Where Did Deepwater Horizon Oil Go?

As chemists studying the Deepwater Horizon disaster, we get this question all the time from family, friends, colleagues, policymakers, and the general public: What happened to the 200 million gallons of crude oil that were released into the Gulf of Mexico over 87 days in the spring of 2010?

What everyone wants is nothing less than a full accounting of the oil that spewed from the damaged Macondo well 50 miles off the coast of Louisiana at an unprecedented depth of 5,000 feet beneath the sea surface. In theory, it’s not unlike tracking your bank account. We are attempting to match the amount of oil that went into the ocean with the amount that subsequently went out of it. But it is much harder than balancing revenue in from your paycheck against bills paid and cash withdrawn from ATMs.

At the height of the disaster, we advised a public that craved a fast, definitive answer that it would take several years to balance the Deepwater Horizon oil budget. Five years later, we’ve made some headway.

Keep in mind that we’re trying to track 30,000,000,000,000,000,000,000,000,000,000,000 carbon atoms (and twice that number of hydrogen atoms) in a hostile, ever-moving environment. The oil traveled in many directions by different means. Some floated to the ocean surface, where it went into the air or was driven by winds and currents into coastal marshes, beaches, and islands. Some oil became entrained into watery layers that circulated within the ocean’s interior at depths of 3,000 to 4,000 feet. Some sank into seafloor sediments. And some of that surfaced oil may have stuck to phytoplankton, which subsequently sank to the seafloor in a phenomenon that some call a “dirty blizzard.”

To complicate matters, unlike your dollars, oil not only moves around, it is also changes composition. Oil is made of thousand of different chemicals with different properties that, under different circumstances, can dissolve, evaporate, be eaten by bacteria, or transform chemically, leaving behind a gunky material we call a “weathered residue.”

Finding a footprint

A seminal paper published in 2012 estimated that of the 200 million gallons discharged into the environment, half never surfaced to form slicks but remained trapped deep in the ocean. We have just completed a study, published Oct. 27, 2014, in The Proceedings of the National Academy of Sciences, on how much of that trapped oil rained down to the ocean floor, and where it fell.

We used data generated by the U.S. government and recently released on the Internet. The data came from 3,000 samples of seafloor sediments collected at 534 locations by a dozen research expeditions that surveyed the Gulf of Mexico in 2010 and 2011.

The Gulf of Mexico is not a pristine location. It has a history of industrial oil pollution, as well as an estimated 200,000 gallons per day seeping from the ocean floor. These “leaky faucets” are a form of chronic pollution that has been going on for thousands of years in the Gulf.

To account for oil from the Deepwater Horizon event and not these other sources, we focused on the concentration and distribution of one compound, 17a(H),21b(H)-hopane, which was in Deepwater Horizon/Macondo oil and most other crudes from the Gulf of Mexico. Since it takes many years for seafloor sediments to accumulate as much as one inch, we argued that elevated levels of hopane from oil from Deepwater Horizon would show up only in the top half-inch of the ocean sediments and that the highest concentrations would be closest to the damaged well.

We compared hopane levels across the Gulf of Mexico, finding a distinctive peak in the top half-inch of sediments within 25 miles of the well. In contrast, we routinely find samples containing Deepwater Horizon oil hundreds of miles away on the beaches of Florida. The footprint we identified for Deepwater Horizon oil delivered to the seafloor was about 2 percent of the footprint for slicks at the surface, showing that oil spread much more widely at the surface than it did in the deep.

Trapped in the depths

We found that 4 to 31 percent of the oil trapped in the deep ocean—the equivalent of 2 to 16 percent of the total oil discharged during the accident—fell within a 1,250-square-mile patch of the deep seafloor. This is a minimum estimate, because it’s likely that we missed some of the oiled sediments, including other suspected oily areas around the well, and some oil was certainly deposited outside the patch.

Inspecting the data a bit more closely, we noticed that the oil was patchily deposited on the seafloor, but most intensively to the southwest of the damaged well. That establishes a trail for peak contamination.

Importantly, we identified hotspots of oil fallout near communities of damaged deep-sea corals. That supports previously disputed findings that these corals were damaged by the Deepwater Horizon spill.

We found peak concentrations of hopane lay at seafloor depths of 4,265 to 5,250 feet, bracketed by lower but still elevated concentrations up to 2,950 feet and down to 5,575 feet. This jibes with our previous studies in 2010 mapping a plume of dissolved oil and gas from the spill that flowed southwest at a similar depth.

We propose that tiny droplets of oil coagulated in these contaminated plumes and sank to the seafloor—possibly when they encountered high densities of particles that naturally cloud the waters near the seafloor. Some oil droplets came to rest at shallower depths in places where the contaminated plume flowed past higher-rising areas of the seafloor.

A similar effect occurs when oil contaminates coastlines; we call it the “dirty bathtub ring.” It was a big surprise that we saw dirty bathtub rings at the bottom of the ocean.

With so many cruises and so many samples, it’s humbling that we could definitively account for only 2 to 16 percent of the total oil spilled. But when you are doing a puzzle, putting any pieces into place eliminates other possibilities and begins to reveal the full picture. Our research illuminated several new pieces in the puzzle to assess damage caused by the Deepwater Horizon spill and give us new insights into the behavior of oil in the ocean. This new knowledge will improve ways to avoid and mitigate oil spills in the future.

Everyone wants the answer to the “oil question” wrapped up conclusively, like your checkbook at the end of the month. But the ocean is hostile and unpredictable, and just getting access to clues is hard-won.

Dave Valentine’s research was supported by the National Science Foundation (NSF). Chris Reddy’s research was supported by NSF and Gulf of Mexico Research Initiative’s DEEP-C Consortium.