Sign Up for Our Newsletter


In the Wrong Place at the Right Time
Scientists serendipitously discovered a new technique for measuring the impact of offshore oil spills. Will we be smart enough to make their happy accident a permanent part of our disaster response?

On April 20, 2010—the day that BP’s Deepwater Horizon oil platform exploded 40 miles off the coast of Louisiana—Tom Ryerson, an atmospheric chemist for the National Oceanic and Atmospheric Administration, was only a few days away from taking off on a project to measure emissions from California power and petrochemical plants. He had already spent six years planning his study, capped by the several months it took to collect and load 20 high-tech instruments onto one of NOAA’s P3 turboprop hurricane-hunting planes—where the machines’ disaggregated parts had to be put back together again, he says, “like assembling a ship inside a bottle.”

But before Ryerson and his fellow scientists could ascend into the smokestack haze and get to work, word came down: There had been a change of plans. As important as their California project was, their boss at NOAA explained, it would simply have to wait. Ryerson, his team, and their specialized instruments were needed in the Gulf to help identify and measure all of the different gases that were evaporating off of the huge—and, at the time, still growing—oil slick, and to see if the air was safe for responders and coastal residents to breathe.

They made their first flight over the spill on June 8th. Even more than seven weeks after the pipeline’s burst, the site still “looked like a disaster zone,” Ryerson recalls. “There were flames everywhere, oceans of oil on the surface. It was dark, and hard to see very far. We thought: ‘Boy, this is just a completely unbelievable mess; we’ll never make sense of it.’”

But after they had flown over it again a few times, he says, “We realized, ‘Hey, this is a tractable problem.’ It looked like a point source to us. We knew how to deal with those.” With the degree of precision afforded to them by all of their onboard technology, Ryerson says, both he and the NOAA administrator who had sent him to the Gulf, A.R. Ravishankara, suspected that the team “could say something really powerful about what was happening, and why.” They quickly realized that their technologically tricked-out airplane equipped them to identify and analyze the very nature of the oil slick itself—including where the oil was coming from, how fast it was moving, and where it was likely to spread. Ryerson had effectively flown his plane into the future of oil-spill response.

To understand just how he did this, you first have to understand how we have conventionally handled offshore oil spills in their early stages. The key to containing any spill is in knowing its “flow rate,” a term denoting how much oil is coming out of a source, and at what speed. Flow rate determines how best to cap a well, how much chemical dispersant to apply in order to neutralize a slick, how many tankers one needs to bring in to siphon oil off the surface, how many booms are needed to form a corral, and—importantly—how much oil is estimated to be spreading, unchecked, underwater. Measured in barrels per day (BPD), the flow rate also tells us, in the end, how much actual damage has been done—and, ultimately, how much an oil company should have to pay to clean it up.

The problem: Flow rate is notoriously difficult to measure with any real precision, especially in deep water. Usually pilots will just fly over a spill, eyeball an estimate of how many square miles the slick appears to cover (and how thick the oil appears to be), then divide that figure by the number of days the spill has already taken to grow to that size. The resulting figure is the median of an extremely wide range of potential flow rates.

Their technologically tricked-out airplane equipped them to identify and analyze where the oil was coming from, how fast it was moving, and where it was likely to spread.

But at the time of the Deepwater Horizon blowout, no proven methods existed for measuring the discharge of oil from a pipe that had snapped a full mile below the water’s surface. BP’s initial flow-rate estimate, widely ridiculed, was 1,000 BPD; government estimates, equally ridiculed, were 5,000 BPD. Not until three weeks after the blowout, when BP finally released the first video of the oil and gas stream, did independent scientists get their own chance to offer estimates. Those turned out to be much more in the neighborhood of what would eventually become the final, “official” figures: 62,000 BPD (on the first day of the spill), tapering off to 53,000 BPD (on the last).

In part due to those early, incorrect estimates, various dispersants that had been employed to break up the oil didn’t work properly. Nor did the technique known as “top kill,” which entails pumping mud into the well to stop the leak before capping the well with cement. These failures highlighted the frustrating inability of seemingly anyone to get an accurate flow-rate measurement. Not that people hadn’t been trying. Camera-wielding robots were sent down to capture video of the hydrocarbon plume, as well as the smaller jets shooting out of fissures in the kinked pipe. Ships anchored nearby used sonar in an attempt to calculate the size and shape of these streams. Such efforts took several weeks—from mid-May to early June—and required constant and direct access to the site of the blowout.

