S13B-2851
Statistical discrimination of induced and tectonic earthquake sequences in Central and Eastern US based on waveform detected catalogs
Monday, 14 December 2015
Poster Hall (Moscone South)
Xiaofeng Meng1, Clara Daniels1, Eva Smith2, Zhigang Peng1, Xiaowei Chen3, Lara S Wagner4, Karen M. Fischer5 and Robert B Hawman6, (1)Georgia Institute of Technology Main Campus, Atlanta, GA, United States, (2)Grinnell College, Grinnell, IA, United States, (3)University of Oklahoma Norman Campus, School of Geology and Geophysics, Norman, OK, United States, (4)Carnegie Institution for Science Washington, Washington, DC, United States, (5)Brown University, Dept. of Earth, Environmental and Planetary Sciences, Providence, RI, United States, (6)Univ Georgia, Athens, GA, United States
Abstract:
Since 2001, the number of M>3 earthquakes increased significantly in Central and Eastern United States (CEUS), likely due to waste-water injection, also known as “induced earthquakes” [Ellsworth, 2013]. Because induced earthquakes are driven by short-term external forcing and hence may behave like earthquake swarms, which are not well characterized by branching point-process models, such as the Epidemic Type Aftershock Sequence (ETAS) model [Ogata, 1988]. In this study we focus on the 02/15/2014 M4.1 South Carolina and the 06/16/2014 M4.3 Oklahoma earthquakes, which likely represent intraplate tectonic and induced events, respectively. For the South Carolina event, only one M3.0 aftershock is identified by the ANSS catalog, which may be caused by a lack of low-magnitude events in this catalog. We apply a recently developed matched filter technique to detect earthquakes from 02/08/2014 to 02/22/2014 around the epicentral region. 15 seismic stations (both permanent and temporary USArray networks) within 100 km of the mainshock are used for detection. The mainshock and aftershock are used as templates for the initial detection. Newly detected events are employed as new templates, and the same detection procedure repeats until no new event can be added. Overall we have identified more than 10 events, including one foreshock occurred ~11 min before the M4.1 mainshock. However, the numbers of aftershocks are still much less than predicted with the modified Bath’s law. For the Oklahoma event, we use 1270 events from the ANSS catalog and 182 events from a relocated catalog as templates to scan through continuous recordings 3 days before to 7 days after the mainshock. 12 seismic stations within the vicinity of the mainshock are included in the study. After obtaining more complete catalogs for both sequences, we plan to compare the statistical parameters (e.g., b, a, K, and p values) between the two sequences, as well as their spatial-temporal migration pattern, which may shed light on the underlying physics of tectonic and induced earthquakes.
