Induced Seismicity Monitoring Project

The Induced Seismicity Monitoring Project is a collaborative effort involving various stakeholders including Geoscience BC, BC Energy Regulator (formerly BC Oil and Gas Commission), Natural Resources Canada, andthe Canadian Association of Petroleum Producers. The project aims to monitor anthropogenic activities, such as hydraulic fracturing, and induced seismicity in northeast British Columbia. Over a span of 7 years (2015-2022), Mahan Geophysical Consulting played a pivotal role in ensuring the project's success by offering seismological expertise and solutions.

Throughout the project, our contributions spanned from installing seismic stations to analyzing ground motion data. This collaborative approach involved working closely with colleagues and partners. By providing insights and technical recommendations, we facilitated a deeper understanding of the relationship between fluid injection and seismic activity, thus addressing potential seismic hazards and risks.

Key Activities:

 1. Seismic Sensor Installation and Maintenance:

The installation and maintenance of seismic sensors spanned over a vast area in northeast British Columbia and western Alberta. This strategic placement ensured comprehensive coverage of seismic activity and fluid injection within the Montney Play, Horn River Basin, Liard Basin, and Cordova Embayment. Collaborative efforts involving research institutions (universities of McGill, Calgary, and UBC) and industry partners were pivotal in the network expansion and data sharing. The expanded regional network can capture very small earthquakes and is a significant milestone in accurately assessing the impact of anthropogenic activities on regional seismicity.

2. Network Design and Evaluation:

Properly designed seismic networks ensure suitability for consistent identification of earthquakes with robust source parameters (time, location, magnitude). During the expansion of the seismic network, we provided maps of the minimum detectable magnitude and location accuracy for different network configurations. The performance analysis of the seismographic network was a decisive factor guiding the subsequent efforts in the densification of regional network within the northeast British Columbia and western Alberta.

3. Data Acquisition and Processing:

As more seismic sensors were added to the network, there was a need for automated processing of waveforms and to associate the detected seismic phases with earthquakes that were not detectable before due to scarcity of stations. Data acquisition and processing were set up through the SeisComp3 system. Waveforms recorded at stations are transmitted in real time to the Natural Resources Canada data center in Sidney, BC, where seismic phases are detected, and events are identified.

4. Velocity Model Development:

Detailed velocity model is essential in calculation of robust earthquake location specially depth of earthquakes that are usually uncertain. New velocity models were developed for northeast British Columbia by taking advantage of the available sonic logs that present velocity measurements. The new velocity models result in significant improvement in delineating earthquake location compared to the very simple model used by Natural Resources Canada.

5. Magnitude Standardization:

Accurate determination of local magnitude for induced earthquakes caused by fluid injection is a vital task for regional seismograph network operators in the Western Canada Sedimentary Basin because specific mitigation measures are required by government regulations when the induced events exceed a predefined magnitude threshold. Local magnitudes calculated by Natural Resources Canada uses the original Richter’s 1935 correction factors that were derived for southern California. This leads to incorrect measurement of magnitudes for earthquakes within the Western Canada Sedimentary Basin as attenuation of seismic waves are different between the two regions. New correction factors were published to consider regional attenuation within the Western Canada Sedimentary Basin and to accurately determine the local magnitude. Calibration and standardization of the local magnitude has led to a significant regulatory change that strikes better balance between the economic development of unconventional resources and the protection of public safety and environment.

6. Earthquake Catalog Compilation:

The significant expansion of the seismic network allowed for more comprehensive earthquake catalogs to be compiled. These catalogs are significant improvements over the routine earthquake reports by Natural Resources Canada with much lower magnitude of completeness and more accurate earthquake location and magnitude.

7. Source Parameters Analysis:

Knowledge of the stress condition within unconventional resources is vital information in resource development. Earthquake focal mechanisms can provide such information. Through analysis of the focal mechanisms of induced earthquakes within the Fort St. John Graben, we found that most of these events are caused by strike-slip movements. The dominant nodal plane has a roughly N60°E strike and dip angles exceeding 60°. Stress inversion of the focal mechanisms provided that the orientation of maximum horizontal stress varies between N22°E and N33°E. Utilizing Coulomb failure criteria and Mohr diagrams, the study estimated that pore-pressure increases of4 to 29 MPa (average of 14.8 MPa) are required to trigger shear slip in the analyzed area.

8. Ground Motion Analysis:

The damage potential of induced earthquakes associated with fluid injection is a major concern in hydrocarbon resource development. Although most induced earthquakes are very small in size and not felt by people, relatively larger magnitude earthquakes have occurred within the northeast British Columbia as the result of hydraulic fracturing injection. The two largest events occurred on August 17, 2015, and November 30, 2018, with moment magnitudes of 4.6. Recorded ground motions and macroseismic intensity perceived by people are two important sources of information for the assessment of ground motions. The isoseismal map of the November 30, 2018, earthquake in the Septimus region of northeast British Columbia shows an approximate 10 km radius around the mainshock for Modified Mercalli Intensities of 4–5. Although shaking intensities from shallow earthquakes (depth ≤ 5 km) can be higher than deep events (depth ≥ 10 km) at close distances (10–15 km), they tend to decrease abruptly at greater distances to become lower than deep events. The localization of large intensities from induced earthquakes within the meizoseismal area warrants special attention in future resource developments and call for systematic intensity data collection within the Western Canada Sedimentary Basin.

9. Seismicity Parameters Study:

The frequency–magnitude distribution of induced earthquakes in several temporal and spatial clusters show fundamental discrepancies between seismicity in the southern Montney Play and the northern area. Using a well-located data set of earthquakes in southern Montney, we observed that there is a depth dependence for the Gutenberg-Richter b value parameter for which seismicity clusters with deeper depths show lower b values than those with shallower depths. These findings have important implications regarding seismic hazard assessments from induced seismicity within the Montney Play.

10. Ground Motion Prediction Equation:

Ground motion prediction equations are vital components in seismic hazard assessments, which provide estimates of ground motion amplitudes for given magnitudes, distances, and site conditions. Ground motion prediction equations were obtained for two regions within the Montney Play of northeast British Columbia using the recorded ground motions by the expanded network.

11. Fluid Injection and Seismic Activity:

Extensive evidence has shown that seismicity is correlated with injection activities both spatially and temporally. Analyzing the controlling factors of injection-induced seismicity and the relationship between parameters of injection and seismicity (such as total pumped fluid volume and magnitude) have been the subject of many studies. We analyzed the seismic activity in the northern Montney Play of British Columbia and its connection with fluid injection (hydraulic fracturing and long-term injection of gas and wastewater disposal) in the region. Spatial and temporal correlation of seismic activity with the fluid injection in the region revealed that these events are better correlated with hydraulic fracturing than other types of injection. Detailed analysis of earthquake locations constrained their depths above the target zone where hydraulic fracturing was taking place. Moment tensor solution for the largest event in this area (August 17, 2015, with moment magnitude of 4.6) shows a predominantly thrust mechanism in the northwest–southeast direction.