Last September, BP (BP) tapped into a vast new oilfield called Tiber, estimated to hold at least 3 billion barrels of crude, or six months' worth of U.S. consumption. Discovered through seismic imaging and data crunched on a supercomputer, the field is nearly six miles beneath the Gulf of Mexico's floor, in a spot where the water is almost a mile deep. To access it, the British oil company and its drilling contractor, Transocean, sank the deepest oil well ever known. The $365 million rig that accomplished this feat was called the Deepwater Horizon, a floating thicket of machinery the size of two football fields, which held a crew of 130 and cost more than half a million dollars a day to rent from Transocean. As a drilling rig, the Deepwater Horizon didn't hang around to pump Tiber's crude. The self-propelled behemoth motored to the site of its next exploration, a perfect expression of humankind's ability to find buried treasure deep beneath the ocean floor and bring it safely to the surface.
Today the wreckage of the Deepwater Horizon lies upside down at the bottom of the Gulf under a mile of seawater in a place called Mississippi Canyon Block 252. Eleven of its crew are missing and presumed dead. Oil from the last well it drilled—not Tiber, but a much shallower, less complex job called Macondo—is spewing out of control from the seafloor, with perhaps 3 million gallons released at last estimate. A swath of the Gulf about the size of Delaware is covered by an iridescent, rusty orange sheen.
If the spill comes ashore, as scientists say it is almost certain to do, it could cause damage rivaling the 1989 wreck of the Exxon Valdez in Alaska's Prince William Sound, despoiling the breeding grounds of countless species in the fragile coastal buffer zone that provides hurricane protection. Already, the oil is threatening some of the most productive and profitable shrimping and fishing grounds in the world, part of a Gulf industry that provides a quarter of the seafood in the U.S. If efforts to seal the well go awry, they could cause far larger volumes of oil to spill. In some scenarios, the Gulf of Mexico loop current could even carry the oil around Florida and up the East Coast. The unfolding environmental disaster might yet become the worst in U.S. history.
The Deepwater Horizon exploded on Apr. 20 at a moment when U.S. energy policy was pivoting decisively in favor of undersea drilling. After two years of fierce partisan debate triggered by the $4-a-gallon gasoline prices of 2008, President Barack Obama and other Democratic leaders dropped their opposition to expanded offshore drilling. On Mar. 31, Obama proposed opening up large areas of East Coast, eastern Gulf, and Alaskan waters beginning in 2012—part of a grand bargain that Administration officials hoped would propel comprehensive energy and climate legislation designed to lead the country beyond petroleum, as BP's slogan famously declared. All of that is now in doubt.
For the oil industry, the Gulf of Mexico had been one of the few bright spots in a dim picture of flattening global production. After several years of decline, the amount of oil pumped in the Gulf has been rising as rigs move into deeper waters. The U.S. Energy Dept. predicted this year that by 2035, offshore oil production from the lower 48 states would grow more than 80 percent to account for about 38 percent of U.S. output, up from 30 percent now.
Until now, Gulf oil production has been not just expanding but reasonably clean, with serious spills so rare that industry and regulator alike behaved as if they were scarcely possible. Catastrophic accidents, it seemed, had been relegated to history by such high-tech gear as "blowout preventers" designed to shut off wells when pressures get out of control. These valves and shears were the last line of defense, a supposedly impenetrable Maginot Line that made other fail-safes unnecessary. The federal Minerals Management Service, which regulates offshore oil and gas production and collects reports on spills as small as a single barrel, was so confident of the system that it exempted BP from filing an environmental impact statement for the Macondo operation. MMS commissioned studies on creative ways to cope with massive well blowouts and never implemented them. It promulgated rules and allowed the oil industry to obey them on a voluntary basis. "It turns out, by the way, that oil rigs today generally don't cause spills," Obama said on Apr. 2. "They are technologically very advanced."
The Deepwater Horizon crisis has shifted the political conversation from seeking opportunity to avoiding risk. The Obama Administration announced a 30-day safety review. Florida Governor Charles Crist, who recently left the Republican Party to run for the U.S. Senate as an independent, backed away from his support for offshore drilling, as did California Governor Arnold Schwarzenegger. "It could have as big an impact as Santa Barbara," says Philip Verleger, an oil industry expert, referring to the 1969 oil spill off the California coast that was the catalyst for the first Earth Day in 1970.
Ultimately, no matter how many new regulations are passed, it will never be 100 percent safe to puncture holes in the ocean floor to drain out oil and gas. And until society manages to wean itself off hydrocarbons, offshore drilling will continue; the world needs the fuel. That said, advances in technology cannot be allowed to blind operators and regulators to the inherent risk. BP and its contractors were aware of other lines of defense. They just opted not to use them. As Mayor Tony Kennon of Orange Beach, Ala., told National Public Radio: "Why are we just now trying to figure out a Plan B? I think that bothers me more than anything right now."
