After the shaking from the 5.8 magnitude earthquake near Pawnee, Okla., subsided Sept. 3, Tandis Bidgoli noticed something unusual.
“I was very concerned,” said Bidgoli, a geologist with the Kansas Geological Survey, “because it didn’t appear there were any foreshocks to that event.”
In other large quakes, such as the 5.7 magnitude near Prague, Okla., in 2011, there had been other, smaller earthquakes before the big one. Scientists interpreted these earlier foreshocks as signs of a big one to follow.
But there were no foreshocks in Pawnee, Bidgoli said. The fault line was new. So she worried: What if the 5.8 quake, the largest recorded earthquake in the region ever, was itself just a foreshock? What if the biggest quake ever was just hinting at an even bigger quake to follow?
“Holy cow,” she said she thought. “If this is in Kansas, this is a game-changer for all of us.”
Holy cow. If this is in Kansas, this is a game-changer for all of us.
Tandis Bidgoli, geologist
A new theory
Bidgoli has been working on building models and maps that could change how scientists look at the earthquakes in Oklahoma and Kansas, which she will soon present to the National Academy of Sciences.
Normally scientists look to find wells within 5 kilometers, or about 3.1 miles, of where an earthquake occurred and figure out which wells increased wastewater before the earthquake that might have caused it, she said.
(Wastewater is a byproduct of most oil production in Kansas and Oklahoma. As fracking has increased, the disposal of wastewater deep under the earth also has increased.)
A recent study by the United States Geological Survey concluded that one or two wells that dispose of wastewater caused the largest recorded earthquake in Kansas in 2014. Bidgoli thinks the blame for earthquakes like this may need to be more widely spread.
The traditional model of looking at wells within 5 kilometers doesn’t explain the huge surge in earthquakes since 2009 in Oklahoma and 2014 in Kansas, Bidgoli said.
For instance, she said, 85 percent of the recent increase in oil production in Kansas occurred in Harper County. But much of the initial earthquake activity occurred in Sumner County, farther away than would be expected by the traditional model.
She thinks earthquakes could be caused by wells 10 miles away or even farther, she said. That’s because so much wastewater is being injected into the ground in certain places that the rock layer has become saturated. When new water is added, it can increase pressure in one spot that is then released tens of miles away.
“A very small pressure change could go a very long distance,” Bidgoli said.
In Harper and Sumner counties the number of wastewater disposal wells has more than doubled in the past few years to about 170, Bidgoli said, and the average volume of water injected from each of those wells has increased. So the total increase in volume has been exponential.
“The injections have produced a pressure field that has extended across the region, which is a potential reason we’re seeing seismicity on such a broad scale,” said Lynne Watney, one of Bidgoli’s colleagues at the Kansas Geological Survey. “It’s not individual wells, but collectively they’re leading to the heightened potential for seismicity.”
And the amount of water injected in Oklahoma far surpasses that of Kansas. The water from Oklahoma could lead to pressure buildups miles across the border, so that Kansas could continue to be susceptible to earthquakes in Oklahoma.
“We can cut back here, but is that going to put a dent in the problem?” Bidgoli asked.
Her model would make it more challenging to pinpoint blame for any one earthquake. If only the nearby wells could be responsible for an earthquake, identifying the cause becomes relatively easy. But if a quake could be caused by a well 15 miles away, identifying a cause and establishing legal blame in court isn’t so simple.
“You might have a well that is the straw that broke the camel’s back, so it may get blamed when a lot of the pressure may be generated far away from the actual earthquake event,” Bidgoli said.
If her studies and subsequent studies prove this model correct, then regulatory agencies may need to change their approach to managing risk.
After a large earthquake, the Oklahoma Corporation Commission usually restricts wastewater disposal in an area that surrounds the epicenter of the quake. It shuts down or reduces the volume of wells within that area. This makes sense according to the established model but may not be sufficient according to Bidgoli’s model.
