EARTHQUAKE STATISTICS IN THE EASTERN TENNESSEE SEISMIC ZONE ARE INCONSISTENT WITH AN AFTERSHOCK SEQUENCE

A key challenge to understanding intraplate seismic hazard is separating aftershock sequences that are not representative of long-term earthquake rates. Here, we use modern frequency-magnitude relations to investigate whether high seismicity rates in the Eastern Tennessee Seismic Zone (ETSZ) are plausibly a transient response to a large but unknown prehistoric event. The expected rate of aftershocks depends on the magnitude of the mainshock: To be conservative, we begin by assuming that the entire ~250-km NE trend of epicenters defines that maximum length of a plausible rupture, which corresponds to a M8 event. Modern frequency-magnitude relations in the ETSZ would require that this putative M8 earthquake occurred 196 years ago, but no such event has been documented since European colonization ~400 years ago. We next take a Monte Carlo approach to propagate uncertainty in b-value, modern annual earthquake rates, the exponent of Omori decay and associated constants. Of these realizations, 96% require the mainshock within the past 400 years, meaning that there is only a 4% chance that ETSZ seismicity solely reflects aftershocks of an unknown (pre-European) event. The lack of mention of large-magnitude earthquakes in Cherokee oral history suggests an even greater minimum age of a mainshock, reinforcing the conclusion that earthquakes in the ETSZ are not simply the product of a large but undocumented earthquake. Considering a more reasonable scenario of a M7.5 500 years ago, only 16% (±8%) of modern activity could be ascribed to aftershocks. Therefore, the present mapping of elevated hazard in the ETSZ is warranted. Similar analyses in other intraplate seismic zones will aid in determining the extent to which current earthquake rates are truly representative of hazard.