Abstract

The characterization of the failure process of induced fractures by hydraulic stimulation is fundamental to understanding the generation and evolution of the discrete fracture network within the reservoir. A more detailed analysis of the fracture mechanism can be a powerful tool for identifying fluid flow paths and proppant placement within the reservoir. For example, during the rupture process the energy release is partitioned into different physical processes for which the relative ratio is changed by the presence of fluids or rotations in the local stress field with relation to the frictional resistance in the fracture plane. Fractures occurring in the same host rock under the same stress conditions are expected to rupture similarly, independent of their size (self-similarity). Changes in the scaling relationships of fractures are indicative of a change in the failure process, host rock or in-situ stress. Correlation of failure process data with reservoir rock properties and the in-situ stress field will help identify regions within the reservoir with characteristic types of failures. Further correlation with hydrocarbon production data can be used to develop efficient treatment plans and production diagnostic tools.

In this study we investigate the failure process of ~ 27,000 microseismic (M < -1) fractures induced during a hydraulic fracturing shale completion program in NE British Columbia, Canada by estimating static and dynamic source parameters, such as dynamic and static stress drop, radiated energy, seismic efficiency, moment tensor, fracture plane orientation, slip direction and rupture velocity. On average, the microseismic events have low radiated energy, low dynamic stress and low seismic efficiency, consistent with the obtained slow rupture velocities. Events fail in overshoot mode (slip weakening failure model), with fluids lubricating faults and decreasing friction resistance. Events occurring in deeper formations tend to have faster rupture velocities and are more efficient in radiating energy. Variations in rupture velocity tend to correlate with variation in depth, fault azimuth and elapsed time, reflecting a dominance of the local stress field over other factors.

Further identification of spatial and temporal distribution of families of events with similar characteristic rupture behaviors, based on either rock formation, depth, source type, fracture plane orientation, stress drop, pad or proppant stage, may be used as a proxy for specific fracture network development and hydrocarbon production. This information may be used to determine reservoir properties, constrain reservoir geo-mechanical models with measured physical parameters, classify dynamic rupture processes for fracture models and improve fracture treatment designs. These will be the focus of future studies.

This content is only available via PDF.
You can access this article if you purchase or spend a download.