The Significance of Sub-horizontal Fractures in the Development of Discrete Fracture Networks Leading to Fracture Growth during Hydraulic Stimulations
The earliest models of hydraulic stimulations considered that the fractures grow in the direction of SHmax in symmetric, bi-wing, and vertical planes around the treatment well. Modeling of these treatments has progressed to incorporate structures like natural foliations in the rock to even crosscutting systems of pre-existing vertical fractures along pre-defined grids. However, a preference by some for the earlier, simplistic model still exists, and by correctly incorporating the three-dimensional fracture network, the calibration of reservoir simulators and geomechanical models can be improved. Observation of the discrete fracture network (DFN) responsible for fracture growth is possible by using microseismicity recorded during these stimulations under appropriate recording conditions. Based on seismic moment tensor inversion and source parameter analyses of observed microseismicity, we have identified that a complex fracture network with varying levels of interconnectivity and consisting of sub-horizontal and sub-vertical fractures appears to be responsible for the overall fracture growth. In one case example of a shale play, we identify a marked preference for sub-horizontal fractures in these DFNs, suggesting that delamination of the bedding planes in these units may represent a significant structural weakness that is exploited by the injection, if only temporarily. This behavior is enhanced by a regional compressive stress field. The moment tensors also reveal that most fracturing is mixed-mode tensile-shear along with more classic shear failures during injection. These observations suggest that the stress-strain conditions in the reservoir are not uniform during these treatments, and fractures of one orientation at any point in time may react differently than fractures at other points in time as a result of stress shedding and transfer. On-going Coulomb-stress analysis of these fractures is being considered to identify stress changes resulting in the initiation and propagation of new fractures. In particular, the role of sub-horizontal fractures will be examined to reveal their response to the treatment. Based on these analyses, it may be suggested that hydraulic fracture stimulation models should include sub-horizontal fractures, fracture complexity (three-dimensional behavior of the DFN), and the degree of interconnectivity for proper assessment.
AAPG Search and Discovery Article #90142 © 2012 AAPG Annual Convention and Exhibition, April 22-25, 2012, Long Beach, California