The April 22, 2017 M 3.8 Earthquake Sequence near Rangely, Colorado

On April 22, 2017, a magnitude 3.8 earthquake occurred approximately 4 km northwest of Rangely, Colorado at 11:48 AM local time (05:48 PM UTC).  There were 15 felt reports from the town of Rangely, CO.  Two aftershocks, approximately 1 km NNE of the mainshock, were located by UUSS.  The first aftershock (ML 2.6) occurred on April 27 at 03:11 AM local time (09:11 AM UTC), and the second aftershock (ML 3.3) occurred on May 3 at 01:42 AM local time (07:42 AM UTC). Based on the moment tensor solution for the mainshock this was a predominantly strike-slip earthquake on steeply dipping planes with the strike either northwest or northeast.  From the distribution of the aftershock locations, we tentatively favor the northeast striking plane.  Eighteen earthquakes within 20 km of the mainshock , with magnitude greater than 2.0, have been catalogued since 1962.  The largest historical earthquake (ML 4.6, March 20, 1995) was located 2.3 km NE of the 2017 mainshock.

The Rangely area was one of the first focus sites for the study of fluid-induced earthquakes.  Some of the first documented induced earthquakes occurred near Rangely in the 1960s and 1970s.  During this time water-flood expansion was being used for secondary oil recovery.  It was a good place to test the correlation between fluid injection and seismic events with a controlled experiment (Rayleigh et al., 1976), and the experiment showed a direct link.  The seismicity during the experiment occurred on a ENE-WSW trending plane.  This is rotated from the current seismicity, but the locations of the seismic events have also migrated through time.  Water based fluid injection ended in 1983; since 1986 injection of CO2 has been used for secondary oil recovery.

Given the proximity of the recent seismicity to the Rangely Oil Field, it is fair to ask if the recent sequence is also induced.  Analysis of this sequence is ongoing, but initial work includes the following results.  An STA/LTA detector (detection threshold 3.5) was run across continuous waveforms from the two nearest stations (O20A and RDMU) for the time period April 22, 2017–May 04, 2017.  Requiring simultaneous detections on both stations, in order to reduce the number of false detections, resulted in one new detected event that occurred on May 3.  Using cross-correlation, we found similar waveforms (CC > 0.5) from the four events (mainshock, two aftershocks, and the new detected event) recorded at station O20A, suggesting possible common source properties.

The lack of close seismic stations makes it difficult to clearly associate these seismic events with oil production efforts.