The fight against the gulf oil spill is already writing lessons for future cleanups. Unfortunately for the gulf coast, outside experts say, many are lessons in what not to do.
Since the Deepwater Horizon rig exploded April 20, outsiders watching the cleanup say, the federal government and BP have made key mistakes that delayed or distracted the effort to stop the spill. Some were probably inevitable: BP engineers and government officials were forced to improvise in the face of mounting disaster.
But other missteps -- seen with the calm and clarity of hindsight -- look as if they could have been avoided.
Officials used "dispersants" to break up the oil. But some experts think that those chemicals caused much of the oil to remain below the water's surface, out of reach of standard cleanup techniques. The first attempt to place a "dome" over the well failed because of a well-known problem called hydrate crystals. The government and BP repeatedly under-estimated the oil's flow, and BP was not ready to capture all of the oil being siphoned up from the well.
In all, it appears that the mistakes have made it harder to fight what President Obama has called a "war" on the oil.
"There have been days where we've actually recovered more oil through containment and recovery than what came out. I think there are probably days that we didn't," Coast Guard Admiral Thad W. Allen said Thursday, when asked whether the war on the oil was being won. "I don't think I could come up with a combination of win and loss days."
In an interview Thursday, BP spokesman Toby Odone said his company faced a daunting problem: a huge leak involving broken machinery at the crushing depth of 5,000 feet. But he said his company's response had been shaped by an official plan.
"We didn't know what state the blowout preventer was in, what state the well was in, what state the riser was in, so there's been a lot of learning," Odone said. "We've learned. I don't know if we've made mistakes. I think we've tried things and we've learned from that."
So far, statistics of the cleanup effort indicate, at best, mixed success. A flotilla of vessels have skimmed 21.9 million gallons of oily water from the gulf, and 5.2 million gallons have been burned.
BP has siphoned 202,000 barrels (8.5 million gallons) to the surface using a "cap" over the leak. That rate increased Wednesday, when a second specialized ship arrived to help with the task. Also, on Thursday Allen said a relief well, being drilled to plug the well far beneath the sea floor, was ahead of schedule.
But the oil is still spreading faster than it can be cleaned from beaches and marshes. On Thursday, the Coast Guard said oil was on about 72 miles across the Gulf Coast, up from 68 on Sunday.
"I would give them a C-plus or B-minus, especially at the beginning. I think it's getting better now. I think they fell into the trap of following standard steps -- the usual procedure you would follow to deal with a situation like this -- and the situation was not quite usual," said Tadeusz Patzek, a professor of petroleum engineering at the University of Texas. "They used up precious time doing things that were not successful."
One of the most-criticized decisions was to use dispersants under the surface of the gulf.
The point, according to the federal government and BP, is to break the oil into smaller droplets, to promote consumption by oil-eating microbes. This week, federal officials said they stand by the decision, although they have directed BP to limit use of the undersea dispersants to 25 percent of what it was originally.
But biologists and petroleum engineers say that by keeping sunken oil below the surface, dispersants have made the country's arsenal against oil spills -- skimmer boats, controlled burns, containment boom -- less effective.
And scientists worry that under the surface, the oil may cause unseen harm to species as varied as plankton and whales. One new survey, by James H. Cowan Jr. of Louisiana State University, indicated a 300-foot-thick "cloud" of oil, 35 miles away from the leak.
William K. Reilly, the former head of the Environmental Protection Agency and the co-chairman of Obama's commission on the spill, said he was troubled by how little is known about the dispersants' effects on the environment.
"I suspect that a lot of the pressure to use dispersants is cosmetic," Reilly said, meaning that they help keep the oil out of sight. He said that he had not allowed them to be used after the Exxon Valdez spill in 1989. "When the salmon fry came out of the hatcheries, they swam under the oil -- which was on the surface -- rather than through the oil."
In addition, outsiders have said BP and the government greatly misjudged the amount of oil flowing from the well. Eight days after the explosion, they produced a combined estimate of 5,000 barrels (210,000 gallons) a day. When outsiders said this appeared too low, the Coast Guard answered that it was already responding to a worst-case scenario.
Now, latest government estimate is that the flow is more like 35,000 to 60,000 barrels (1.47 million to 2.52 million gallons) a day. And when the cleanup effort began using the new "cap" to siphon oil and gas away, it became obvious that there was more of it than BP's ships could take. Vessels are being brought in from other parts of the world and are expected to arrive by the end of June.
"They seriously underestimated the nature of the blowout and the rate of the well," said Nansen Saleri, chief executive of Quantum Reservoir Impact, an oil technology consulting firm. "Of course, they don't have the processing and handling capacity at present."
Another misstep occurred early in the crisis, when BP tried to lower a containment dome onto the collapsed riser pipe. Within moments of being lowered over the pipe, hydrates formed and clogged the pipe leading from the top of the dome.
BP has acknowledged that its engineers were surprised by the amount of hydrates that formed.
"We're all kind of Monday morning quarterbacking here," said Bruce Bullock of Southern Methodist University. But he said the dome effort was clearly "a mistake. The hydrate problem at that depth and that temperature has been so obvious for so long, that it shouldn't have been underestimated."