Microseismic source interpretation is often used to gauge the spatial extent of hydraulic fractures in unconventional reservoirs. It is usually difficult to interpret this data beyond a gross approximation of the maximum extent of joint activation and fracture propagation, either due to under-resolution of instrumentation, the inversion process, or possibly the interpretation of the moment tensor. Together, these issues lead to a challenging environment for making more complicated operational decisions based on microseismic data.
The approach taken here is to capture micro-seismicity as an emergent phenomenon from the simulation of flow and geomechanics in a fractured reservoir. This model is then applied to an at-scale stimulation operation, including natural fractures sampled from geostatistical distributions of fracture size, orientation, and constitutive parameters informed by field measurements. The results of this study are compared to field observations and are shown to capture some of the key features of microseismicity, including the spatial distribution of events with time, the frequency-magnitude relationship, and the observed source mechanism tensor distributions.
Additionally, such simulations offer the advantage of testing hypotheses regarding the origin of different observed phenomena. Here we use the simulations to test whether we can reject the hypothesis that stress perturbations alone (due to the reorientation of the stress tensor in the vicinity of propagating hydraulically driven fractures) can explain the majority of microseismic events. It is shown that such "dry" microseismicity, i.e., seismicity not connected with the arrival of a fluid front, cannot be rejected, challenging the assumption that microseismicity is indicative of connectivity between the well-bore and locations of events in the reservoir. However, it also suggests that encoded in the micro seismicity may be additional information about the composition and stress state of the reservoir.