Updated: September 28, 2017
On September 2, 2017, at 05:57 pm MDT, an Mw (moment magnitude) 5.3 earthquake occurred beneath the Aspen Range in southeastern Idaho. The epicenter of the earthquake was located 15 km east of the town of Soda Springs, Idaho, near Sulphur Peak (yellow star, Figure 1). This shock was reported felt by more than 1800 people in southeastern Idaho, western Wyoming, and northern Utah.
Moment tensor solutions determined by long-period waveform modeling by the U.S. Geological Survey (USGS), Robert B. Herrmann of Saint Louis University (SLU), and the Global Centroid-MomentTensor Project (GCMT) show that the mainshock was caused by dominantly normal slip on either (1) a fault dipping 39° to 60° to the east or east-southeast or (2) a fault dipping 33° to 51° to the west or west-southwest (Table 1). The SLU moment tensor solution is the only one of the four solutions in Table 1 that indicates a significant strike-slip component to the faulting. For the other three solutions, the rake angles all differ by less than 10° from pure normal slip (-90°). The long-period waveform modeling also shows that the center of the fault slip, or centroid, was located at a depth of 9 to 16 km below the earth’s surface (Table 1).
Temporary Seismic Stations
The University of Utah Seismograph Stations (UUSS) and USGS have collaboratively installed a network of eight temporary seismic stations in the region of the Sulphur Peak main shock in order to (1) enable better locations, especially focal depths, for the ongoing aftershocks of this earthquake; (2) provide data that can be used, along with relative location techniques, to improve the locations of the mainshock, foreshocks, and the early aftershocks; and (3) capture near-source ground motions from large aftershocks and less likely, but possible, earthquakes comparable in size to the Sept. 2 mainshock or larger (Figure 1 and Table 2). The improved locations for the earthquakes in the sequence will help to provide a better understanding of the relationship between seismicity and faults in southeastern Idaho and elsewhere along the eastern boundary of the Basin and Range Province, including the Wasatch Front region of Utah. The ground motion recordings will add to the database of ground motions that is used to predict large, damaging ground shaking from future moderate to large earthquakes.
The first two temporary stations were installed by UUSS staff inside buildings in the towns of Soda Springs and Georgetown, Idaho, on Sept. 5 and 6 MDT (both on Sept. 6, UTC, which is 6 hours ahead of MDT). These two stations are equipped with 3-component strong-motion instruments, and were installed primarily to record potential strong, damaging ground motions in populated areas with buildings and other structures. These stations are also proving to be very useful for aftershock locations. The other six temporary stations have USGS equipment that provides 3-component broadband velocity data plus 3-component strong-motion accelerometer data. These six stations were installed by UUSS and USGS staff between Sept. 7 and 12. Two of these stations were sited within the aftershock zone in order to provide control on aftershock focal depths. The other four stations were distributed around the activity at distances of 35 to 50 km in order to provide azimuthal control on aftershock locations, in combination with existing permanent stations. Continuous waveform data from the eight temporary stations are being telemetered to UUSS and USGS recording centers, and are publicly available from the Incorporated Research Institutions for Seismology. The eight temporary stations, along with permanent seismic stations in the area (six within ~80 km) operated by the Idaho National Laboratory, the University of Utah, and the USGS , are providing excellent data on the ongoing aftershock sequence.
Foreshocks and Aftershocks
UUSS detected nine foreshocks of magnitude 1.3 (UUSS coda magnitude, Mc) to 4.1 (USGS body wave magnitude, mb) during the 34 minutes prior to the Mw 5.3 mainshock. The last and largest foreshock occurred 85 sec before the mainshock. The mainshock has been followed by a very energetic aftershock sequence consisting of more than 1900 locatable events, including 26 of magnitude 4.0 or greater and 147 in the magnitude range 3.0 to 3.9 (as of 6:00 pm MDT on Sept. 28). The largest aftershock to date was an Mw 5.0 shock that occurred 8 km south of the mainshock at 03:47 am MDT on September 10. This aftershock was reported felt by nearly 200 people, mostly in southeastern Idaho with a few reports from western Wyoming and northern Utah.
The aftershock epicenters form a north-northwest-trending zone that is approximately 20 km long and up to 10 km wide, as shown in Figures 1 and 2 and in maps and cross sections showing USGS preliminary multiple event relocations. The mainshock epicenter is in the northeastern part of the aftershock zone on Figure 1 and 2, but near the middle of the northeastern edge of the aftershock zone defined by the preliminary USGS relocated epicenters. The aftershocks generally migrated from north to south with time during the first ~9 days after the mainshock, as shown by the color-coded epicenters in Figures 1 and 2 and in the animation below. However, many of the more recent aftershocks have occurred near the middle of the aftershock zone, as best seen in Figure 2. The focal depths of the aftershocks relocated by the USGS are nearly all between 2 and 18 km below sea level, which is approximately 4 to 20 km below the local ground surface given the average surface elevation of ~2 km in the aftershock zone. However, most of the relocated aftershocks have depths between 4 and 12 km below sea level, which is approximately 6 to 14 km below the surface.
