Nearby seismograms show that the 7 November 2016 (6 November local time), M5 earthquake near Cushing, Oklahoma is two earthquakes of about M4.8 occurring about 1.4s apart. Such “rapid-fire” occurrence of two earthquakes may have contributed to the damage in Cushing; in general, the hazard from such multiple earthquakes may be underestimated. The occurrence of two events may also explain difficulties with magnitude and mechanism determination for this earthquake.
Identification of two events
The 5s long seismograms below are from (upper) station OK914 about 3km SW of Cushing and (lower) OK915 3km S of Cushing. Both seismograms show two distinct sets of P and S energy arrivals for the earthquake (P1 and P2 on blue, vertical trace; S1 and S2 on yellow and magenta, horizontal traces).
The 5s long seismograms below are from (upper) station OK031 about 8km SW of Cushing and (lower) OK034 about 6 km NE of Cushing. These two seismograms also show two sets of P and S energy arrivals for this earthquake (P1 and P2 on blue, vertical trace; S1 and S2 on white, horizontal traces).
This duplication of P and S arrival energy strongly suggests that there were two distinct earthquake events, about 1-2s apart.
Indeed, the relationships between the arrivals – P1 at OK031 being earlier than P1 at OK034, while the opposite is the case for P2, suggest that the second event was closer to OK034 than to OK031 and was to the NE of the epicenter of the first event:
Seismograms from a more distant station, OK033 about 14km NE of Cushing, also show the two sets of P and S energy arrivals (upper, ground acceleration; lower, ground velocity):
Formal event locations
The relative locations of two events can be formally determined using the 20 P and S arrival times from the 5 stations:
The first event is at the position of the Oklahoma Geological Survey epicenter for this earthquake, while the epicenter of the second event falls about 2.5km ENE of the first event, near the NW limit of Cushing, and along the trend of aftershocks and background seismicity determined by the Oklahoma Geological Survey:
The two events have similar waveform amplitudes and thus are likely to have similar magnitudes. Assuming the overall M5 magnitude represents the sum of energy for the two events, then the magnitude of both would be about 4.8.
Increased shaking duration and hazard
The strong, S wave shaking from each event alone has a duration of less than 1s. But the combined duration of strong shaking from both events, plus a 1.4 delay between events, gives a total duration of strong shaking of about 2.5s:
An increases duration of strong shaking will, in general, produce more damage. This effect, plus the location of the events very close to and nearly below Cushing, may account for “substantial damage” to structures in Cushing.
The total duration of shaking for the two M4.8 events is likely longer than for a single, M5 event, in which case the hazard from multiple, “rapid-fire”, smaller events can be larger than that from a single, slightly larger event. Such hazard could be underestimated when only the expected size of a single, largest event is used to estimate hazard.
Two events pose difficulties for magnitude and mechanism determination
Most magnitude and mechanism determination algorithms assume a single event. So the occurrence of two events close in time with overlapping waveforms can cause problems for these algorithms. This may be the case for magnitude estimates for this earthquake, which range from 4.5 to 5.3 (USGS), and for the USGS W-phase moment-tensor mechanism which shows large uncertainty.
Two events: “doublet”, or “stopping phase”, or…
The two events identified from the seismograms are 1.4s apart and separated by about 2.5km. The expected subsurface, fault rupture length for an M4.5-5 earthquake is about 2-3km (Wells & Coppersmith, 1994), so the second event may have occurred near the end of rupture of the first event. Since the two events have similar size, they could be considered together as “doublet” events, if the second event initiated after termination of rupture of the first event. Alternatively, the second event could be a large burst of energy from increased rupture at the end of the first event, or, conversely, a “stopping phase” from very abrupt ending of rupture of the first event. There are other possible rupture scenarios to explain the two events – more detailed analysis and modelling of the seismogram waveforms should shed some light on these issues.
- Earthquake parameters: USGS Earthquake Hazards Program Oklahoma earthquake event page
- Seismogram waveforms: Strong-motion accelerometer data from US Geological Survey Networks GS and NQ. The facilities of IRIS Data Services, and specifically the IRIS Data Management Center, were used for access to waveforms, related metadata, and/or derived products used in this study. IRIS Data Services are funded through the Seismological Facilities for the Advancement of Geoscience and EarthScope (SAGE) Proposal of the National Science Foundation under Cooperative Agreement EAR-1261681
- Seismogram analysis and visualization: SeisGramWeb and SeisGram2K
- Earthquake location: NonLinLoc – probabilistic, global-search earthquake location in 3D media
- Wells, D.L. and Coppersmith, K.J., 1994. New empirical relationships among magnitude, rupture length, rupture width, rupture area, and surface displacement,Bulletin of the Seismological Society of America vol. 84 no. 4 974-1002
- 2016/11/09: Added stations OK914 and OK915 and updated locations and several figures. Added “Two events: “doublet”, or “stopping phase”, or…” section.