S52C-04:
Spatiotemporal Aftershock Complexity in the November 8th 2011, Prague, OK Earthquake: Insights into the Role of Damage Zones in the Seismic Cycle

Friday, 19 December 2014: 11:05 AM
Heather M Savage1, Katie M Keranen2, David Paul Schaff1 and Caitlin C M Dieck1, (1)Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY, United States, (2)Cornell University, Ithaca, NY, United States
Abstract:
Although most faults are surrounded by a halo of fractured rock known as a damage zone, it is not clear what role damage zones play during the seismic cycle on mature faults. Here, we present a superbly-located foreshock-mainshock-aftershock sequence surrounding the November 8, M5 Prague Oklahoma earthquake, which demonstrates most aftershocks are located within the damage zone surrounding the fault. The 2011 Prague, Oklahoma sequence included three M5+ earthquakes along three different faults over a three-day period. The November 8th event was third in the sequence and captured with an array of nine seismometers with ~2 km spacing, allowing for precise event location. We located more than 1000 foreshocks and aftershocks within a 14 hour time window, and relocated these aftershocks using waveform cross correlation and HypoDD. Because of the location precision, we can use these events to investigate spatial and temporal complexity of the foreshock and aftershock sequences.

First, we compare the aftershock distribution to fracture distributions within damage zones surrounding faults. The aftershock sequence localizes to a reasonable damage zone thickness given the rupture length of the event, according to previously documented scaling between fault length and damage zone thickness. Furthermore, the aftershock density is constant within the fault zone, but falls off precipitously outside of the damage zone. Most aftershocks in this sequence occur within the first hour after the mainshock, and there is some indication of temporal migration of aftershocks away from the fault. Finally, foreshock activity along this fault was limited to the intersection with the fault that had hosted a M5.7 earthquake two days prior. Because this earthquake is potentially linked to fluid waste disposal, we interpret our results in terms of hydraulic pressure changes during the foreshock-mainshock-aftershock sequence.