Tony Hayward, BP's chief executive officer, slid into a booth at Copeland's, a restaurant on the main drag of Houma, a southern Louisiana town of strip malls and honky tonks. It was May 2, some 12 days since the crisis began. His company had lost $30 billion, or 16 percent, of market value, and Hayward, over dinner with a Bloomberg News reporter, was not bothering with the obligatory CEO pose of absolute confidence and control. He was sweaty and haggard; he picked at his ravioli and only sipped his beer. The 52-year-old had been going flat-out for days, and despite his weariness, he retained the determination that won him the top job over a pack of rivals at the London-based oil giant. He repeated a phrase he says was coined by Winston Churchill: "When you are going through hell, keep going."
He was succinct about the future of offshore drilling. "It all depends on how successful we are with our response," he said. "If we deal with the situation in a way that minimizes the environmental impact, it will cause some debate. If the environmental impact is serious, as a consequence there won't be much, if any, extension of offshore drilling."
Hayward heard about the blowout at breakfast in London on Apr. 21, about four hours after it began; he says his first reaction was "unprintable." For the first few days he kept a low profile. BP judged that Transocean (RIG), owner of the drill rig, should take the lead. Later Hayward deferred to U.S. government agencies such as the Coast Guard. In early May, BP changed strategies and put Hayward in front of the cameras. The company hired Marine Spill Response to deploy four 210-foot oil-skimming ships and two planes to spray dispersants on the oil. As the crisis mounted, Hayward tried everything at once. The only problem was that, even at a cost of $7 million a day, none of it stopped the leak.
A mile below the surface, things can go to hell in an instant. The pressures and temperatures at work are otherworldly. Imagine an elephant sitting on your chest, and you get a small sense of the weight of rock and water pressing down on the reservoir of oil and gas miles below the surface. To keep the superheated, supercompressed fluids from shooting upward like a volcanic eruption before the well is finished, drillers fill the hole completely with a heavy, synthetic "mud." Then, to finish the well, they inject a high-tech cement. Each well requires its own unique formulation of mud and cement. The cement is supposed to go down the middle of the drill pipe—a seven-inch tube surrounded by a larger pipe called the casing. When it reaches the bottom of the drill pipe, it oozes up into the gap between the pipe and its casing before drying in place, forming an impenetrable seal.
At least, that's what's supposed to happen. The Deepwater Horizon accident occurred at the final stage of the job, as the rig crew was preparing to put a temporary seal on the well and move on to another site. The exact circumstances aren't likely to be known for months, though it's clear that pressurized natural gas was able to infiltrate upward, meaning the seal was imperfect. It's the same thing that happens when you crack open a soft drink and tiny bubbles of carbon dioxide expand and rise—the difference being that natural gas is explosive. Mistakes like this are not rare. A 2007 MMS study found that cementing was a factor in 18 of 39 Gulf of Mexico blowouts over 14 years. The pressure surge from a gas bubble even has a nickname: the kick. There are procedures for recementing, but those take time and money; each extra day of leasing the drilling rig costs about $500,000. Halliburton (HAL) was in charge of cementing, under BP's direction. Robert MacKenzie, a former cementing engineer who is now a securities analyst for FBR Capital Markets (FBCM), says he wants to know whether BP ordered a so-called cement bond log test to evaluate the cementing. Such a test would have determined whether or not a remedial cement job was necessary. BP declined comment.
When all else fails, the last line of defense in a subsea well is the blowout preventer, which uses valves to close off the flow of oil and gas if pressures get too high. If the regular valves fail, the doomsday feature of the blowout preventer is a series of shear rams—essentially giant pairs of hydraulically powered scissors that are supposed to close off the opening by slicing through all the pipes. The pressure the shear ram can exert is greater than the pressure on the struts of a landing, fully loaded Boeing 747 as the rear wheels hit the runway, says Satish Nagarajaiah, a Rice University professor of mechanical and civil engineering. The Deepwater Horizon's blowout preventer had five hydraulic rams. For reasons that are still unknown, they didn't do the job.
The blowout preventer, made by Houston-based Cameron International (CAM), failed to close off the flow of oil and gas. The gas came up the pipe to the rig and then, heavier than air, settled in low spots. Survivors later said they heard a thump, then a hissing. The mammoth rig began to shake, and explosions began at about 10 p.m. on Apr. 20, sending flames hundreds of feet into the air. To Captain Michael Roberts, a commander of a supply vessel that arrived just a few hours later, "it appeared as if the sun was coming over the horizon," he told CNN. Nearly two weeks after the accident, rescue crews still had not managed to get the blowout preventer working. "We have found that there are some leaks on the hydraulic controls" of the blowout preventer, Bob Fryar, senior vice-president of BP's exploration and production operations in Angola, in southwestern Africa, told the Houston Chronicle.
