Reducing wastewater disposal from oil and gas extraction could significantly decrease induced or human-caused earthquake activity in Oklahoma by the end of the year, returning it to natural levels within just a few years, a new modeling study in the 2 December issue of the journal Science Advances estimates.
Already, as predicted by the model, a mandate issued earlier this year calling for a reduction in wastewater disposal has proven effective in curbing Oklahoma's seismic activity.
Earthquake activity in north-central Oklahoma overlaps with saltwater injection sites produced by oil and gas extraction. | Langenbruch and Zoback (2016), Science Advances
Earthquake activity in Oklahoma has increased significantly in the past few years, following the injection of increasing amounts of wastewater from oil and gas extraction back into the ground. Between 1979 and 2009, the natural seismic activity in Oklahoma was relatively constant, producing about one magnitude 3 or greater earthquake each year. In 2015, however, following several years of increased saltwater injection, which is known to trigger the release of accumulated rock pore stress underground, approximately 900 magnitude 3 or greater earthquakes occurred in the north-central region of the state.
In response, in February and March 2016, regulators called for a 40% reduction compared to 2014 levels in the volume of saltwater being injected into seismically active areas of Oklahoma. Oil and gas extraction operators were requested to meet the reduced injection rate by the end of May 2016.
Cornelius Langenbruch and his colleague Mark Zoback at Stanford's School of Earth, Energy & Environmental Sciences, co-authors of the Science Advances paper, sought to understand how the mandated reduction in injection rates would affect quake rates in the state.
"We are in contact with the regulators in Oklahoma," Langenbruch said, "and we know they are interested in science-based decision making tools."
Langenbruch and Zoback built a statistical model linking changes in saltwater injection rates to seismicity rates, calibrating its predictions against data from large-volume injection wells in Oklahoma and thousands of injection-related earthquakes in the state.
Ultimately, the researchers' model showed that the mandated reduction would be an effective step toward mitigating the seismic hazard of induced earthquakes — significantly reducing quake rates by the end of 2016 and returning them to historic levels within just a few years. For 2017, for example, Langenbruch said their model predicts about 250 magnitude 3 and larger earthquakes compared to approximately 900 in 2015.
The study introduces a model for assessing how a combination of past seismicity and future wastewater injection volumes impact future earthquake probabilities — a critical resource as increases in injection-related seismic events are difficult to forecast.
"There are no past approaches for assessing how wastewater injection volumes in Oklahoma impact future earthquake probabilities," said Langenbruch. "Our model is the first seismic hazard model [to be able to do this]."
"It has been denied for many years that the earthquakes in Oklahoma are caused by injection of wastewater. Finally, the industry, regulators and researchers are working together to mitigate the seismic hazard."
Regulators will be happy to hear that the mandated volume reduction is resulting in a declining number of induced earthquakes, according to Langenbruch. "We already can observe a significant decrease of the number of widely felt earthquakes," he said. "That's the good news."
The bad news, he explained, is that the earthquake rates in seismically active areas are still comparatively high and will take several years to get back to the natural background rates.
One reason for this staggered effect is the delay between injection of wastewater into the ground, where it causes increases in rock pore pressure that spread over time, and occurrence of an earthquake, which may not be triggered until the creeping pressure meets a vulnerable fault. When large volumes of saltwater are injected into a highly permeable rock formation in Oklahoma called the Arbuckle group, for example, pressure builds slowly over time.
"The pressure increase caused by injection needs several months to travel the distance from the injection wells to the critically stressed preexisting faults in the crystalline basement," Langenbruch said. "It is a very slow process and the pressure increase caused by the massive injection during the past years will keep on propagating through the underground for years. Where it finds critically stressed faults, earthquakes will be triggered. This is why the risk of potentially-damaging earthquakes will remain high for some more years."
The model developed here can also be adjusted for additional regions of interest, where wastewater disposal is happening at high levels.
"It has been denied for many years that the earthquakes in Oklahoma are caused by injection of wastewater," Langenbruch said. "Finally, the industry, regulators and researchers are working together to mitigate the seismic hazard."