Remote Dynamic Triggering of Earthquakes in Three Unconventional Canadian Hydrocarbon Regions Based on a Multiple-Station Matched-Filter Approach

We investigate the occurrence of remote dynamic triggering in three Canadian unconventional hydrocarbon regions where recent fluid injection activity is correlated with increasing numbers of earthquakes. We select mainshocks with an estimated local peak ground velocity exceeding 0.01 cm/s occurring between 2013 and 2015, when station coverage was increased to monitor injection activity. A twofold approach, using continuous waveform data and an enhanced earthquake catalog created using a multiple-station matched-filter detection algorithm, suggests that remote dynamic triggering occurs at all three regions. The waveform-based approach shows evidence for direct triggering in the surface wavetrain of the mainshock, as well as directly afterward in the coda. The enhanced catalog approach shows qualitative increases in earthquake rates at all three regions that are both immediate, and in some cases, sustained over 10-day time windows and are corroborated with two types of statistical tests: a p-value test to quantify the statistical significance of earthquake rate change following a stressing event and an interevent time test that provides a statistical measure of changes in seismicity rates. The occurrence of both direct and delayed triggering following transient stress perturbations of < 10 kPa in all three regions suggests that local faults may remain critically stressed over periods similar to the time frame of our study (similar to 2 yrs) or longer, potentially due to high pore pressures maintained in tight shale formations following injection. The results interpreted in the context of injection history and recent poroelastic modeling results may have implications for the mechanisms of remote triggering. Namely, triggering via poroelastic stresses may provide a unifying mechanism that can explain both delayed and immediate triggering observations.