Reservoir management plans de-optimize with time as technology and the business environment change or as new reservoir information becomes available. Reservoir characterization is the process of creating an interdisciplinary high-resolution geoscience model that incorporates, integrates, and reconciles various types of geological and engineering information from pore to basin scale. The reservoir data are then conceptually and quantitatively modeled and compared to the historical production data and fluid flow distribution patterns within and beyond the limits of the reservoir to match well production histories and predict their behavior. The goals of reservoir characterization are to simultaneously (1) maintain high displacement efficiency, (2) optimize high sweep efficiency, (3) provide reliable reservoir performance predictions, and (4) reduce risk and maximize profits. Notice that in addition to the technical concepts that we normally associate with "characterization," maximizing profits is an essential element of this process. Papers from the Fourth International Reservoir Characterization Technical Conference (1997), sponsored by the U.S. Department of Energy, this publication is a unique compilation of 27 papers covering every aspect of reservoir characterization and has been a popular AAPG publication since that time. Using an interdisciplinary approach, the papers address qualitative information as well as integrated quantified data and culminate in a fully integrated study.