Dallas-Fort Worth earthquakes coincident with activity associated with natural gas production

O 31 October 2008 and the following day, numerous Dallas-Fort Worth (DFW) residents called 911 to report experiencing several small earthquakes, accompanied by loud booming noises and the shaking of walls and furniture. Using data recorded by regional seismic stations, the USGS National Earthquake Information Center (NEIC) located nine earthquakes with magnitudes between 2.5 and 3.0. On 16 May 2009, this scenario repeated itself, as local residents felt three earthquakes and the NEIC located four (largest = magnitude 3.3). A third sequence of felt events began on 2 June 2009, approximately 65 km southwest near the city of Cleburne, Texas, but has not yet been studied in detail (Figure 1). For various reasons these earthquakes caused concern locally and induced a spirited response from the media. News stories raised the possibility that “drilling” to recover natural gas from the Barnett Shale might cause such earthquakes. Tarrant and Dallas Counties, home to Fort Worth and Dallas, respectively, constitute a major urban center with a combined population of about four million, and an underlying worry was that these earthquakes might be harbingers of larger, more damaging events to come. Following the October earthquakes, seismologists from Southern Methodist University (SMU) borrowed six, threecomponent, broadband seismographs from an NSF-supported instrument pool and operated them at sites in Tarrant and Dallas Counties (Figure 2) between 9 November 2008 and 2 January 2009. Although the NEIC reported no felt earthquakes during this period, the SMU stations recorded numerous local events, including 11 earthquakes between 20 November and 2 December with exceptionally well-recorded P and S arrivals on three stations, AFDAD, AFMOM, and CPSTX. Th e present paper summarizes our analysis of seismograms of the DFW sequence and reports precise locations for 11 well-recorded but “non-felt” events. Using seismological data and other information available in the public record, we show that: (1) In 2008 prior to 29 October, we detected no earthquakes occurring near DFW, including earthquakes too small to be locatable by the NEIC; (2) the 11 hypocenters have a preferred focal depth of 4.4 km and lie along a 1.1km SW-NE line; and (3) the mean epicenter estimate of the 11 events is less than 0.5 km from a 4.2-km deep saltwater disposal (SWD) well where injection began on 12 September 2008, seven weeks before the DFW focus became active. On the basis of time and spatial correlations, we conclude the DFW sequence may be the result of fl uid injection at the SWD well, but we are puzzled as to why earthquakes occur at this particular location but not near other SWD wells in the region. Finally, we discuss the DFW earthquakes in the context of regional historical seismicity, which includes both Dallas-Fort Worth earthquakes coincident with activity associated with natural gas production