The 2014 Weld County, Colorado, Earthquakes: A developing case of induced seismicity?

Friday, 19 December 2014
William L Yeck, Anne F Sheehan, Matthew Weingarten and Jenny Nakai, University of Colorado, Boulder, CO, United States
On June 1st 2014 (UTC), a M 3.2 earthquake occurred in Weld County, Colorado. Weld County is largely aseismic and this event was unexpected. There were no events in the ANSS Comprehensive Catalogue within 50 km of the earthquake. Weld County is the center of active oil and gas development, including many high-volume Class II wastewater injection wells. Injection wells have been linked to an increasing number of earthquakes throughout the US in recent years. Due to the lack of background seismicity in the area and the proximity of the earthquake to several injection wells, the University of Colorado requested seismometers from IRIS PASSCAL’s Rapid Array Mobilization Program in order to study further seismicity. Seismic stations were deployed within 3 days of the June 1st event.

We report on our ongoing findings from this deployment. To date, we have located 89 earthquakes and have detected over 600. These numbers continue to grow as we collect and analyze further data. Earthquake magnitudes remain small with only three earthquakes above M 2.0 recorded by our network, the largest of which was an M 2.6 earthquake on June 23rd 2014. Earthquakes locate in a small cluster (~2 km radius) centered near a high-volume injection well. This well operated at injection rates greater than 300,000 barrels/month since August 2013 and injects at a depth near the sediment-basement contact. Prior to our local deployment, the closest seismic station to the June 1st event was > 100 km away and therefore the evolution of seismicity prior the June 1st earthquake is poorly constrained. In order to better understand the temporal evolution of seismicity, we use match-filtering on data from these distant stations and find the earliest matching event on November 11th 2013. Due to the strong spatial and temporal correlation between these events and injection operations, we find it likely that these earthquakes are induced.

In response to the ongoing seismicity near the well, the Colorado Oil and Gas Conservation Commission (COGCC) ordered the high-volume well shutdown for 20 days. The well has since restarted injection at lower rates, with the potential to increase injection rates over time. The shutdown and resumption of smaller injection volumes provides the unique opportunity to directly evaluate the seismicity’s relationship to well operations.