Workers uncovered a tar mat weighing some 18,000 kilograms just offshore of a natural barrier island in Louisiana in the summer of 2013. Although the tar mat turned out to bear more sand than oil, it represented another small fraction of the hydrocarbons that went missing after BP's blowout in the Gulf of Mexico in 2010. The sum of all the dispersed oil located thus far, from tar mats to oily marine snow, hardly accounts for at least four million barrels of oil spewed into the cold, dark bottom of the Gulf of Mexico from the deep-sea well named Macondo five years ago.
Like any good mystery, this one may never be solved. Of that four million barrels or more spewed after April 20, 2010, more than a million remain missing, according to the best estimates of the U.S. government.
Mystery plagued BP's blowout from the beginning. Initial oil company estimates claimed just 1,000 barrels per day flowed into the deep—an underestimate off by at least 50 times, as measured by a device that assessed the actual pressure of the escaping oil attached later in the spill. "I felt like a general on the battlefield. There was a fog of research out there," says biologist Christopher D'Elia of Louisiana State University and dean of the School of the Coast and Environment. "We didn't know where the oil was, what it was like, where it was going, how it was being dispersed."
This is more than an academic exercise because the total amount of oil spilled will determine the total value of fines faced by the multinational oil company. The federal government's initial estimate concluded that 4.9 million barrels of oil spilled from the Macondo well over 87 days, of which 17 percent was captured at the wellhead, 25 percent evaporated or dissolved and 32 percent was burned, skimmed or dispersed chemically or naturally. That left more than one million barrels out there as tar mats, tar balls, plumes or buried in sand and sediments. Although a federal judge ruled earlier this year that the well spewed just four million barrels in total, he also concluded that more than three million entered Gulf waters, much of which remains out there. "They're still battling in court what this enormous settlement is going to be," D'Elia adds. "But it's always likely the truth is somewhere in the middle."

Given the uncertain, debatable measurement of the spill itself, perhaps the total amount could be calculated by all the Macondo well oil found in the Gulf or surrounding coastlines since the spill began. Scientists documented several plumes of oil drifting in the deep, including one that stretched 35 kilometers long, two kilometers wide and 200-meters thick in the months during which oil spewed into the Gulf. Much of that oil appears to have sunk to the seafloor, settling in a layer of "oil fluff," says biogeochemist Samantha Joye of the University of Georgia. That oily marine snow covers at least 3,200 square kilometers of the Gulf floor, according to research by biogeochemist David Valentine of the University of California, Santa Barbara. This mix of oil, mucus, shells, microscopic corpses and other detritus forms the top layer of the deep-sea sediment, as revealed by hundreds of cores pulled up from the bottom. But the oil appears in some patches and not in others just a few meters away, which Valentine, for one, attributes to the oil-forming droplets that hit some areas but not others: "We think it's some kind of misting of oily particles that are raining down on the seafloor."
Then there's the oil that made it to the swampy shoreline of the Gulf coast despite the best efforts of booms, dispersants and even ill-advised, hastily constructed barrier islands that quickly washed away. That oil can still be found along more than 1,600 kilometers of coast, especially Barataria Bay in Louisiana, among other regions, such as the tar mat off the island of Grand Terre or the tar balls that continually wash ashore. "Whether the tar balls are Macondo or not is always the question," D'Elia points out, given the many natural sources and other, smaller spills in the Gulf region.
The oil that disappeared into the sediment, whether in the marsh or the bottom of the sea, will remain there forever, however. Microbes in these sediments seem incapable of eating all of this oil—thus it will become a permanent part of the geologic record, especially the biggest hydrocarbon molecules. "If you look the cores, there's still that layer of sedimented oil on the surface but it doesn't reek of hydrocarbons anymore," Joye says. "The volatiles are gone."
And although microbes, sunlight and other natural processes eliminated much of the oil, whether it be in the deep, at the surface or on the shore, much of it remains—somewhere. "People are underestimating how much evaporates," argues marine chemist Chris Reddy of the Woods Hole Oceanographic Institution. As for the rest, he thinks it's still at the bottom, noting that the oil traveled only tens of kilometers at depth, compared with hundreds of kilometers when oil made it to the surface. In fact, the biggest concentrations of unexpected oil have been found within 40 kilometers of the Macondo wellhead, which also suggests that much of the missing oil may have sunk to the bottom of the sea. "It's not exactly missing," Valentine adds." At the same time we don't know exactly where it is either."
The Gulf of Mexico is a fairly big sea, after all—and even 210 million gallons of oil is a drop in the bucket of 643 quadrillion gallons of water. "You can easily account for the missing oil on the [continental] shelf, deep water, marshes and beaches," Joye says. That may be so, but definitive proof of that is lacking—and no smoking gun may ever be found. As Joye adds, in a sentiment echoed by her peers and even the Congressional Research Service: "I don't think we're ever going to be able to close this oil budget." The mystery of where much of the oil that spewed from BP's Macondo well ended up may never be solved.