Master of disaster reflects on Deepwater Horizon's enduring lessons

At some point in the next week, BP will likely initiate the "bottom kill" procedure that permanently plugs the Macondo well, bringing to an end the worst maritime oil spill in American history. No more 24/7 video of oil gushing into the Gulf of Mexico. No more weekly tutorials on the intricacies of deepwater oil drilling. No more sludge cloud shadowing the Obama administration's every move in the 2010 summer of discontent. Now only the clean-up and long-term repercussions remain to sort out.

Perhaps no one has a better first-hand grasp of the Deepwater Horizon disaster than retired Adm. Thad Allen, the national incident commander who also coordinated the federal response to Hurricane Katrina in 2005. Recently, National Journal spoke with Allen about lessons learned from the crisis and federal response, and how they might affect future policy. Edited excerpts from that interview follow.

NJ: How do you respond to critics who say the federal government's response to the Deepwater Horizon disaster was too slow given the magnitude of the problem?

Allen: Look at the actual timeline. The explosion on the Deepwater Horizon occurred on April 20th. As commandant of the Coast Guard, I got a call just before midnight that there was an uncontrolled fire on a rig in the gulf, with an unknown number of people killed and injured. That night the Coast Guard evacuated a lot of people from the site of the explosion, and we launched a two-day search for the 11 workers who were never found, even as we moved lots of equipment towards the site. Then, early in the afternoon on April 22nd, the entire rig collapsed and sunk. Hours after the rig sunk, I was in the Oval Office along with [Homeland Security Secretary Janet] Napolitano, briefing President Obama on our initial response. So I don't buy the argument that we were slow in responding. I certainly didn't lean back in the saddle.

NJ: Did you immediately understand the severity of the crisis?

Allen: As events unfolded, the enormity of the problem started revealing itself. We weren't dealing with a single, monolithic oil slick like the 11 million gallons that spilled from the Exxon Valdez. This was an uncontrolled discharge, with 53,000 barrels each day spewing in different directions depending on the prevailing winds and currents, creating hundreds of thousands of separate oil slicks. The United States had never dealt with that situation before. Very quickly we were forced to spread our assets from the southern Louisiana coast to the Florida panhandle. That's when we realized that the required response was going to dwarf what was anticipated in BP's response plan.

NJ: Why did response plans seem so outdated and inadequate to the magnitude of the crisis?

Allen: Basically because oil spill response is all predicated on the lessons of the 1989 Exxon Valdez disaster. The legislation that came out of that disaster focused on tanker safety and phasing out single-hull oil tankers, on making sure the party responsible for the disaster meets its liability requirements, and on cleanup as directed by the Oil Pollution Act. That was the regulatory scheme established for responding to oil spills.

However, in the 10 years after that accident, while we were primarily focused on the safety of tankers and the Alaska pipeline, oil drilling was moving offshore and going deeper underwater. So the technology changed, and the overall response structure didn't keep pace with those changes and the emerging threat. You could say the same thing about Coast Guard inspection regimes, which we are in the process of rethinking. Right now, for instance, the Coast Guard is not required to approve a company's oil spill response plan, because that goes through the Minerals Management Service. I suspect that will change in the future.

NJ: Given that BP seemed so culpable in causing the disaster, did it make sense that the company also had such a prominent -- some would say dominant -- role in the cleanup effort?

Allen: Well, in the regulatory regime created after the Exxon Valdez, BP was the "responsible party" in both statute and regulation, which meant that it had to bear the costs associated with the spill. For that to happen, however, we had to bring them into the command structure to write the checks for everything from boom to catering. As the "responsible party," BP was also required to have contractors in place to clean up the spill, while the government had oversight over that operation. The public didn't understand that arrangement very well. The notion of BP having such a key role in the response after seeming to cause the problem understandably didn't sit well, and that relationship was tough to manage. BP had divided loyalties, so to speak. It was responsible to the public for the cleanup, but at the same time it had a fiduciary responsibility to its shareholders.

NJ: Do you think that divided responsibility should be addressed?

Allen: Well, I think we need to take a very hard look at the role of the "responsible party" in the command and control of a cleanup operation after an oil spill. You need someone in the command post to represent the oil industry, but it might be better if they didn't have a fiduciary connection to a specific corporation. BP might have taken the resources needed for the cleanup and put them into a blind trust, for instance, that was administered by a trustee who actually writes the checks. That might mitigate the appearance of a conflict of interest in the public's mind. Ultimately, we need to decide what we really mean by "responsible party" in these types of situations. It's a very interesting public policy question.