In August, the Kansas Corporation Commission voted to expand its protected area that is now many times larger than what it was in 2015. It stretches from Goddard in the northeast toward Kiowa to the southwest, and from Murdock in the northwest down to the Oklahoma border south of Wellington.
Injections of wastewater are now limited in an area that extends about 2,000 square miles and touches the outskirts of southwestern Sedgwick County.
More, better data
In order to prevent earthquakes based on water pressure, scientists need more and better data than they currently have, several geologists in Oklahoma and Kansas said.
The Kansas Geological Survey currently has only one well near Wellington where it measures underground water pressure, and the Oklahoma Geological Survey has eight. It is expensive to drill more wells for study and legally difficult to take over private existing wells.
Many of the 17,000 wells in Kansas shut down for routine maintenance, anyway, so scientists could get in for a few days to study the underground pressure without causing much disruption if the well owners will let them, Watney said.
Geologists also need to know where the old faults are. Faults are like an old break in a piece of wood that has been glued back together: Under ordinary stress, the glue holds. But when it faces extra stress, like what is happening with increased water pressure, the glue can snap.
These breaks might have happened over a much longer period of time in the past, but now the process is in fast forward.
It’s like you sped up the tape recorder.
Lynne Watney, geologist
“It’s like you sped up the tape recorder,” Watney said.
“We have these faults in the subsurface, primed, ready to have an earthquake,” Bidgoli said. “They have been accumulating stress for millions of years, and just need a very small change in pressure to trigger that seismicity.”
We have these faults in the subsurface, primed, ready to have an earthquake.
Tandis Bidgoli, geologist
Scientists don’t know where many of those faults are, or more critically, how big they might be. The bigger the fault is, the more energy that is released when the earth moves, and the more potentially dangerous the shaking.
Bidgoli has recently submitted a grant to study the Nemaha Ridge, the underground mountain range that passes beneath Wichita, between Omaha and Oklahoma. Scientists still don’t know that much about the Nemaha or the Humboldt Fault Zone, she said.
Oil and gas companies likely already have maps of some of these unknown faults. The bigger companies use 3-D imaging to help them identify good spots to drill. These 3-D maps also can identify fault lines. But the maps are proprietary.
Companies are reluctant to share for free what they have spent so much money on, Bidgoli said. They’re also afraid of lawsuits.
“There is a little bit of fear about liability,” Bidgoli said. “If you know there is a fault, and it is near your disposal wells, and let’s say there are earthquakes happening, do you become liable?”
Bidgoli avoids alarmist language about the threats Kansas faces. There is a lot of uncertainty about how large a quake could occur here with the information scientists currently have, she said.
Scientists know that the New Madrid fault line on the eastern side of Missouri has produced massive earthquakes in the past, Bidgoli said, because the geology of the rivers left a record of the past shaking.
Kansas and Oklahoma don’t have that same geologic record. It could be because the area hasn’t gone through earthquakes that intense, Bidgoli said. Or it could be that the earth just didn’t leave clues as obvious, and it hasn’t been as well studied.
“I don’t think these induced earthquakes are going to go away anytime soon,” she said. “After the next oil boom, who knows what is going to happen at that point? It’s all very, very concerning.”
Bidgoli is as worried about how unprepared Kansas and Oklahoma are as she is about the uncertainty of how big an earthquake could strike.
She grew up in California, and people there would never put a big bookcase or heavy painting above a bed without anchoring it to the wall, she said.
“I worry about cities like Wichita and Oklahoma City,” she said. “I don’t think they have the kinds of structural enhancements that San Francisco might have. Even in a city not as big, it might be quite damaging. The populace is just not prepared in the same way as people who have encountered frequent earthquakes.”
The day after the Pawnee earthquake she was invited to speak to an introductory geology class.
“The first thing I told the students was to be prepared,” she said. “Have a plan and take those heavy things down above your bed.”