Technical CommentsSeismology

Comment on “How will induced seismicity in Oklahoma respond to decreased saltwater injection rates?” by C. Langenbruch and M. D. Zoback

See allHide authors and affiliations

Science Advances  09 Aug 2017:
Vol. 3, no. 8, e1700441
DOI: 10.1126/sciadv.1700441

Figures

  • Fig. 1 Annual exceedance probability as a function of p and b values in 2017 and 2025.

    For 2017, the probability of a M ≥ 5 event varies between 37 to 80% and remains significantly above the tectonic probabilities. Probabilities remain far above historic (that is, tectonic) values even in 2025. (A) Modified Omori aftershock decay following L&Z for the complete catalog (black) and a catalog with aftershocks removed (gray). The best-fitting p values are 1.4 (yellow) and 1.2 (red), respectively, which is significantly smaller than p = 2.0 (green) in the study of L&Z. (B and C) Magnitude distributions (colored circles) for different time periods in the study of L&Z (see legend) and maximum likelihood b value fits (colored lines) with Mc = 3.0 (triangles) for western and central Oklahoma. The best-fitting values of b vary between 1.14 and 1.46 (see legend). (D) Annual exceedance probability for tectonic events before 2009 (gray) and predicted behavior based on different values of p and b in 2017 (orange) and 2025 (blue). The values used in the study of L&Z are shown as solid blue and orange curves.

  • Fig. 2 Trends in earthquake numbers are largely uncorrelated from trends in cumulative seismic moment, suggesting a different behavior of seismicity rates of small-magnitude versus large-magnitude events.

    (A) Cumulative seismic moment (red) and cumulative earthquake number above M = 3 (black) with earthquakes M ≥ 5 highlighted by black arrows and vertical dashed lines. The inset shows a map of the study region and earthquake locations. W.O., western Oklahoma; C.O., central Oklahoma; (B) Injection (Inj.) (blue) and seismicity rates (red) in central Oklahoma as well as the triggering threshold (blue dashed line) for induced seismicity in the study of L&Z. None of the M ≥ 5 events occurred at times when injection rates exceeded the triggering threshold even when accounting for the proposed time delay (green lines) by L&Z. (C) Same as (B) for the western Oklahoma region.