NJ: Do you think it's a problem that the oil industry has a monopoly on the technologies involved in deep-sea drilling and oil-well capping?

Allen: By law, the oil companies had to essentially create a capability in the private sector to respond to oil spills after the Exxon Valdez. The decision was made by government to rely on private contractors. As you point out, that reliance was most acute at the wellhead, which was five miles below the surface of the ocean. There is no government in the world that owns the means to do deep-sea drilling. Neither the Navy nor the Coast Guard had anything like that capability. The technology was entirely in the hands of private companies, so the government's role at that point became one of oversight. An overarching question as we look to the future is whether that capability should be solely in the hands of the private sector, or do you want some measure of that capability in the public sector so that the government can mount an immediate response?

NJ: Doesn't that question seem all the more important given how little time and energy BP spent in preparing an adequate spill response?

Allen: One problem we ran into was that during normal operations, all of the oil produced in the gulf is shipped back to shore via pipelines. When we had to bring oil to the surface after the accident, there was no obvious way to transport or collect it. To make that happen, BP had to bring a floating production system from the North Sea that uses tankers to shuttle the oil to shore. To bring the oil to the surface, we brought in freestanding, floating pipes called "risers" that are used off the shore of Angola. So our solution amounted to the North Sea meets Angola in the Gulf of Mexico. Lashing all that together took 85 days, because none of it had been put together that way in the past. So one lesson we learned is the need for a system like that on day one, rather than on day 85. The oil companies are already thinking hard about such a system.

NJ: As was the case with Hurricane Katrina, there seemed to be significant tensions, disconnects and finger-pointing between federal, state and local authorities. Is that inevitable in trying to mount "whole of government" responses to far-reaching disasters?

Allen: I think these efforts will always be, in some ways, unique and a work in progress. Any time there is a gap between what local officials want and what they see being done on the federal level, there's going to be pointed discussions about the best way forward. And to paraphrase Tip O'Neill, all oil spills are local. They manifest themselves differently in different places, depending in part on varying types of local government and political structures. I'm there to provide unity of effort, for instance, and the law assumes I interface with state officials, who in turn interact with their local officials. In places where you have more autonomous home rule, such as Louisiana's parishes, however, the challenge of smoothly integrating federal, state and local responses is greater. We also ran into the problem that some of the affected areas along Louisiana's coast were really isolated and difficult to get to, and that only added to the complexity of the operation.

NJ: Would you change methods for estimating the scope of an oil spill, especially in light of widespread suspicions that BP and the government underestimated the amount of oil dispersed into the gulf?

Allen: I think for any future oil spills we should rely only on official government estimates based on the findings of an independent team of scientists. That was ultimately the solution we adopted. There was so much angst over how much oil was spilling that I created a flow-rate technology team of scientists led by the head of the U.S. Geological Survey. They estimated that the well was spilling 53,000 barrels a day into the gulf, plus or minus 10 percent. That's how we came up with the top-line figure of 4.9 million barrels. That's a lot of oil.

NJ: Is it enough oil to cause you personally to question the wisdom of deepwater drilling?

Allen: Whenever I'm asked that question, my reply is the same: That's way above my pay grade. I will say that in this case we had a "fail-safe" system that turned out not to be fail-safe. So if we are going to continue to allow drilling at 5,000 feet below the ocean's surface, on a seabed that only robots can reach and where operations resemble Apollo 13 more than a standard oil drilling operation, then we had certainly better know how to deal with another failure if it were to occur.

NJ: You've had a direct hand in responding to devastating crises ranging from the 9/11 attacks and Hurricane Katrina to the earthquake in Haiti and the gulf oil spill. Have you drawn any overriding lessons about the nature of government responses to such destructive incidents?

Allen: When considering future responses to big events like these, I think we will have to decide on a social contract that spells out what citizens can expect from their government. Because the universe of potential interventions, and the expectations of the citizenry, are both growing in ways that outstrip traditional funding sources and statutory guidelines. For instance, what's the government's responsibility for dealing with the long-term socioeconomic and behavioral health impacts of these events? Nowhere in government statute or regulations will you find guidance on how to deal with those kinds of issues. I don't know if a whole society can acquire post-traumatic stress disorder, but you definitely see disaster fatigue set in after these major events. You can see it in the gulf region right now. So we as a nation are ultimately going to have to deal with the public policy issues raised by these big national traumas.