A plume or not a plume? That was the question for scientists, oil company employees and government officials in the early days of the oil spill into the Gulf of Mexico from BP's Macondo 252 well.

"Most oil accumulates on the surface, historically," noted marine biologist Jane Lubchenco, administrator of the National Oceanic and Atmospheric Administration, at a press conference on May 12. After all, many of the hydrocarbons in oil are less dense than water and therefore should quickly rise to the surface. "Subsurface oil is a unique feature of this spill," Lubchenco said, a feature that government officials occasionally seemed to downplay.

NOAA and partners "have been doing extensive testing for subsurface oil to see if it's there and, if so, what concentrations and where," Lubchenco added at a July 22 press briefing. "What we're finding is that there is subsurface oil right near the vicinity of the well head, and as one goes farther away from the well head the oil is highly dispersed," in concentrations ranging from four to seven parts per million between 1,000 and 1,300 meters down. On August 18, she added, "The oil that is subsurface is highly dispersed, it's in parts per million in the water column, and it appears to be biodegrading relatively quickly."

But now scientists have presented definitive evidence that a significant amount of hydrocarbons well above normal concentrations for Gulf of Mexico waters formed an at least 35-kilometer-long plume at a depth of 1,100 meters that followed the contours of the seafloor to flow at 6.5 kilometers per day in a west-southwesterly direction away from the Macondo well. "What we found is that a subsurface hydrocarbon plume existed," says ocean physicist Richard Camilli of the Woods Hole Oceanographic Institution (WHOI). "It was created by the Deepwater Horizon Macondo well."

Such a plume could cause problems for sea life, ranging from the direct toxicity of the oil itself to microscopic plants to the physical coating of marine animals and the formation of low- or no-oxygen dead zones. But any biological or seafood impacts remain unknown at this point.

And the life that benefits from an oil spill—hydrocarbon-eating microbes—appeared not to be biodegrading the oil in the plume quickly at all, at least in late June when the surveys were made. "Where we studied we did not see any evidence of a dead zone," or an area where available oxygen in the seawater has been used up by microbes busily eating the hydrocarbons, says WHOI marine chemist Chris Reddy. Some computer modeling has shown that such dead zones could form as microbes degrade the hydrocarbons. "You can't tell a microbe when to do what and where to do it. They're like a teenager given a chore."

Marine mission

Researchers from WHOI—armed with a National Science Foundation rapid-response grant and an autonomous underwater vehicle dubbed Sentry equipped with a mass spectrometer—cruised the Gulf of Mexico from June 19 to June 28 on the University of Rhode Island research vessel Endeavor. Using Sentry to sniff out subsurface oil, the researchers found a plume of hydrocarbons that the underwater robot crossed in and out of 19 times during its "hunter-gatherer" mission. "We hunt the plume and gather samples," Camilli says.

More literally, the free-swimming Sentry hunted the plume, and whenever it detected unusually high levels of hydrocarbons, it reported back to the Endeavor acoustically—some 57,000 chemical samples in all. Scientists on board also collected water samples—some of which had to be picked up by other boats and delivered to shore mid-cruise because of their short shelf life as their contents degrade. "They are as clear as spring water," says Reddy, who collected the samples along with representatives from BP, NOAA and the Environmental Protection Agency. "I don't think I ever smelled oil. It's not some raging river of chocolate syrup."

But there are significant hydrocarbons, specifically the benzene, toluene, ethylbenzene and various xylenes Sentry was set up to sniff out. The continuous deepwater plume had such hydrocarbon concentrations of 50 micrograms per liter, indicating additions of at least 5,500 kilograms per day from somewhere, most likely the spewing Macondo well. For comparison, that is more than twice as much added by all the natural oil seeps in the northern Gulf of Mexico. And a more diffuse plume existed at shallower depths between 50 and 500 meters. Sentry tracked the plume for 35 kilometers from the source and had not found its end before operations had to be called off because of the approach of Hurricane Alex. "It therefore appears likely that the plume extends considerably beyond the survey bound," the researchers wrote in the paper presenting their findings, published online in Science on August 19.

Previous experience suggested that such plumes might form; the Ixtoc blowout in the Gulf of Mexico in 1979 created a plume of at least benzene stretching for 40 kilometers, even though that well was only at a depth of 50 meters. And previous research, such as Project "Deep Spill" conducted by the Minerals Management Service in June 2000 off the coast of Norway, also suggested that such plumes could form when oil was released at depths of 800 meters. But that experiment released oil for only two hours and therefore was unable to determine whether the plume it generated was a temporary effect or something more permanent. "It's not enough to establish a steady-state plume," Camilli says. "To get a plume that can track out to 35 kilometers, you have to have a continuous release in order to create a continuous plume."

"Continuous release" describes the Macondo well between April 20 and July 15, when the oil was finally shut in the well. "This plume was incredibly stable at depth," Camilli notes. "It only varied by 10 to 20 meters [in depth] at a time."

Microbes did not seem to be affecting that stability too much, drawing down oxygen at a rate of one micromole per day. That rate "is considerably faster at that depth than at any other location [at that depth] but also considerably slower than the rate of microbial activity in the surface," says WHOI geochemist Benjamin Van Mooy, because of the cold temperatures at depth. "The rate of hydrocarbon degradation in the plume was about 10 times slower than at the surface."

Known unknowns
It remains unclear whether the plume formed as a result of physical interactions between the water, hydrocarbons, temperature and depth—or whether subsurface application of chemical dispersants that break the gushing oil into globules played a role. "We don't know the process," Camilli says.

And it remains unclear if the plume detected by the WHOI scientists was the only plume or if others formed as well. Scientists from the University of South Florida found oil droplets likely to have come from the Deepwater Horizon blowout coating the sediment in DeSoto Canyon to the east-northeast of the Macondo well in August. "There may have been others in other places at other times," Camilli says. "We just don't know."

Also unknown: all the different hydrocarbon constituents or dispersed oil that made up this plume, though the specific set of hydrocarbons measured here typically compose 1 percent of the Louisiana sweet crude that came from the Macondo 252 well, according to Reddy. Nor are its potential biological impacts known. "With time, we will be able to sit down with toxicologists and ecologists, when we have a full data set, and see whether or not the extent, the amount and the duration of exposure would cause any damages," Reddy says.

But it is now known that, at least in cases like the Deepwater Horizon blowout, plumes will form. "What we found went against conventional wisdom," Reddy adds. "Plumes can be formed. They can be continuous and coherent, and they can move at a known velocity."

And it is broadly consistent with the findings of other research cruises studying this oil-spill disaster, like the one from the Monterey Bay Aquarium Research Institute in early June and one undertaken in mid-June by geoscientist John Kessler of Texas A&M University and colleagues. "Scientifically, there was no controversy. This is exactly what we would have expected," Kessler says. "The controversy is in the court system as people try to quantify BP's culpability."

In fact, the data gathered here will be used to assess the actual natural-resource damages involved in the Deepwater Horizon disaster. Meanwhile, large quantities of hydrocarbons are still out there, especially methane, which made up the most significant part of the hydrocarbon release, according to Kessler. "Dissolved oil is decreasing rapidly," he says. "Methane is not, and based on our calculations it will be there for a long, long time."

It is already clear that the hydrocarbon plume lasted throughout June, and likely longer. "This plume persisted at this depth interval for months," the researchers write. Camilli adds, "As to where it's gone and what will become of it? I don't know."