As oil continues to spew from the floor of the Gulf of Mexico for a 21st consecutive day, British Petroleum executives, government leaders and the oil industry as a whole are working to develop new methods to contain the leak after other efforts fell flat.
Over the weekend, a plan to cap the leak with a 100-ton containment dome failed when icy residue clogged the device, which was intended to pump oil from the site of the leak to a waiting ship. Now the dome sits on the seabed, away from the spill, according to BP. The containment dome strategy came after BP's robots failed to trigger the well's blowout preventer device and clamp the leak nearly a mile underwater.
The company said it has already spent some $350 million since the April 20 explosion that started the disaster, including efforts to stop the leak and contain the damage. Still, oil continues to gush at a rate of about 210,000 gallons a day.
Despite the hundreds of vessels, thousands of workers and 1 million feet of boom deployed to help gather the oil and polluted water, evidence suggests that the oil has already made landfall. Ed Overton, a Louisiana State University professor of environmental studies, tested a sample of the tar balls that landed on Alabama's Dauphin Island over the weekend and estimated with 95 percent certainty that they came from the recent oil spill.
So what are the options going forward?
BP is again using underwater dispersant chemicals to break up the oil before it reaches the surface. That tactic was abandoned last Wednesday after questions about the chemical's environmental impact were raised, but BP said it's using the same approach again, in collaboration with the Environmental Protection Agency and the National Oceanic and Atmospheric Administration.
There has been little scientific study of the ecological impact of using underwater dispersant, which experts said is essentially like using dish soap and no more harmful than the oil itself.
The company said in a statement today that it would try again to use a containment dome to gather the oil.
"A second, smaller containment dome is being readied to lower over the main leak point," BP said, noting that this time the equipment was designed to deal with the kind of icy clog that sidelined the larger dome. Hedging its bets, the company wrote that "This operation has never been done before in 5,000 feet of water."
BP is also pursuing other strategies. One option would be to cut the riser pipe of the well, and pump the oil to a ship on the surface using a larger pipe. It could be risky, because cutting the pipe would actually increase the flow of oil.
"That's a very tempting option," Philip Johnson, petroleum engineering professor at the University of Alabama told the AP. "The risk is when you cut the pipe, the flow is going to increase. ... That's a scary option, but there's still a reasonable chance they could pull this off."
Another choice is what BP calls a "top kill" option. Under that plan, the well would be injected with junk, like rubber tire shards and golf balls, to clog it and stop the leak.
None of these methods are guaranteed to work, though, because it's the first time they've been tried at these depths.
"There is a lot of real-time learning going on," said BP CEO Tony Hayward at a briefing at the company's command center in Houston today. "We are doing it for real for the first time."
BP does believe, though, that relief wells currently undergoing drilling will offer a definitive solution if all other methods fail. Hayward insisted today that the wells "will assure ultimate success," when they are completed, but that could take up to three months. The well is already under way, which means if all other attempts to stop the leak fail, oil could be flowing into the Gulf of Mexico and the fragile Mississippi Delta for weeks to come.
ABC's Barbara Garcia and the Associated Press contributed to this report.