Updated at 7:36 EST.

The White House may have ignored the advice of federal scientists when crafting a heavily criticized draft report last August accounting for oil spilled into the Gulf of Mexico by the Deepwater Horizon disaster.

Several scientists raised concerns about the draft of the so-called oil budget, according to e-mails between scientists at the National Oceanic and Atmospheric Administration, U.S. EPA and the U.S. Geological Survey. The e-mails -- a sampling of thousands of documents -- were obtained by E&E along with letters sent today to the White House from the nonprofit Project on Government Oversight and Rep. Raúl Grijalva (D-Ariz.).

At issue is the Aug. 4, 2010, interagency report that accounted for the spilled oil, indicating the amounts that had been burned, skimmed and dispersed. The document claimed all but 26 percent of the oil could be accounted for at that point, and the Obama administration used the document to paint an optimistic picture of the situation in the Gulf, a move that was criticized by academics, lawmakers and environmentalists who said the report -- and the administration's presentation -- downplayed the severity of the spill.

Carol Browner -- President Obama's top energy and climate adviser, who announced last night she will step down from her White House post -- said the report indicated the vast majority of the oil was "gone." The White House later admitted she was speaking off the cuff. And findings of the the August draft were ultimately backed by an independently peer-reviewed final draft, which was released in November.

But the sampling of internal e-mails made public today highlights that some top federal scientists working on the report also had concerns about the draft's presentation ahead of its release and immediately afterward.

NOAA Administrator Jane Lubchenco wrote in an e-mail to Commerce Department officials and White House communications staffers on the morning of the report's release that she was "concerned to hear that the oil budget report is being portrayed as saying that 75 percent of the oil is gone."

"It's not accurate to say that 75 percent of the oil is gone," Lubchenco wrote. "Fifty percent of it is gone -- either evaporated or burned, skimmed or recovered from the wellhead. 24 percent has been dispersed, and although much of this is in the process of being degraded, it is not 'gone' yet."

The internal e-mails go on to highlight two other instances when federal scientists disagreed over decisions to gloss over scientific uncertainties and to simplify information in the report. At least one of those came as a result of White House "pushback," according to the e-mails.

Danielle Brian, executive director of the nonprofit watchdog Project on Government Oversight (POGO), and the group's investigator, Paul Thacker, expressed concern about the e-mails in the letter today to the White House.

"Based on e-mails in our possession at this time, POGO is concerned that the White House may have ignored expert advice from agency officials and pressured scientists to make changes during the development of the report in order to advance a public relations agenda," they wrote.

White House spokesman Clark Stevens said, "The federal response required extensive coordination and involved a broad range of scientists and experts from across, as well as outside, the federal government. These e-mails reflect close and effective coordination and healthy scientific debate among numerous federal agencies and the White House."

NOAA spokesman Justin Kenney responded similarly, saying the documents showcase "healthy scientific debate that is the hallmark of good policymaking, especially during a national crisis."

"Contents of these e-mails show the effort made to communicate real-time scientific information to the American public quickly and clearly," Kenney added. "We were clear in our release, the report and at the press briefing how those numbers should be interpreted."

Stevens added that the oil budget has since been extensively peer reviewed, and Kenney said that revised oil budget report released in November provides the technical documentation for the calculations and validates the early results released in August.

EPA's concerns

The original oil budget -- using the Aug. 2 federal estimate of the spill at 4.9 million barrels -- says 16 percent of the spilled oil was dispersed naturally and 8 percent was dispersed using chemicals. A later revision of the document, out in late November, increased the estimate of chemically dispersed oil to 16 percent.

But EPA scientists had cautioned officials at other agencies against making misleading characterizations about the federal government's dispersion efforts.

"The percentages are very rough and should not be considered accurate. We still do not believe we should in a public document try to distinguish between naturally and chemically dispersed oil in the ocean," EPA Deputy Administrator Bob Perciasepe wrote in a July 31 e-mail. "These calculations are extremely rough estimates, yet when they are put into the press -- which we want to happen -- they will take on a life of their own. We should combine these two categories."

That advice was ultimately rejected because "the goal is to show chemical dispersion as part of the federal response to the spill," responded Stephen Hammond, chief of USGS's Emergency Operations Office in the National Geospatial Program, in an e-mail to Perciasepe and others.

Perciasepe defended EPA's suggestion. "I think you are making a mistake on the separate estimates of dispersal, but I have no additional arguments other than it is not verifiable and we will be trying to explain it for the rest of our time on this," he wrote back that night.

Grijalva, for one, is concerned about the outcome of the exchange.

"While there is room for legitimate internal debate about scientific issues, this exchange gives the distinct impression that the White House was more concerned about public image than scientific accuracy in describing the effectiveness of its cleanup efforts," the Arizona Democrat wrote in a separate letter today to Obama.

Perciasepe and EPA scientists are not included as contributors or consultants on the final version of the report. At least one EPA scientist had been listed in an earlier version of the document.

"It is our understanding that EPA scientists were not completely comfortable with the findings in the published report," Brian and Thacker wrote in their letter to the White House.

An EPA spokesman said EPA scientists were not listed among the contributors because they only offered technical feedback and were not involved in developing the report itself.

White House 'pushback'

The internal e-mails also indicate the White House rejected NOAA scientists' suggestions about describing the amount of spilled oil as a range rather than a precise figure.

An early version of the report shows the size of the spill expressed as a range of between 3 million and 5 million barrels. Days later, Mark Miller of NOAA's Office of Response and Restoration wrote, "We have received strong pushback from [the White House] on the cumulative total used in our graphic being more than the official 4.93 million barrels."

Later drafts and the final version of the report use the more precise figure.

"It is vital that the public be able to trust the decisions made by scientific agencies -- and this trust should not be compromised to advance a public relations agenda," the POGO letter says. "The American people need to know the hard facts, including any uncertainties, in order to make informed decisions."

The letter continues: "We are concerned that White House officials may have removed the uncertainty that typically comes with scientific measurements and settled on a more definitive number to make the public feel more comfortable. We feel that the public would be better served by understanding that the government's handling of the Gulf spill is filled with numerous uncertainties."