Ground Motions for Induced Earthquakes in Oklahoma

Source Spectral Properties of Small to Moderate Earthquakes in Southern Kansas

“…The Milan event itself has a slightly lower estimated Δ σ (3.6 MPa) than would be expected by extrapolating the scaling trend of equation though still nearly an order of magnitude greater than the median value of the data set as a whole. The early aftershocks of Milan have lower than expected Δ σ values that appear to increase with time, consistent with observations of the Prague, OK, sequence (Sumy et al, ; Yenier et al, ). In early 2015, we observe another temporal cluster of high Δ σ events, with several other sequences later in 2015 and 2016 exhibiting analogous behavior.…”

“…Because ground motion intensities at high frequencies are controlled primarily by stress drop (Baltay et al, , ; Boore, ; Douglas & Edwards, ; Yenier & Atkinson, ), its characterization is of fundamental interest to studies that aim to develop ground motion prediction equations for induced events (Atkinson & Assatourians, ; Atkinson et al, ; Yenier et al, ). In this study, we observe quantifiable time‐dependent and depth‐dependent variations in stress drop, both of which are in accord with the conclusions of Yenier et al () and Atkinson and Assatourians () for ground motions of recent seismicity in Oklahoma. This consistency suggests that the results we present could potentially serve as a basis for future studies focused on quantifying the influence that spatiotemporal and depth‐dependent variations in earthquake source properties may have on the observed ground motion amplitudes of induced earthquakes in the central United States.…”

“…Scientific consensus has attributed much of the elevated seismicity rate to anthropogenic activity and in particular to the injection of wastewater from the oil production process into the Arbuckle Group that is stratigraphically above the granitic basement (e.g., Buchanan, 2015;Ellsworth, 2013;Ellsworth et al, 2015;Rubinstein & Mahani, 2015;Walsh & Zoback, 2015;Yeck et al, 2017). The abrupt increase in seismic hazard within this region (Petersen et al, 2016(Petersen et al, , 2017 has spurred numerous observational studies focused on connections between fluid injection and seismicity rates (Barbour et al, 2017;Choy et al, 2016;Goebel, 2015;Keranen et al, 2013Keranen et al, , 2014Weingarten et al, 2015), the source properties of a subset of the larger events (Boyd et al, 2017;Choy et al, 2016;Cramer, 2017;Sumy et al, 2017;Walter et al, 2017), and observed ground motion amplitudes (Atkinson & Assatourians, 2017;Atkinson et al, 2016;Hough, 2014;Yenier et al, 2017). Although these studies have rapidly advanced scientific understanding of these earthquakes, there is still much that remains unanswered, and the nonstationary nature of the seismicity warrants continued monitoring.…”

Are Ground‐Motion Models Derived from Natural Events Applicable to the Estimation of Expected Motions for Induced Earthquakes?

“…Taken together, these factors suggest that California events having depths of 6-10 km should be reasonable proxies for CENA events (regardless of whether they are natural or induced) having depths of 2-6 km. The veracity of this claim is demonstrated using a database of induced-event ground motions for M ≥ 3:8 at distances < 50 km (Assatourians and Atkinson, 2017;see Data and Resources).…”

“…Response spectra are shown for an M 4 event at R for shallow events, the effect is obscured by the steep scaling of the stress parameter value with focal depth. Based on a limited sample of induced and very shallow natural events, Yenier and Atkinson (2015b) and Yenier et al (2017) found no compelling evidence that the value or scaling of the stress parameter for induced CENA events differs from that of natural events. The behavior of the stress parameter with depth and magnitude is illustrated later in the article (see Fig.…”