S54A-02:
Some differences in seismic hazard assessment for natural and fluid-induced earthquakes

Friday, 19 December 2014: 4:15 PM
Arthur McGarr, US Geological Survey, Menlo Park, CA, United States
Abstract:
Although there is little doubt that fluid-induced earthquakes contribute significantly to the seismic hazard in some parts of the United States, assessing this contribution in ways consistent with hazard assessment for natural earthquakes is proving to be challenging. For natural earthquakes, the hazard is considered to be independent of time whereas for fluid-induced seismicity there is considerable time dependence as evidenced, for instance, by the dramatic increase in recent years of the seismicity in Oklahoma. Case histories of earthquakes induced by the development of Enhanced Geothermal Systems and wastewater injection at depth illustrate a few of the problems. Analyses of earthquake sequences induced by these operations indicate that the rate of earthquake occurrence is proportional to the rate of injection, a factor that, on a broad scale, depends on the level of energy production activities. For natural earthquakes, in contrast, the rate of earthquake occurrence depends on time-independent tectonic factors including the long-term slip rates across known faults. Maximum magnitude assessments for natural and fluid-induced earthquake sources also show a contrast in behavior. For a natural earthquake source, maximum magnitude is commonly assessed from empirical relations between magnitude and the area of a potentially-active fault. The same procedure applied to fluid-induced earthquakes yields magnitudes that are systematically higher than what is observed. For instance, the maximum magnitude estimated from the fault area of the Prague, OK, main shock of 6 November 2011 is 6.2 whereas the magnitude measured from seismic data is 5.65 (Sun and Hartzell, 2014). For fluid-induced earthquakes, maximum magnitude appears to be limited according to the volume of fluid injected before the largest earthquake. This implies that for a given fluid-injection project, the upper limit on magnitude increases as long as injection continues.