Marcellus Shale model stimulation tests and microseismic response yield insights into mechanical properties and the reservoir discrete fracture network

Abstract

We have developed stimulation tests of a model discrete fracture network (DFN) in the Marcellus Shale reservoir, Morgantown, West Virginia. The microseismic response observed from the modeled stage is characteristic of that observed in several stages along the length of the horizontal well, so that the workflow developed in this paper can be easily extended to other stages in this and other Marcellus Shale wells. The model DFN is designed using log data, including fracture image logs from vertical pilot and horizontal wells. Data from these wells provide geomechanical properties, fracture trend and intensity, and stress orientation. Microseismic cluster trends provide additional constraints on geomechanical model development. Results from stimulation tests are used to modify the reservoir DFN and geomechanical model. Modifications ensure consistency with borehole observations. Fractures observed along the length of the horizontal well consist predominantly of one set, whereas two sets are observed in the vertical pilot well. These two sets are required in the model DFN to reproduce the stimulation trend inferred from microseismic data. Northeast asymmetry in the microseismicity associated with hydraulic fracture treatment is interpreted to result from a horizontal drop of [Formula: see text] toward a previously drilled well. The asymmetry is interpreted to result from stress reduction associated with treatment of an earlier parallel well, the presence of a cross-strike structure parallel to the well, or a combination of the two. Limited downward growth, inferred from the microseismic response, required an increase of the minimum stress in model strata underlying the Marcellus.