The Geological Susceptibility of Induced Earthquakes in the Duvernay Play. Issue 4 (19th February 2018)

• Record Type: Journal Article
• Authors: Pawley, Steven; Schultz, Ryan; Playter, Tiffany; Corlett, Hilary; Shipman, Todd; Lyster, Steven; Hauck, Tyler
• Is Part Of: Geophysical research letters. Volume 45:Issue 4(2018)
• Page Start: 1786
• Page End: 1793
• Publication Date: 2018-02-19
• Subjects: induced seismicity -- hydraulic fracturing -- machine learning -- earthquake susceptibility; Geophysics -- Periodicals; Planets -- Periodicals; Lunar geology -- Periodicals; 550
• DOI: 10.1002/2017GL076100
• Languages: English
• ISSNs: 0094-8276

Abstract

Abstract: Presently, consensus on the incorporation of induced earthquakes into seismic hazard has yet to be established. For example, the nonstationary, spatiotemporal nature of induced earthquakes is not well understood. Specific to the Western Canada Sedimentary Basin, geological bias in seismogenic activation potential has been suggested to control the spatial distribution of induced earthquakes regionally. In this paper, we train a machine learning algorithm to systemically evaluate tectonic, geomechanical, and hydrological proxies suspected to control induced seismicity. Feature importance suggests that proximity to basement, in situ stress, proximity to fossil reef margins, lithium concentration, and rate of natural seismicity are among the strongest model predictors. Our derived seismogenic potential map faithfully reproduces the current distribution of induced seismicity and is suggestive of other regions which may be prone to induced earthquakes. The refinement of induced seismicity geological susceptibility may become an important technique to identify significant underlying geological features and address induced seismic hazard forecasting issues. Plain Language Summary: The likelihood of a well inducing an earthquake is described as a function of the underlying geology using a machine learning model; geology input in our model allows for the extrapolation of this likelihood. Key Points: The likelihood of induced earthquakes from hydraulic fracturing is captured spatially This spatial likelihood is described as a function of the underlying geology The modeled likelihood suggests areas of further research into geological factors which contribute to induced earthquakes