One purpose of monitoring earthquake activity in northeastern North America is to discover which geologic structures are seismically active in this region. If seismically active structures can be found, they can be studied to decipher their seismic history and their potential for strong earthquakes. No seismically active geologic structures have yet been confirmed in the northeastern United States (Ebel and Kafka 1991). The only earthquake with observed surface faulting in northeastern North America took place in the Ungava Peninsula of northern Quebec in 1989 (Adams et al. 1991). Other than some minor offsets of glacial striations (Oliver et al. 1970), no geologic evidence of Holocene surface faulting in the northeastern United States has been reported in the literature. Furthermore, the seismicity that has been detected by modern regional seismic networks in the northeastern United States does not align convincingly along known or suspected geologic structures. Nevertheless, the persistence of small-earthquake activity over time and the historic occurrences of past damaging earthquakes (e.g., Ebel 1996; Ebel 2000; Ebel et al. 2000; Ebel 2006) indicate that there must be some seismically active structures in the region that are capable of hosting earthquakes above magnitude 6.0. Because such earthquakes are capable of causing significant damage, there is great incentive to learn which structures are seismically active in this heavily populated region.
Between fall 2006 and spring 2007, a sequence of earthquakes took place near the town of Bar Harbor on Mount Desert Island on the coast of Maine (figure 1). The largest earthquake in the sequence was Lg-magnitude (MLg) 4.2. It caused several rock falls in Acadia National Park near Bar Harbor, forcing the closure of several hiking trails and one road (figures 2 and 3; http://www.nps.gov/acad/photosmultimedia/earthquakephotos.htm). Acadia National Park has many steep (almost vertical) rock faces of jointed granite, and some rockslides occur annually due to weathering effects. In this earthquake sequence, only the largest of the events generated sufficiently strong ground motions to generate rock falls of the unstable slopes. A water well at McFarland Hill near Bar Harbor that was being monitored by the U.S. Geological Survey showed an unusual drop in water level of about 2 m immediately following this event (the water level data can be accessed through the Web site http://waterdata.usgs.gov/nwis/gw). The MLg 4.2 earthquake was felt over the southern two-thirds of Maine with a few felt reports from New Hampshire (see the community Internet intensity map at http://pasadena.wr.usgs.gov/shake/ne/STORE/Xtib1_06/ciim_display.html) and was the largest event centered in Maine since 1988. A total of 38 earthquakes were detected by seismic stations at regional distances from the Bar Harbor area from the start of the sequence on 22 September 2006 until the end of 2006. Two more events were detected in the spring of 2007. The purpose of this paper is to report on an analysis of the relative locations of the Bar Harbor earthquakes detected by the regional seismic network and to use the results of that analysis to assess what geologic structure might have been seismically active in this earthquake sequence.