The Boom (41 page)

Over the next couple months, Flatirons voluntarily did something unusual. It spent liberally to figure out what had gone wrong with the well—and left behind a valuable, detailed record. Flatirons hired the Schlumberger oil-field service company to run an ultrasonic image log, a tool that uses sound pulses well above the range of the human ear to find gaps between the pipes and cement. The test results were not encouraging. At points, the “cement was so spotty and unbonded,” there could be multiple places where gas was flowing into the man-made annulus, according to Flatirons. “We saw gas-filled channels in the cement, but you can’t really tell how large they are,” Jeff Jones, a managing director at Flatirons, told me in an interview. Like blowing air through a straw into a thick milkshake, the gas had likely entered into the cement slurry before it hardened and created long channels that extended for hundreds or maybe thousands of feet. Flatirons collected a gas sample at the surface and sent it off for analysis, looking for isotopic fingerprints to determine if it had come from the Marcellus or from a shallower zone. The analysis showed it was coming from gassy sands only a couple thousand feet deep.

Flatirons then ran a noise log. Developed by Exxon in the 1970s at the same time that Cooke developed the RDT, a noise log is essentially a powerful microphone that listens to the well. It is so sensitive that it can hear a leak and help determine if the leak is a liquid, gas, or both. Flatirons worried that a lot of gas was moving up the well, possibly into the aquifer, and not making it to the surface. “The noise log is one of those things that you’d kind of rather not know, but we decided we needed to know,” explained Jones. “If there was any way we were possibly moving gas into the aquifer, we needed to do something.” But since the leaks were small, the gas couldn’t be heard. “Most of the tools aren’t set up to find a small quantity of gas,” he explained.

The company also sent a sample of the cement to an independent lab for testing. The results weren’t good. To ensure that cement works as advertised, the industry has established quality guidelines. One requirement is no more than twenty teaspoons of fluid be lost into the rocks from a set amount of cement in thirty minutes. If too much liquid escapes into the rock, the cement can become too dense to pump and won’t go where it is needed. In the laboratory testing, the cement failed spectacularly. It lost over a hundred teaspoons. Flatirons determined that the cement had grown too chunky to spread evenly and provide a good barrier. Instead of a solid seal, it had left behind channels. But there was some good news. As these tests were run, the volume of gas leaking to the surface decreased, and testing the water aquifer didn’t turn up any evidence of contamination.

Flatirons’s investigation left me unsettled. The company used ultrasonic logs, noise logs, and even gamma ray logs to pinpoint the leak, but had failed to figure out where the gas was entering the wellbore. These tests cost nearly $200,000, according to Flatirons, and weeks of detailed work. If Flatirons had found a leak, it could have used perforating guns to blast small holes in the pipe and squeeze in more cement. Later that summer, in a presentation to the state, Flatirons argued against this remedy. It might seal the leaks, or maybe not, but it could also make it harder to frack the well. And it would cut into its profits. The well might have to be scrapped. The state decided that since the leak appeared to be going away on its own, Flatirons wouldn’t have to try to fix the well.

Flatirons isn’t the only company in the Marcellus to have problems with leaking gas. “Many other operators in Pennsylvania have been confronted with these problems,” Flatirons engineers said in a paper they wrote about the well. The state convened a group of regulators and companies to study the problem. One issue the group has taken up is what is an acceptable level of leakage. The Flatirons well was leaking 270 cubic feet daily. The average US home uses 200 cubic feet daily. Flatirons pointed out state regulations for underground storage caverns that store gas for peak wintertime usage were allowed to leak up to 5,000 cubic feet a day of gas without any repercussions. The working group decided this was a good starting point. Leaks happen. Fixing them is hard.

There was one more tool that Flatirons could have tried: Schlumberger’s isolation scanner. This is the superdeluxe tool that BP had ordered, and then skipped, aboard the Deepwater Horizon. Schlumberger introduced this tool in 2006 to improve on older tools that it says suffered from limitations and generated ambiguous results. Flatirons asked Schlumberger about using an isolation scanner, said Jones, the Flatiron executive, but was told none were available in Pennsylvania. To bring one in would quadruple the cost of running it down the well.

In Pennsylvania, the drilling industry doesn’t want to be forced to use this tool. In January 2011 the state issued a draft of new rules on testing a well’s mechanical integrity. The state reserved the right to require the driller use an isolation scanner. The Marcellus Shale Coalition, an industry group that lobbies on behalf of the state’s drillers, requested the draft language be changed. Specifically, it wanted to replace “isolation scanner” with “ultrasonic logging tools.” A month later, these less precise and less expensive ultrasonic logging tools proved inadequate in finding the Flatirons’s well leak.

“If I had a more precise tool, maybe it would have showed us where there was a minor channel,” said Jones. This might have changed his mind about ordering remedial cement. “Maybe I would have a different opinion if I had other tools available,” he added.

