Stochastic Modelling of Induced Seismicity Clusters in Central Alberta

Date: 4/26/2019

Time: 06:00 PM

Room: Fifth Avenue

For the past several years, hydraulic fracturing has been associated with the increasing seismicity rate in the Western Canada Sedimentary Basin (WCSB). Despite the high density of injection wells in the WCSB, induced earthquake clusters exhibit 1) large spatial and temporal variability and 2) swarm-like behavior that may not follow traditional recurrence laws. Site-specific statistical analyses of such induced seismicity are further challenged by the limited size and completeness of earthquake catalogs. Taking advantage of on-site industrial monitoring and recently-developed detection methods, thousands of M > 0 events were identified to be associated with two distinct hydraulic fracturing well-pads in central Alberta, Canada. In this study, three catalogs are examined and simulated: 1) the SO2 region where “traffic-light” protocol is in effect, 2) the November 2016 cluster containing multiple sub-clusters of earthquakes and several M ~3 events (Tony Creek Dual Microseismic Experiment, Eaton et al., 2018), and 3) the January 2016 cluster characterised by a mainshock with M 4.1 (Wang et al., 2017). To evaluate and characterize the recurrence patterns of these earthquake clusters, we employ a modified Epidemic Type Aftershock Sequence model, where we approximate the rate of seismicity as a linear superposition of the rate due to past earthquakes and the normalized effect of the operation of hydraulic-fracturing wells. Our results support that both the injection strategies and geological features are the key factors that control the occurrence of induced seismicity. The modeled earthquake rate can be used in the Monte-Carlo based stochastic earthquake simulations for a short-term earthquake hazard assessment accounting for nearby hydraulic-fracturing activities.

Presenting Author: Ruijia Wang

Authors