Hayward said he was mystified that the blowout preventer failed, though failure isn't a rare event. The last-ditch shear ram is rarely tested under real conditions because of the destruction it causes. In a 2002 laboratory test for MMS, researchers found that three of six shear rams failed. Seven other makers declined to be tested.
Within a day of the accident, BP had sent as many as eight underwater robots to the scene. Hayward and other company executives watched from a special room inside the company's suburban Houston complex as the robots, which look like sleds and are painted yellow to be visible at ocean depths, did their work. Tethered to a mother ship by wire, they fed video images to their pilots, who use joysticks to move the vehicles and manipulate their tool arms. BP executives hoped the robots could pull external levers and get the blowout preventer to snap shut. A BP spokesman says the company no longer expects the robots to get the device working.
That doesn't mean they are being put back in their boxes. Robots are the only eyes and ears that BP and other responders have below the sea. The company uses them constantly to monitor the wellhead and mile-long length of pipe that has fallen to the bottom of the Gulf. BP and its contractors are now deploying the robots in a new attempt to shut down the undersea leaks. The company is constructing a four- story steel funnel that it will try to lower over the most serious of the three. Once the device is in place, a cap attached to a pipe will be placed on top. The hope is that the captured oil, which is lighter than water, will rise up the pipe to a special surface vessel where it can be separated from water and disposed. BP estimates that if successful, the funnel could divert 85 percent of the oil flow. On May 5 the company succeeded in stopping the smallest of three leaks, a dribble from an oil-filled section of pipe that had broken off.
As the spill spread, BP assembled at least nine airplanes and hundreds of ships for cleanup. Its logistics base for the Gulf of Mexico, in Houma, has been turned into a command center where uniformed U.S. Coast Guard personnel mingle with specialists from BP and other oil companies such as ExxonMobil (XOM). "There is no expense spared," says Jacqueline Michel, president of Research Planning, an oil-spill consulting firm based in Columbia, S.C. One fear: Because oil is still flowing, a cleanup exercise might be in vain—and even damaging to the most fragile areas, because it might have to be repeated.
Before the accident, Hayward seemed to have BP on the right track after succeeding his mentor John Browne, who resigned amid personal scandal in 2007. Browne did the megadeals—a $62 billion merger and acquisition of Amoco in 1998, a $32 billion deal with Arco in 2000—that propelled BP into the supermajor class. Yet the company during his tenure was hopelessly accident-prone in the U.S., its most important operating area. In 2005 a blast at a BP refinery in Texas City, Tex., killed 15 people and injured more than 180. In 2006 oil leaking from a pipeline forced BP to shut down part of its key Prudhoe Bay oilfield in Alaska. Hayward has improved BP's safety record, accelerated much-delayed projects, and backed away from some of Browne's investments in alternative energy that Hayward deemed uncommercial. Profits more than doubled for the quarter ending Mar. 31, to $5.6 billion.
More than any other major, BP has staked its future on aggressive projects in the deepwater Gulf. "We don't do simple things," said Andy Inglis, the company's exploration and production chief, in an interview last summer. "We are prepared to work on the frontier and manage the risks." Using seismic imaging techniques, BP made big discoveries including Thunder Horse, now the second-largest producer in the U.S. at about 300,000 barrels per day. Last September's Tiber discovery was one of the world's most important for the year.
The Macondo prospect, where the accident occurred, was a much simpler job than Tiber. The well was only half as deep. Hayward insisted the accident was the fault of the drilling company, Transocean, not BP. "It is not our rig, not our equipment, not our systems," he told the BBC on May 3, before edging toward a recognition of political reality. "There may be a need to do more oversight of drilling equipment," he said at dinner.
Clearly, the industry and its regulators will have to revisit how they operate. Inquests into the Deepwater Horizon disaster may find that complacency played a crucial role. Take one example: The four-story-tall steel funnel that BP fabricated in Port Fourchon, La., and plans to lower over the largest of the three leaks, was not expected to be operational until three weeks after the accident. Shouldn't the industry already have several such chambers on standby in offshore drilling regions? This isn't exactly a fresh idea: In 1979, Brown & Root lowered a 40-foot-wide "steel sombrero" over the blowing Ixtoc I well in the Bay of Campeche, Mexico, which spilled 35 million barrels before the sombrero helped stanch the flow.
