GEOSCAN Search Results: Fastlink


TitleGround motion from M 1.5 to 3.8 induced earthquakes at hypocentral distance < 45 km in the Montney Play of northeast British Columbia, Canada
AuthorBabaie Mahani, A; Kao, HORCID logo
SourceSeismological Research Letters vol. 89, no. 1, 2017 p. 22-34,
Alt SeriesNatural Resources Canada, Contribution Series 20170245
PublisherSeismological Society of America (SSA)
Mediapaper; on-line; digital
File formatpdf; jpg; html; csv
ProvinceBritish Columbia
NTS93O/09; 93O/16; 93P/09; 93P/10; 93P/11; 93P/12; 93P/13; 93P/14; 93P/15; 93P/16; 94A/01; 94A/02; 94A/03; 94A/04; 94A/05; 94A/06; 94A/07; 94A/08; 94A/09; 94A/10; 94A/11; 94A/12; 94B/01; 94B/08; 94B/09
AreaSeptimus; Graham; Fort St. John; Dawson Creek
Lat/Long WENS-122.5000 -120.0000 56.7000 55.6000
Subjectsgeophysics; environmental geology; fossil fuels; seismology; earthquakes; seismicity; petroleum industry; hydrocarbon recovery; hydraulic fracturing; earthquake magnitudes; seismographs; seismic waves; attenuation; seismic velocities; strong motion seismology; amplitude spectra; seismic risk; earthquake risk; Montney Play; Canadian National Seismic Network
Illustrationslocation maps; plots; graphs; tables; spectra
ProgramEnvironmental Geoscience Shale Gas - induced seismicity
Released2017 12 13
AbstractGround-motion amplitudes from small potentially induced earthquakes in the Montney Play of northeast British Columbia (MP BC) are used to evaluate the site condition, attenuation, and strong-motion duration. The dataset includes waveforms from 219 events with local magnitude ranging from 1.5 to 3.8 recorded at hypocentral distance < 45 km by the local seismographic stations operated by energy companies, complemented with some waveforms from the regional seismographic stations of the Canadian National Seismic Network. Horizontal-to-vertical (H/V) spectral ratio of ground-motion results in amplification factors of 2.5-7.9 (0.4-0.9 in log unit) and 1.6-12.6 (0.2-1.1 in log unit) for seismographic stations in the Graham and Septimus areas, respectively. Following the Hassani and Atkinson (2016) methodology, VS30 values are estimated for each seismographic station using the fundamental frequency associated with the peak H/V ratios. In this study, ground-motion prediction equations (GMPEs) are obtained for the Graham and Septimus areas for the entire magnitude range (1.5-3.8) and for the combined dataset for magnitudes above 2.5. The obtained geometrical spreading coefficients for the Graham area suggest higher decay in ground-motion amplitudes than those in the Septimus area for the peak ground velocity (PGV), peak ground acceleration (PGA), and response spectral acceleration (PSA) at frequency of 10Hz, whereas the difference is insignificant at lower frequencies. Although, the geometrical attenuation of ground-motion amplitudes for magnitudes above 2.5 is higher than Atkinson (2015) for PGV and PGA, it is comparable for PSA at frequencies 1, 2, 3.3, 5, and 10 Hz. Duration of the strong ground motion for waveforms with PGA > 1 cm=s2 is also calculated from two widely used methods (bracketed [Db] and significant [Ds] durations). Both Db and Ds are strongly correlated with PGA (Db increases while Ds decreases with PGA). Moreover, Db appears to be correlated with magnitude (increases with M), whereas Ds is correlated with distance (increases with hypocentral distance). Although the database in this study includes waveforms with relatively large ground acceleration (PGA of ~115 cm=s2 from an M 3 event), duration of the strong ground motion is, however, very short at higher acceleration levels (e.g., 50 cm=s2) with a typical value of a fraction of a second.
Summary(Plain Language Summary, not published)
We collected ground motion data recorded at close distances to fluid injection sites in two areas (Graham and Septimus) of northeast British Columbia. These data were used to derive the ground motion prediction equation for the two regions, which can be used to better characterize local seismic hazards. We found that the seismic wave amplification and attenuation are both higher in Graham than that in Septimus. We also found that, although the ground motion caused by shallow induced earthquakes can be large, the during of strong shaking is actually very short, on the order of a fraction of a second. This short during is unlikely to cause significant damage to buildings and/or infrastructures.

Date modified: