This article, written by Assistant Technology Editor Karen Bybee, contains highlights of paper SPE 133299, ’Prediction of Uncertainty in Exploration of Unconventional Gas Reservoirs,’ by Mohan Kelkar, SPE, Kyle Bonney, SPE, and Phillip Bonney, SPE, University of Tulsa, originally prepared for the 2010 SPE Annual Technical Conference and Exhibition, Florence, Italy, 19-22 September. The paper has not been peer reviewed.
Once a new unconventional play is identified, companies start acquiring sizable land holdings within the same play. Before making acquisitions, the companies would like to know how much to pay for the bonus, what the strategy is to initiate the drilling program, and how to determine the uncertainty related to estimated ultimate recovery (EUR) from these wells. Most companies use the data from existing wells within the same play or a similar play to estimate what type of production to expect from the land to be acquired.
Shale-gas plays are becoming increasingly important in the world. Shale-gas plays possess some unique characteristics, which are not typical for conventional gas reservoirs. One of the relevant characteristics is the areal extent. Unlike conventional reservoirs, whose areal extent is limited and discontinuous (e.g., fluvial channels can appear and disappear), shale plays extend over many square miles. Drilling a well is almost guaranteed to be successful in penetrating a particular shale formation, but this does not mean the well will be successful. Other intrinsic parameters of the shale such as mineralogy, presence of natural fractures, brittleness, and stress anisotropy can play an important role in determining the success of fracturing in the shale. In addition, external parameters also have played an important role. Many operators believe that with the correct fracturing technique and well configuration, areas that are not successful today can become successful tomorrow.
Because of this belief, the acquisition of shale properties and further evaluation are performed very differently than in most conventional plays. Companies acquire large swaths of acreage (thousands of acres) without worrying about small variations of physical properties within the play and base their estimates of reserves on average characteristics.
Two specific questions addressed in the full-length paper are the following.
1. If a company is interested in acquiring new acreage in a shale play, how would a company evaluate that acreage? If analogous data are available from an adjacent area, which has been developed, how can they be used effectively to quantify potential uncertainties in the acquisition? Is there a way that the variation in potential EURs can be determined correctly?
2. Assuming that acreage is acquired, can a company track its success by comparing the observed values with the expected results? For example, if the company has an initial 10 well drilling program, what is the accepted variation for the observed results before a company can say that the observed results fall within the bounds of predicted uncertainties?