For the past several years, every time a single barrel of oil gets pumped up from deep under the Earth’s surface in Oklahoma, 10-15 barrels of salty, often contaminated wastewater gets pumped up with it. After being separated from the desired oil and gas, the water is injected back into the Earth, far enough down that the oil operators hope the watery byproduct won’t contaminate groundwater supplies in the region. But, much like many toddlers, this part of the country hates injections.
A few years ago research decisively linked wastewater injection to increased earthquake activity in the area, where the number of significant earthquakes has been increasing steadily year after year. In 2013, there were 109 earthquakes with a magnitude of three or higher. Earthquakes of magnitude three feature shaking that is strong enough to be felt, but not usually enough to cause major damage. In 2014, that number went up to 585. In 2015 it was up to a whopping 906, the equivalent of 2 or 3 earthquakes per day.
But there’s hope: A study published recently in Science Advances finds that just as quickly as those numbers climbed, they can also fall.
Unfortunately, it will take a long time for earthquake activity to drop back to normal, pre-injection levels. Large earthquakes – like the one that shook Cushing, OK last month or the 5.8 magnitude earthquake that shook Pawnee, Oklahoma in September – are still going to occur as previously injected water makes its way through the Earth, a process that could go on for another 5-10 years.
“I think it’s really important to understand that this triggering process cannot be switched off. If the probability of a damaging earthquake is decreasing right now, it doesn’t mean that it’s low. It’s still high, and there might be damaging earthquakes in Oklahoma over the next few years,” says Cornelius Langenbruch, the lead author of the study. “You really have to be patient, and be aware that this is not the end of induced earthquakes. It will take some time before it gets back down to the background rate.”
Wastewater injection is considered to be a different process than hydraulic fracturing (also known as fracking), where water mixed with sand and chemicals is injected deep into the ground with the express purpose of breaking up rocks to release oil or natural gas reserves. Wastewater injection is (for the most part) as simple as sending wastewater into the ground and leaving it there. Fracking has also been linked to earthquakes, most recently in Canada. But in Oklahoma, the earthquakes are being induced by wastewater injection.
Wastewater injection rates in Oklahoma had been going down at around the time that the regulations were introduced, but not out of concern for the planet. “Most of the volume reduction is not driven by mandated volume reduction, it’s driven by the market,” Langenbruch says.
With oil prices sliding in recent years, production at wells in the area was in decline, and with it, wastewater injection was also scaled back. The regulations ensure that even if oil prices and production rise again, there will be limits on how much wastewater can be injected into the ground.
The exact conclusions of this study are localized, and dependent on many factors: the rate of injection, the types of rock that the wastewater is injected into, and the general geology of the area. All of these factors (and many others) feed into the size and frequency of the earthquakes. But by changing the variables, this mathematical model could be applied in other situations around the world. In some ways, it already has been. The mathematical model that Langenbruch and his colleagues used in their paper was adapted from a similar equation that looked at injection rates in geothermal and hydraulic fracturing operations in Europe.
Langenbruch and colleagues hope to further refine the model with direct measurements at wells in the region, which are now available for scientific research. “Right now, we don’t know exactly what is the pressure increase caused by the injection of the fluid, because there are no accurate measurements of the pressure in the formation where the wastewater disposal happens,” Langenbruch says. But by measuring the exact pressure at different points underground, the researchers should be able to get a better sense of how the injected water is behaving over time, and how that relates to the earthquake hazards in the state. “If we can see where the pressure is going up and where it is decreasing, this might allow for a more localized hazard model,” Langenbruch says.
The current earthquake hazard model shows the likelihood of earthquakes of different magnitudes striking an area during a specific time period. But until now, it relied on records of previous earthquakes to draw up that assessment. Now, Langenbruch says, researchers can also include information about injection rates into the hazards model, resulting in a more accurate risk assessment.