National Incident Commander Thad Allen told a presidential commission investigating the BP oil spill on Monday that wildly understated federal estimates of the oil flow "weren't consequential in any of the decision-making I did."
But the co-chairs of the Oil Spill Commission, which is charged with assessing what lessons should be learned from the disaster in the Gulf, weren't buying it.
"It's not entirely clear to me how it could be that flow rate would not affect the response," co-chair and former EPA administrator William K. Reilly said at a mid-day press conference.
"It's a little bit like Custer," chimed in co-chair and former senator Bob Graham. "He underestimated the number of Indians on the other side of the hill and he paid the ultimate price for it."
Indeed, despite administration protestations to the contrary, the federal response when it came to mobilizing skimmers and testing for oil beneath the water's surface -- not to mention getting the well shut -- has been found wanting.
Why the Obama administration chronically lowballed estimates of how much BP oil was spewing into the Gulf of Mexico remains a mystery, however, after a top NOAA official easily sidestepped the commissioners' tepid questioning.
National Oceanic and Atmospheric Administration senior scientist Bill Lehr repeatedly testified that estimating the flow of the spill was a difficult and challenging undertaking.
But he provided no new insights into why, despite much higher estimates by independent scientists, NOAA stuck to BP's estimate of 5,000 barrels of oil per day for five weeks, then slowly increased its estimate, in fits and starts, until finally concluding after the well was capped that the real flow had topped out at a staggering 62,000 barrels per day.
As one of Lehr's fellow panelists, Ian MacDonald, put it: "5,000 barrels was not in the right ballpark, for whatever reason." MacDonald, a Florida State University oceanographer and one of the independent scientists who questioned the federal estimates early on, further noted that "It took a long time to catch up to reality."
Meanwhile, contrary to White House assertions that "the vast majority of the oil is gone," MacDonald warned that more than 50 percent of the oil remains in the Gulf of Mexico in the form of "highly durable material" that is now buried along the coast and on the sea floor.
Colleagues have found signs "that the oil sank from view so it was no longer clearly visible, but formed layers that persist to this day," MacDonald said. One problem with that: "Any storm event tends to resuspend them."
MacDonald also put forward a new possible explanation for why BP's flow estimates were so far off (though not for why the government paid them any heed).
MacDonald said he noticed over the weekend that tables in BP's laughably inadequate regional oil spill response plan include oil thickness estimates as much as 100 times smaller than NOAA guidelines. So using satellite photographs of the slick -- and BP's tables -- to calculate the amount of oil spilled would result in a dramatic understatement.
"A pressing question," MacDonald wrote in his prepared statement, "would be why the Unified Command authorities apparently relied on BP's internal, and evidently erroneous standards instead of using NOAA's guideline that was formulated on the basis of international agreed upon standards."
Graham and Reilly both acknowledged that they still don't understand why government officials were so wrong, and for so long. Graham generally raised the issue of "deference to the industry." Reilly recalled a scribble on a NOAA white board during the early days of the spill indicating that officials immediately realized the leak could be much bigger.
"One of the real, significant sources of public distrust and discontent about the response was the constantly changing numbers about the spill," Reilly told Lehr, who was also point man on the overly rosy late-August "oil budget" report that administration officials then misleadingly cited in a PR campaign to put the spill behind them.
"It was a challenge," Lehr said. "You always tend to estimate conservatively when you're the government."
Why not consider the worst case, instead of the best case, Reilly asked.
"It was a very difficult, challenging process," Lehr replied.
"Having learned from this, I assume next time we'll do better?" Reilly asked.
"I hope so," Lehr responded, unconvincingly.
Dan Froomkin is senior Washington correspondent for the Huffington Post. You can send him an e-mail, bookmark his page; subscribe to his RSS feed, follow him on Twitter, friend him on Facebook, and/or become a fan and get e-mail alerts when he writes.