Over the years, researchers funded by MMS have proposed a variety of ideas for coping with well blowouts; none has gotten past the drawing board. In 1985, Brown & Root recommended that an oil tanker be retrofitted to collect huge volumes of oil and seawater funneled from a blown well and to pump the liquid into separating tanks. Estimated cost: $59 million. Without commenting on the merits of that particular idea, industry consultant Robert Peterson says it's common for the industry to neglect promising innovations. "The transfer of the technology from the lab to a commercial application is where you see the highest mortality rate in terms of good ideas," says Peterson, a Houston-based oil and gas expert at Charles River Associates. A spokesman for the Minerals Management Service did not respond to questions about past research.
One reason for complacency is that MMS's own statisticians never flagged the possibility of a serious spill. By design, the department's Oil-Spill Risk Analysis model focuses on the likely frequency of spills rather than how big one might get. Projecting from the history of small spills over the past several decades, MMS in 2007 predicted just five spills of 10,000 barrels or more per 100 billion barrels of oil produced. Ten thousand barrels is about what the Macondo well is thought to be leaking every two days. The routine spills that MMS used to calibrate its prediction model "have no bearing on dealing with major spills, predicting their frequency, or getting ready for them," says Zvi Ziegler, a mathematician at the Technion-Israel Institute of Technology in Haifa, Israel. The MMS press office did not return calls asking about the agency's statistical methodology.
Critics of MMS have repeatedly called it a captive of the companies it regulates. Last September the Interior Dept. shut down an oil royalty program run by the agency after audits found that MMS was undercollecting millions of dollars worth of royalties. The Interior Dept. inspector general's office found that several MMS officials had "frequently consumed alcohol at industry functions, had used cocaine and marijuana, and had sexual relationships with oil and gas company representatives." In the latest embarrassment, BP was one of three finalists for the agency's annual safety award for major offshore producers. The ceremony, scheduled for May 3 at a conference in Houston, has been postponed.
Even with MMS as its minder, BP felt threatened by overregulation. In a letter to regulators last September, Richard Morrison, the company's vice-president of Gulf of Mexico production, wrote to oppose new rules meant to reduce the risk of injuries and spills, saying that voluntary programs "have been and continue to be very successful."
The biggest question now is unanswerable: Whether the Macondo blowout was a fluke in an otherwise safe system or an indication of more trouble to come. Earlier this year, 19 members of Congress wrote to mms asking for its response to accusations by an anonymous former contractor that BP did not complete crucial engineering drawings and other paperwork for subsea components of its Atlantis project, which began producing oil in 2007. The ex-contractor supplied what he claimed were internal BP e-mails expressing concern about the incomplete documentation. In one such e-mail made public by the self-proclaimed whistleblower, a BP executive on the Atlantis team named Barry C. Duff writes that the incomplete documentation "could lead to catastrophic operator errors." Food & Water Watch, the organization that publicized the accusations, has refused to release the whistleblower's name to the news media at the person's request. An MMS spokesman declined to comment on the allegations. In a Jan. 15 letter to the House subcommittee on energy and minerals, BP did not dispute the authenticity of Duff's e-mail but denied that it had mismanaged records or jeopardized the project.
Some of the industry's most reliable supporters have been taken aback by the Macondo spill. "This is like a nuclear plant going out of line; it is too risky for the environment," says Nansen Saleri, a former chief of reservoir management at Saudi Aramco. "This should be a lesson to learn and improve; offshore resources are too important to be written off."
Thus far President Obama has not backed off his proposal to expand offshore drilling. Aides say the Administration is using the 30-day safety review to devise rules that could make offshore drilling safer and, perhaps, rescue his delicate compromise of more drilling along with more efforts to boost clean energy and combat global warming. Regardless, the crisis is likely to further splinter an already skeptical Congress. Senator Frank Lautenberg (D-N.J.) is leading a drive to increase the industry's liability for economic damages, which is currently capped at $75 million per incident. As for BP's Hayward, he will need both savvy and luck to avoid having this incident define his tenure, just as the Texas City refinery explosion defined his predecessor's. Noting that he has spent his first three years as CEO restoring BP's fortunes, Hayward says: "My task for the next three years is to put this event behind us."
A Comfortable Safety Record—Then Disaster
Annual oil spills from U.S. offshore platforms (In thousands of barrels)
2000 - 0.3
2001 - 0.2
2002 - 1.4
2003 - 0.8
2004 - 0.6
2005 - 8.0
2006 - 1.2
2007 - 0.1
2008 - N.A.
2009 - N.A.
2010 - 70 (Estimated total as of May 5, assuming 5,000 barrels per day)
With Joe Carroll, Jim Efstathiou Jr., Hans Nichols, Jim Polson, and Jessica Resnick-Ault