Evidence of possible induced seismicity in the Wabash Valley Seismic Zone from improved microearthquake locations

Abstract To evaluate the spatial structure of seismicity in the Wabash Valley of southern Indiana and Illinois, we analyzed data from a temporary seismic network that included 10 three-component, short-period seismometers and a 10-component dense phased array. We produced the first comprehensive catalog of local earth-quakes recorded during the 211-day deployment, from November 1995 through June 1996. The results are dominated by a cluster of 534 microearthquakes, with magni-tudes ranging from 0.6 to 1.8, near the town of New Harmony, Indiana, that were detected and recorded only by the phased array. The remarkable similarity of the cluster events in terms of waveform, magnitude, and temporal distribution suggests a tightly spaced grouping from a single seismic source. We relocated the earthquakes using improved relative arrival times of P and S waves computed by time-domain cross-correlation of the vertical beam traces and by complex cross-correlation of the horizontal beam traces, respectively. Additional constraints on absolute locations were applied using a graphical method of arrival-time difference analysis for six earthquakes recorded by two network stations outside the phased array. The resulting locations define a tight spatial grouping about a zone of post-Paleozoic faulting just west of the Indiana-Illinois border, in White County, Illinois. Average source depths from phased-array location methods and waveform modeling with synthetic seis-mograms indicate, in contrast to larger events in the region, that the earthquakes occurred at depths less than 4 km, within the sedimentary section of the Illinois Basin. We propose that these earthquakes are artificially induced events, likely related to water injection for the purpose of secondary recovery of petroleum in the Illinois Basin. The primary evidence for this includes: (1) tight spatial clustering of earth-quakes; (2) unusually shallow earthquake depths; (3) good spatial correlation of the relocated hypocenters to existing wells and oil fields; and (4) highly repetitive events with only a small range in magnitude