A Sea in Flames - Carl Safina [134]
“You can do Monday-morning quarterbacking,” Lubchenco says to me a bit pointedly, “but at the time there was extraordinary effort. We mobilized ships and planes and people, assuming that it would be really bad.”
“How,” I posit, “is it possible to assume it’s really bad when you kept putting out flow estimates that were one-sixtieth of the—”
Allen interrupts: “I was irritated over the early declarations about whether this was one or five thousand barrels. I listened to that and I thought, ‘This is crazy—nobody knows.’ ”
Lubchenco wants me to appreciate that “for weeks and weeks, the primary focus was stopping the flow.” She adds, “We all thought the early estimates were very low, but we didn’t have a way of getting better estimates initially. In every other spill, the amount of oil was estimated from the surface. But this time a lot of oil was staying deep. We were really frustrated early on because BP said, ‘No, we don’t want anybody else down there.’ ”
“They were just focused on stopping the oil,” says Allen. “That’s when it’s our job to say, ‘Here’s our national priorities; here’s how we’re gonna do it.’ With the amount of remotely operated vehicles moving around down there, it is stunning we didn’t have a major accident. We had ROVs bump into one other and knock things off; that happened twice, almost with major consequences. It’s like the eight near midair plane collisions. There was an unbelievable amount of stuff going on within one square mile; that never happened before. The area where oil was coming to the surface sometimes had thirty-five vessels all in one square mile. But at one point,” he adds, “I said, ‘We’re gonna establish an independent estimate of flow rate.’ ”
Lubchenco continues: “Thad ordered them to let the group go down and make those measurements.”
Allen takes another sip of his coffee and glances at his watch. I know we don’t have a ton of time left. He poses the next question himself: “Could we have stopped it sooner? In hindsight, we might have saved two to three weeks. But we were operating with an abundance of caution.”
“We had very good reason to believe,” Lubchenco informs me, “that the well had been damaged. The concern was, if we built too much pressure in it, the oil would start breaking out through the seafloor. And then—it would be completely uncontrollable.”
“That’s the Armageddon we all feared,” Allen adds. “So we had eighty-five days of a different spill coming to the surface in a different way in a different place every day, depending on winds and current conditions. We had a hundred thousand different patches of oil from Louisiana to Florida. Because the oil spill contingency plan didn’t call for enough equipment, we were behind the power curve for six or seven weeks. We had three kinds of responses to the oil: skim it, burn it, or disperse it.”
I point out that all of the oil was coming out of one pipe: “You had your hands around it right there at the source, and you let it get away.”
“We just did not have the ability to capture it at the source,” Lubchenco replies. “And we didn’t have the boats, skimming equipment, right weather—or the kind of oil—to let us just put a ring around it and capture it at the surface. It sounds like that should have been easy,” she says. “But it was not.”
“When the Deepwater Horizon’s pipe broke,” Allen affirms, “there was nothing capable of capturing that oil. A system should have been in place, but it wasn’t, because we were focused on tanker spills. Oil spill response in this country is based on Exxon Valdez.”
“That’s painfully obvious,” Lubchenco concurs.
“The assumption was, ‘We’ll never have a failed blowout preventer.’ ”
But we know blowouts happen, I insist. What they finally installed on July 15 to stop the leak, they should have already had in a warehouse before April 20.
“I couldn’t agree more,” Lubchenco affirms.
“We’re not gonna sit here and defend the fact that it wasn’t there,” Allen says.
Of the kinds of things that were there, Lubchenco says that she hadn’t anticipated how much their response options depended