In this paper we describe a novel method for focal-depth determination of induced seismic events. Our approach involves joint interpretation of microseismic and induced-seismicity waveform observations along with multicomponent surface seismic data. The method operates using parallel workflows for processing induced-seismicity data and P-P and P-S data. The output is a set of calibrated P-P times for the microseismic events, which thereby enables the events to be co-rendered and visualized with the seismic data, thus providing stratigraphic control on source locations. The method requires Vp and Vs time-depth control from coincident multicomponent seismic data and is achieved by registration of P-P and P-S reflections from equivalent horizons. Hypocenter vertical locations are initially expressed as the zero-offset focal time (2-way P-P reflection time) and then converted to depth by leveraging methods available for time-depth conversion of the surface seismic data, as well as well ties using synthetic seismograms. Application of this method requires high-quality P- and S-wave picks for microseismic events, which are extrapolated to zero offset. This approach avoids the necessity to build and calibrate a 3-D velocity model for hypocenter location, nor determination of accurate absolute origin times. This method also implicitly accounts for factors that are often ill-constrained for most velocity models, e.g. velocity anisotropy, since these factors similarly affect both the induced seismicity and the 3-D seismic travel times. We apply our new method to an induced seismicity dataset with events up to ML3.6, recorded using a shallow-well monitoring array in Alberta, Canada. Reconciling the seismic processing datum with the microseismic datum was found to be a critical, but not insurmountable, challenge. The inferred focal depths place most induced events at, or above, the treatment depth.
View full article as PDF (0.18 Mb)