Ground-motion analysis of an induced-seismicity sequence associated with a hydraulic-fracturing stimulation in a Duvernay shale play, western Canada

The recent increase in the development of the Duvernay shale play, located in the Fox Creek area in central Alberta, Canada, has been closely associated with a sharp increase of seismic activity observed in the same area since 2013. This has been linked with the hydraulic fracturing stimulations required to increase the productivity of low-permeability reservoirs, making necessary the implementation of dense and close seismic monitoring during and after these stimulations to better understand the seismic hazard from such anomalous seismicity, especially at close distances from the stimulated reservoirs. This manuscript presents a case study where a multi-stage hydraulic-fracturing stimulation in the Duvernay-Fox Creek shale was monitored with a near-surface array composed by 1C and 3C short-period geophones deployed in 27-metres deep boreholes, six broadband seismometers, and one strong-motion accelerometer. During the monitored stimulation, 17 seismic events were detected by the regional seismological network of Alberta with reported magnitudes between 1.5 and 3.6. The hypocentre location, seismic moment, moment magnitude, source radius, radiated energy, and focal mechanisms of these 17 events were re-calculated based on the recorded seismograms from the local monitoring array, and on a velocity model derived from a 3D seismic survey and a dipole sonic log from a vertical well located near the stimulated reservoir. The Peak-Ground Velocities (PGV) and Peak-Ground Accelerations (PGA) generated by these events were measured from the waveforms recorded from the six broadband seismometers and the strong-motion accelerometer -located less than 5 km away from the events hypocentre-, and complemented with the same ground-motion parameters measured with regional seismometers located up to 500 km away to obtain a Ground-Motion Prediction Equation (GMPE) for the Fox Creek area calibrated for local and regional distances. Finally, the site response for the local monitoring stations were determined by the direct measurement of the S-wave velocity of the first 27 metres from the short-period geophones deployed in the monitoring boreholes, and from the implementation of the Horizontal-to-Vertical Spectral Ratio (HVSR) method.