The well was fracked, after a six-month delay, and the amount of gas leaking to the surface continued to decline. It is too small an amount to even flare off. “It is now down to the amount of gas that is probably equivalent to one cow,” he said. Methane emission from cows, through flatulence and belching, varies based on diet, whether the cow is lactating, and other factors. Scientists have estimated that a cow emits between eight and sixteen cubic feet of methane a day.

After the magic trick, Gearhart suggests that we adjourn to the nearby McDonald’s for lunch. After eating a small hamburger, Gearhart reaches into his wallet, pulls out a dollar bill, and then counts out 72 cents in change. It’s enough for two vanilla ice-cream cones. He offers to buy one for Cooke’s grandson, if the young man would stand in line to buy them both. He agreed.

Afterward, we head back to Gearhart’s office for a quick tour of the manufacturing side of his business. He introduces us to various workers soldering circuit boards onto small tools and shows off a machine built to calibrate gyroscopes. He makes a magnetic survey tool called the Geo-Shot. It goes down a newly drilled well and reports on the well’s direction. States require independent verification of the underground reach of a well. Drilling under a neighbor’s property without permission is a century-old problem. It amounts to stealing oil and gas. This practice was dramatized memorably in the 2007 movie
There Will Be Blood
. After years of holding out, a penniless landowner finally offers to lease some land to the deranged oilman portrayed by Daniel Day-Lewis. But the oilman says that he has already drained the land illegally. “I drink your milk shake!” he taunted.

The Geo-Shot prevents illicit milk shake drinking. Texas has had several scandals involving wells drilled at a slant to steal neighboring oil, and since 1949, the state has required the type of directional survey that Gearhart’s tool produces.

Making them, Gearhart says, is a good and steady business. At the end of the tour, Cooke takes my elbow and leans in close. “His data is required by regulators,” he says. “Someday, if proof was required that a well will not leak, we could supply that proof. It’s a long way from being required.”

In the months afterward, the lack of interest or demand by the drilling industry in a better and cheaper cement-leak detection tool created headaches for Cooke and Smiley’s new Well Integrity Technology Company. State regulators couldn’t insist it be used because there were no tools available on the market. Until Cooke and Smiley had a working RDT, they couldn’t find a drilling company to test it out.

When the tool was finally located in Bakersfield—the one in the attic turned out to have been loaned years earlier and lost—Smiley hoped that Gearhart could refurbish it. This would let them run it down wells to test out if it was better than a conventional cement bond log. But in early 2013, Gearhart informed him that it would cost $150,000 to design and manufacture a prototype for testing. This was a steep price, said Smiley, because margins are pretty tight in this part of the oil-field service market. And Cooke would have to come up with the money for the prototype himself. If the prototype was lost in the test well, a not-infrequent occurrence, he would have to pay more for another.

Meanwhile, Cooke had thought up a new polymer to be used to frack wells. Compared to standard fracking fluids, it required less water and was biodegradable, so it was more environmentally benign. Promising to lower the cost of fracking a well, this technology had drawn interest from the industry and secured a government research grant.

The RDT tool was going nowhere. As of this writing, Smiley hopes to find a partner—a drilling or oil-field service company—willing to pay for the prototype and run some tests. But interest has been practically nonexistent.

“We’re still holding out for some company who is interested in seeing if this works,” Smiley said. “We don’t have enough time to be knocking on everyone’s door.”

Would a modern RDT improve the industry’s wells? It is possible. As with any invention, it needs a long period of tests and real-world use to determine its usefulness. But Cooke’s experience shows that the industry isn’t clamoring for better tools, and neither are regulators or landowners. At my urging, my father asked Chesapeake what tests it had used on the well at the Farm. His phone calls were bounced between offices and he never got an answer.

13

PANDORA’S FRACK

On December 1, 2012, Dallas opened a new nature and science museum on the edge of downtown. The building is a light gray cube with an irregularly layered concrete façade that looks like stacked rock strata. From a distance, it appears as if a block of earth has risen out of the ground, shed its soil, and dried out. Entering the building, visitors head toward an escalator that carries them up into the ceiling and then through a cutback in a compressed space surrounded by poured cement walls. The path leads to a fifty-four-foot escalator that climbs past a long diagonal bank of windows. Near the top of the cube, the escalator deposits visitors in a sunlit balcony from which the downtown Dallas skyline is visible.

Rising up, surrounded by cement, a passage through rocks—this passage felt strangely familiar when I visited the museum. A few hours later, it struck me that I had followed the path of a hydrocarbon molecule coming up through the earth, traveling in a man-made opening in the rocks, while hemmed in by concrete. Deposited at the balcony, I had exited the well and entered the modern world with the view of skyscrapers and an eight-lane freeway. Maybe I had too much fracking on the brain. I contacted the architect who designed the Perot Museum of Nature and Science, Pritzker Architecture Prize–winner Thom Mayne. An email from Arne Emerson, an associate of the architect, confirmed that the façade was inspired by underground rock layers, but then suggested that the rest of my impression was the product of an active imagination. “Your interpretation,” he wrote, “is one of the best things about experiencing a great piece of architecture—it evokes a response and is also both personal and subjective.”