USA: Forecasting seismicity from wastewater disposal in Oklahoma
Mandated wastewater injection reductions in effect since 2016 are inadequate for preventing future, large-magnitude earthquakes in the state, according to a new induced seismicity model.
By Terri Cook
Earthquake activity in Oklahoma has risen sharply since 2012. Numerous studies have attributed this uptick in seismicity to the disposal of wastewater underground, much of which is associated with gas and oil drilling operations. As operators inject large volumes of brine into the region’s thick Arbuckle formation, the fluid raises the pressure in the surrounding rock and soil high enough to overcome frictional forces—and causes faults in the underlying crystalline basement rocks to slip.
In 2016, in response to the increased seismicity, Oklahoma regulators mandated a 40% reduction in the volume of wastewater that could be injected in two areas of northwestern Oklahoma that had experienced large numbers of tremors. Although earthquake activity in the state has since declined, new research by Dempsey and Riffault suggests these reductions may not be enough to prevent future magnitude 5 events.
To better understand how induced seismicity responds to decreasing wastewater injections, the team developed a series of models that incorporate pressurization from wastewater injection coupled with models of triggered seismicity. The researchers then adapted one of these models to western Oklahoma and used it to forecast earthquakes in the region under several injection rate scenarios, all of which are permissible under the 2016 mandate.
The findings indicate that stabilizing the injection rate at September 2018 levels, which are 25% lower than the mandated maximum, would result in an initial decline in earthquakes of magnitude 3 or greater. But by 2025, following the establishment of a new equilibrium seismicity rate, the number of tremors in the region would increase to between 1.4 and 4.1 events per month. Should wastewater injections increase to the maximum allowed, the researchers predict that the number of earthquakes of this size could climb to as many as 15 per month.
The results also indicate a high likelihood that a magnitude 5 or greater earthquake could occur in Oklahoma prior to 2025, with a 14%–46% chance at 2018 injection rates and an 81%–90% chance at the mandated maximum injection rates. On the other hand, the authors suggest that further cuts—by 50% of the September 2018 injection rates—could cause seismicity to disappear from the region by 2021.
This conclusion, which differs significantly from the results of earlier research that utilized a different type of failure model, suggests the mandated reductions are inadequate for preventing large-magnitude earthquakes. The contrasting results highlight the need to improve scientific understanding of the underlying causes of injection-induced seismicity on local scales.