By contrast, over the course of just two eight-hour flights made on June 8th and 10th, Ryerson and his team came to realize something that neither they nor any other previous oil-spill response team had ever considered. The particular instruments they were carrying aboard their plane had placed them in a unique position to calculate the flow rate with startling accuracy—in the space of roughly half a day. With the aid of tools like their mass spectrometer, which can instantly analyze the aromatic components of oil, or of another instrument that uses laser beams to measure methane emissions, Ryerson could get real-time readouts of the oil vapors being sucked in by the plane’s vacuum pumps and map the slick’s spatial distribution. Air samples his team collected in special canisters and sent off to a lab immediately after their flights added another level of exquisite chemical detail.

Ryerson’s instruments didn’t detect any atmospheric methane as his team flew over the slick—indicating that the gas was actually dissolving in the water en route to the surface, as were lighter carbons, (like propane and butane) and hydrocarbons (like benzene and toluene). Of the hydrocarbons that did make it to the surface, he knew that their lighter compounds—such as gasoline—would start to evaporate as soon as they hit the air. Because he could measure the rate of that evaporation with such accuracy, Ryerson was able to create a map that had both spatial and temporal significance—a time-stamp revealing the exact moment a particular patch of oil had risen to the surface, and the direction it had taken to get there.

“Immediately,” he says, “we knew something that had been confounding a lot of other folks” who feared the oil might be disseminating and spreading below the water’s surface. The data Ryerson was receiving instead told him that the oil compounds that weren’t dissolving were flowing to the top in more or less of a straight column, and then spreading out. “That makes a fundamental difference in how it gets where it gets,” he says, “and how it affects ecosystems downcurrent and downrange.”

Ryerson had effectively flown his plane into the future of oil-spill response.

Two years later, off the coast of Scotland, the same flyover technique had a chance to prove its worthiness a second time—and once again it was shown to represent a remarkable improvement over other means of calculating flow-rate. All the same, nearly four years after the Deepwater Horizon explosion, and despite the urging of many who can vouch for its efficacy and accuracy, the technique hasn’t been officially adopted by first responders. “We were ready to go—but just by luck,” Ryerson says. “The next time something like this happens, we won’t be able to pull this tool out of the toolbox. And there will be a next time. These things typically happen every 10 or 15 years. But it could happen again tomorrow.”

Had he and his team known they could measure flow rate, and had his plane been dispatched to the BP blowout site as soon as it happened, Ryerson believes, it could have saved massive amounts of time, energy, and money that were ultimately wasted on failed cleanup efforts informed by bad flow-rate estimates. At the moment, the various instruments needed for the technique to work optimally are scattered across the country. The process of collecting them, assembling them on an aircraft, and calibrating them just so takes money and time.

But Ryerson has calculated just how much of both it would take—and by federal disaster-response procurement standards, surely it’s a bargain: With just one year and $5 million, he believes, we could have a new and improved version of his old tricked-out plane gassed up and ready to fly, in a few hours’ time, to the site of our next offshore oil spill. When it wasn’t being used for disaster response, he notes, the plane could be used to take super-accurate measurements of other kinds of emissions—such as those given off by fracking operations.

Along with Ravishankara, his former NOAA boss (now a professor at the University of Colorado), Ryerson has been pressing to make this vision a reality. Of the many challenges faced by disaster-response teams as they tried to cap the well below the Deepwater Horizon, navigating the warm and gentle waters of the Gulf wasn’t among them. “Just imagine this happening in the Arctic,” Ravishankara says. Robert Haddad, an official at NOAA’s Office of Response and Restoration, has imagined that very thing. In the course of routine troubleshooting, he says, his office looks “at the potential for dealing with spills in [places like] Alaska’s Chuckchi Sea. How are you going to get to that?” If we’re going to continue to allow drilling in hard-to-reach places, he says, we need to be honest about the possibility “that something might happen there.”

Should it happen there or anywhere, Ryerson champions the flow-rate flyover “as a beautiful technique that’s been completely proven now” in two different instances. “It could be a game-changer in the next one,” he adds. “But we’re still not quite there yet.”

Like this article? Donate to NRDC to support OnEarth's groundbreaking nonprofit journalism.

image of Kim Tingley
Kim Tingley is a regular contributor to OnEarth and the New York Times Magazine. She has an MFA in nonfiction writing from Columbia University and in 2012 received a Rona Jaffe Foundation Writer’s Award, given annually to six female writers who demonstrate excellence and promise in the early stages of their careers. MORE STORIES ➔