Since the first modern Marcellus completion in Washington County, Pennsylvania in 2004, over 2000 Marcellus wells (horizontal and vertical) have either been permitted or drilled through October 2009 in the Appalachian Basin. The Marcellus play has become one of the most widely discussed plays in the nation, with numerous operators drilling and completing wells in Pennsylvania, West Virginia, Ohio, and New York. While the play is still in its infancy, reported production rates and reserves compare very favorably to other established North American shale plays.
The play's wide geographic range has led to numerous and varied completion schemes being utilized. One primary concern has been determining the optimal size and type of stimulation treatment for a given area. One premise has been that equivalent EUR's should be less of an optimization determinant than production curves, especially early-time production, as revenue generated closer in time to capital spent is worth more than far-into-the-future revenue, and thus completions that lead to high initial production rates are more desirable for optimum rates of return and improved net present values. This paper will describe an unsuccessful attempt to reduce the size (primarily proppant volume used) of the frac treatments, thus lowering the overall completion costs, without sacrificing higher-rate, early-time production. The ultimate conclusion was that a substantial reduction in proppant volume, even though it was replaced with additional fluid volume, had a detrimental effect on fracture volume created, early-time production, and EUR. As this optimization process continues, some wells like the reduced sand volume wells discussed in this paper will require restimulation to maximize overall fracture network creation. Two successful horizontal shale refracture stimulations will also be described in this paper.