Company disputes government figures but has fought release of its own data on how much oil leaked into Gulf of Mexico in 2010

The unveiling of the first criminal charges in the Gulf of Mexico disaster could force BP to disclose a closely guarded secret – its internal estimate of how much oil actually gushed out of its stricken well.

It's quite literally a billion-dollar question. The justice department, which announced the charges on Tuesday, against a former BP engineer, is also suing the oil company for damages in a civil case.

Those fines under the Clean Water Act will be decided by the amount of oil that flowed into the Gulf, up to $4,300 per barrel if the release is the result of gross negligence. By the government's account, which estimated the well released more than 4m barrels of oil before it was brought under control, that could mean penalties as high as $17.6bn.

But BP has always disputed the government figures, and those of independent scientists. It has also fought in court to keep its own internal estimates of the flow rate a secret.

Now the affidavit released on Tuesday suggest that BP knew more oil was coming out of the well in the early days after the explosion on 20 April 2010 than it was reporting to the federal government or the public.

The discrepancy could have sweeping legal implications for the oil company in civil and criminal proceedings arising from the Gulf of Mexico disaster.

A day after the explosion, Kurt Mix, the former engineer charged on Tuesday, began modelling the potential flow rate from the BP well, according to the affidavit. He shared his estimates with an unnamed supervisor, suggesting the well was gushing between 64,000 and 138,000 barrels of oil a day.

At the time, however, BP and the coast guard were telling the public there was as little as 1,000 barrels of oil coming out of the well. BP gradually raised its estimates in the days and weeks before the well was finally brought under control in July 2010.

However, the oil company refused at the time to even discuss how much oil was coming out of the well, claiming that it was a distraction from efforts to control the well.

The federal government adopted a similar position – much to the frustration of environmentalists and scientists. "The flow rate has never impacted the response," BP America's chief operating officer, Doug Suttles, told the New Orleans Times Picayune in June 2010. He went on to say the flow rate was "irrelevant".

The oil company and the federal government initially claimed it was impossible to arrive at an accurate estimate of the spill. However, independent scientists came up with a 70,000 barrel a day flow rate in May 2010 that turned out to remarkably close to the federal government's final estimate.

However, the affidavit released on Tuesday suggests that the flow rate was crucial to the success of BP's efforts to stop the well, with a procedure known as Top Kill. At a time when BP executives insisted Top Kill had a 70% chance of success, Mix and other engineers were privately warning that the procedure had little chance of working if the flow rate was more than 15,000 barrels a day. "Too much flow rate – over 15,000 and too large an orifice," Mix warned in a text to a supervisor.

In public, though, BP officials continued to say Top Kill was going according to plan for another two days.

Those arguments are set to continue in a federal court in New Orleans on Wednesday, where a judge is reviewing the $7.8bn settlement reached with 100,000 individuals for economic and medical claims.

A few hours after Mix's arrest, the justice department filed papers demanding BP produce its internal estimates of the flow rate, The Times-Picayune reported.

"It appears that BP intends to argue that the United States internal flow rate work should be produced while BP's should be protected," the justice department wrote. "That position is neither fair nor grounded in the law."

(By Suzanne Goldenberg, Guardian.co.uk, 25/04/2012)