Low clouds and heavy snow have grounded aircraft in Valdez since Sunday, so no one really knows what is happening to the oil spilled by the tanker Exxon Valdez a month ago. But dueling computer models continue to spew numbers on the disposition of the oil, and those numbers have taken on a life of their own.
K.T. Koonce, a senior Exxon USA vice president, declared earlier this week that 70 percent of the oil spilled by the Valdez has evaporated, dispersed or been collected.
"Exxon has been quite successful in oil recovery during the past week," he said. "We have been recovering about 3,000 barrels (126,000 gallons) per day, and are rapidly eliminating the oil remaining in the Prince William Sound."
But people who were on the water of the Sound said they saw lots of oil, experts said the barrels of recovered oil for which Koonce took credit were probably mostly water, and the amount Exxon claims has disappeared is only an estimate based on models which some scientists say are unreliable in these conditions.
Scientists and agency officials in Valdez all have their own numbers, but say they have little practical meaning, and that they will not be used to decide when the Sound is clean.
"I think almost no one takes them seriously except the press," said Oceanographer Jerry Galt of the National Oceanographic and Atmospheric Administration.
Measuring millions of gallons of oil spread over thousands of miles of water is a superhuman task, but anyone can make an estimate. While even such basic numbers as the amount of oil originally spilled from the tanker continue to change, Exxon releases daily reports which claim to pin down how much oil has evaporated.
Hugh Brown, Exxon's expert on the subject, declined to comment on the subject in detail. "We don't want to get in a numbers game here, and I'm pretty sure the public relations people are going to come up with a statement on it," he said.
Joe Tucker, of Exxon's public affairs department, was unable to provide information on the subject Tuesday and Wednesday.
Brown acknowledged that Exxon's evaporation figures are based on computer models combined with estimates based on past experience.
State and Coast Guard officials say they don't know the source of that information, but it probably comes from computer models like their own. The models are the electronic equivalent of looking at a table in the back of a book to find out what happened to oil elsewhere in similar circumstances.
But a scientific paper says such modeling doesn't work on this kind of oil in this kind of water.
James Payne and G. Daniel McNabb Jr. tested the oil in a wave tank in Kachemak Bay in 1984 and then wrote that modeling the evaporation and break up of the oil, "is too complex even for sophisticated computers."
Not all scientists agree with that conclusion, but no government agency says it trusts the numbers enough to use them to make decisions. The Coast Guard and Alaska Department of Environmental Conservation don't even put stock in the amount of oil Exxon says it collects each day.
"We know there are rough figures on the amount of oil they are recovering, but included in those numbers is an awful lot of water," said Bill Lamoreaux, the state's onscene coordinator.
During its weeks in the ocean, the oil has bonded with water in an emulsion that ranges in consistency from scrambled eggs to heavy syrup. When workers pick up a gallon of the substance, as much as 80 percent of each gallon is water, Lamoreaux and others said. The state plans to send samples of all the oil collected to a lab to find out how much of it really is oil.
The amount of oil recovered can mean money to Exxon. A state fine imposed on oil spillers, informally called the dollarspergallon law, gives a spiller credit for gallons recovered.
When the oil bonded with the water the total volume of material that needs to be cleaned up increased, Lamoreaux said, and exceeded the total that was originally spilled.
The amount of the original spill has also been revised again. The initial estimate was 11.3 million gallons. Exxon revised that to the widely quoted 10.1 million. Now, the state says 11.3 million gallons escaped the tanker.
"That (10.1 million gallon figure) was just an estimate at the time and people locked onto it," said Lamoreaux.
He also pointed out an apparent error in the amount of oil Exxon at one time claimed to have burned. Only one successful test of oil burning was conducted, shortly after the spill. Lamoreaux said the head of that operation told him 15,000 to 30,000 gallons were burned in the test.
But for weeks Exxon reported daily that 19,000 barrels had been burned. Lamoreaux said the company apparently got mixed up between gallons and barrels. Instead of burning 8 percent of the oil, the test in fact burned .15 percent, he said.
Exxon's Brown would not answer questions on how much oil came out of the tanker, how much was burned, or other issues arising from Exxon's statements on the numbers.
The confusion about numbers may be presenting an unrealistically rosy picture of the spill to the public, but regulators say they have no intention of using the computergenerated information to make decisions about the spill.
"I'm not sure how a computer is going to make decisions about the cleanup," Lamoreaux said. "We've got oil on the water. We're going to have real people decide if the oil is still out there."
Lt. Kathleen Donohoe of the Coast Guard said the same thing. "We're avoiding giving out numbers because we know they're inaccurate," she said. "We're making decisions based on the oil we see in the water."
But while that oil, and oil that has spread out into the Gulf of Alaska, may be impossible to recover, that doesn't mean it will disappear.
The scientific literature predicts that only small quantities of the oil will dissolve into the water. Galt said tests have shown that prediction to hold true.
He guessed, based on past studies, that 40 percent has evaporated and otherwise been shed from the water's surface. Twenty percent to 30 percent may be on beaches, he said.
The rest is still floating around, getting ever harder to find.
"If you have 10,000 square miles of scattered tar balls, you're not going to see it very much," Galt said. "You could say that's lost, that's gone, or dispersed. But it's an awful lot like cigarette smoke. As you get far away, you can't see it anymore. But that doesn't mean it's not there."
story 100 of 380
story 35 of 42