Moment tensor solutions determined by the USGS for 60 aftershocks that occurred between Sept. 2 and 16 show mostly normal or oblique-normal faulting on north-northeast- to north-northwest-striking fault planes, with some strike-slip faulting on northeast- and/or northwest-striking fault planes. The centroid depths for these USGS moment tensors range from 4 to 18 km below the surface. Moment tensor solutions determined by Robert B. Herrmann of Saint Louis University for 70 aftershocks that occurred between Sept. 2 and 19 exhibit a range of faulting mechanisms similar to that of the USGS moment tensor solutions. However, in comparison to the USGS results, the SLU moment tensor solutions tend to show larger components of strike-slip motion (see the figures under “Context”). The centroid depths for the SLU moment tensor solutions range from 5 to 12 km below the surface. This depth range is somewhat narrower than the depth range for the USGS moment tensor solutions, although 80% of the USGS centroid depths are within the SLU centroid depth range of 5 to 12 km. The differences between the USGS and SLU moment tensors are probably due at least partially to (1) differences in the velocity model used for the waveform modeling and (2) methodology differences, including the fact that the SLU moment tensors are constrained to have double-couple (pure shear-slip) mechanisms whereas the USGS moment tensors are not.
The Mw 5.3 Sulphur Peak earthquake cannot, at the present time, be attributed to any known fault. The USGS Quaternary Fault and Fold Database of the United States (QFFDUS) does not show any active faults in the immediate vicinity of the Sulphur Peak earthquake or within its aftershock zone (solid black lines, Figures 1 and 2). The closest Quaternary faults to the activity on the USGS QFFDUS map are the west-dipping Eastern Bear Lake fault and the east-dipping western Bear Lake fault, which form the north-trending Bear Lake graben south of the aftershock zone (Figure 1; Haller and Lewis, 2010 a, b). Structural interpretations of seismic reflection profiles by Evans (1991) and Evans et al. (2003) indicate that the Eastern Bear Lake fault is the master fault, and that the Western Bear Lake fault terminates above or at the Eastern Bear Lake fault. The IdahoGeological Survey (IGS) map of “Miocene and Younger Faults in Idaho” (Breckenridge et al., 2003) shows both the Eastern and Western Bear Lake faults extending farther north than is indicated on the fault map accompanying the USGS QFFDUS (dashed black lines, Figures 1 and 2). On the IGS fault map of Idaho, the surface trace of the Eastern Bear Lake fault continues northwestward along the southwestern edge of the Aspen Range to 42° 47′ N. The surface trace of the Western Bear Lake fault continues northwestward past the town of Soda Springs to the western side of Blackfoot Reservoir at 42° 54′ N. These northwestward extensions of the Eastern and Western Bear Lake faults on the IGS fault map are indicated by dashed black lines on Figures 1 and 2 because we consider their locations to be approximate for the following reasons: (1) the original IGS fault map is at a scale of 1:1,000,000, and (2) in many places these fault traces plot more than a kilometer from the edges of the mountain ranges, where young faults are typically found in the Basin and Range Province. Both the IGS and USGS fault maps indicate that the age of most recent movement on the Eastern Bear Lake fault is much younger south of Montpelier (< 10,000 to 15,000 yrs) than to the north (< 750,000 to 1.6 million yrs). The IGS fault map shows the age of most recent movement on the Western Bear Lake fault as < 10,000 yrs everywhere, whereas the USGS fault map shows it as < 15,000 yrs south of Montpelier and < 1.6 million yrs elsewhere.
The possible northwestward extension of the west-dipping Eastern Bear Lake fault runs 5 km southwest of the epicenter of the Sulphur Peak mainshock. Therefore, it does not appear possible for the Sulphur Peak earthquake to have occurred on the Eastern Bear Lake fault. We consider it unlikely that the earthquake occurred on the Western Bear Lake fault because this fault would need to cut the Eastern Bear Lake fault in order to extend as far east as the mainshock epicenter, a possibility that is inconsistent with the structural interpretations of Evans (1991) and Evans et al. (2003). The aftershocks of the Sulphur Peak earthquake span the mapped surface trace of the Eastern Bear Lake fault in map view (Figures 1 and 2). However, considering that the aftershock depths are four kilometers or more beneath the ground surface, it is likely that the aftershocks lie beneath the west-dipping surface of the Eastern Bear Lake fault. The Sulphur Peak earthquake may have occurred on an unmapped and/or buried fault in the Aspen Range.
- ASI4 installed 9/6/2017
- ASI5 installed 9/6/2017
- ID05 installed 9/7/2017
- ID06 installed 9/8/2017
- ID07 installed 9/8/2017
- ID08 installed 9/9/2017
- ID09 installed 9/9/2017
- ID10 installed 9/12/2017
Animation of automatic and reviewed UUSS earthquakes (September 2-11)
- Why do I see more earthquakes on the webicorders than are reported on the USGS website? While the large earthquakes are located and posted to the USGS website quickly, many small earthquakes are not located until time and resources allow. These earthquakes are located outside of the UUSS authoritative area, but UUSS records many stations in the area and generates automatic computer generated earthquake locations for the area. As time and resources allow more earthquakes will be located and posted to the USGS website as well as the ANSS Comcat Catalog.
- Are these earthquakes related to Yellowstone or the Wasatch Fault? This sequence is not related to the June-August Yellowstone earthquake swarm. It is not related to the Wasatch fault, but does provide a good reminder that Utah is earthquake country and we should be prepared.