Robert Bea has an unusual specialty: He studies disasters. As one of the world’s leading experts in catastrophic risk management, the former Shell Oil Co. executive sifts through the wreckage to unravel the chain of events that triggers accidents. The blunt-spoken civil engineer has spent more than a half-century investigating high-profile engineering failures, from the space shuttle Columbia’s horrific end to the explosion of the Deepwater Horizon oil-drilling rig in the Gulf.
A professor emeritus of civil engineering at the University of California, Berkeley, Bea’s disaster autopsy methods — such as looking at the organizational breakdowns that lead to calamities — have been widely adopted. Although policymakers and corporate honchos seek his counsel, sometimes they don’t like what he has to say — witness the flak he took from BP during the Deepwater Horizon probe.
Now in his mid-70s, Bea’s voice is raspier, but his critical faculties are undimmed. On a crisp fall day, he talked with DISCOVER in his comfortable one-story house in Moraga, a leafy suburb east of Berkeley, about what causes catastrophes.
You have said that engineering failures aren’t the chief culprits behind disasters, pointing instead to human and organizational failures — inadequate safety protocols, corporate hierarchies, conflicting egos or just plain laziness. Was there an “aha moment” when this became apparent?
When I was involved in the investigation into the Piper Alpha disaster, when an explosion destroyed an Occidental Petroleum oil-drilling platform in the North Sea, killing 167 men in 1988. The external investigation team that had been hired by Occidental into what caused Piper Alpha found it was a corporate culture that had gone bad, had lost its way.
I was part of that team all the way through the Lord Cullen Commission hearings in London, and I had to listen to one of my friends explain to the Cullen Commission why he and his colleagues had turned off the smoke alarms on the platform because the operating crew was doing a routine maintenance procedure late in the evening. Unfortunately, for over a month, certain alarms had been disabled to prevent unnecessary shutdowns on the rig — in some cases as a response to practical jokes. But turning off the alarms was one of the reasons they got caught by surprise.
Ironically, two years before, I was brought in to advise Occidental on risk management for Piper Alpha because they were having gas releases, pipes were leaking. Of course, you didn’t have to be very smart to say, “Yeah, we’ve got a problem — it’s called rusty pipes. And we’ve got problems with people not doing what they should be doing, and people who don’t understand what’s happening.”
One evening, during the first year of the investigation, I saw spread out on the reception table of the Occidental offices a copy of the London Times newspaper with a great, big, bold headline that said, “Occidental puts profit before safety.”
It had a picture of one of the bandaged, beat-up, horribly scarred survivors from the disaster who was telling this to the newspaper. What this survivor was observing is true. If you don’t have profitability, you don’t have the resources to invest in achieving adequate protection. What the tension is, is having the discipline and the foresight to make those investments before you’re in trouble.
When I came back to Berkeley after the investigation was completed, I realized that for the past 50-some-odd years of my career, I’d been working on 10 percent of the problems. I’d been working on normal engineering things, and 90 percent of the problems are humans and/or organizations.
We often have ample warnings before catastrophes hit, but we tend to ignore them until it’s too late. Why?
The problem is attention span, particularly in this country because we are a pretty young country. Our knowledge of history is very limited. We are extremely blessed. Lots of good things attract our attention. It’s a noisy environment, really noisy. It’s unusual to find people who are comfortable sitting in a room by themselves thinking.
You could say the eruption of Mount St. Helens was certainly painful, but it actually affected relatively few people and then disappeared into that strong noise environment. At that point people say, “Well, it’s never happened to me.
I can’t even remember my parents talking about it, and I’ve got these new things to play with, and they require attention,” like Facebook and Twitter. And suddenly, we have flitted from something that is difficult and painful to think about back to something that is enjoyable.
You seem to be suggesting that people have trouble dealing with issues over the long term. Are there other examples?
Well, global climate change is a perfect one, or rising sea levels. It’s happening slowly. People love living by the beach, so they build a beautiful home on a concrete slab, on top of the sand a few feet above sea level, and [ignore the fact] that the sea level is [rising]. So thinking about these slowly evolving long-term things, it is painful. It says, “Well, I might have to move my home. I really enjoy the beach,” and we don’t like to give those things up.
Is this inability to think long term also true of organizations — corporations or government agencies?
Yes. The equation for disaster is A + B = C. A is natural hazards, things like hurricanes, gases and liquids under pressure that are extremely volatile. They’re volcanoes. They’re tsunamis. They’re natural, and there’s nothing unusual about them. B is organizational hazards: people and their hubris, their arrogance, their greed. The real killer is our indolence.
So human error is the kindling that escalates a natural hazard — a hurricane, a tsunami, chemicals under pressure — into C, a catastrophic disaster. Can you give me some examples?
Hurricane Ike. Galveston, Texas, got completely wiped out in 1900. Thousands of people got killed. So the U.S. Army Corps of Engineers built a seawall on Galveston Island, and that sucker has gone through every major hurricane since 1900.
But people think that if a storm hasn’t happened since they lived there, somehow it can’t happen to them. This is where B comes in — the hubris and shortsightedness. Because a hurricane hadn’t flattened the city in decades, civic leaders decided to let people build at sea level again. And when Hurricane Ike came through in 2008, it was just like Berlin at the end of the second world war. Everything was gone.
Before Superstorm Sandy, I wrote that the subways were going to flood, but no one did anything. Mayor Bloomberg even hired some of my engineer friends from the Netherlands to come to New York City and advise him about building gates to cut off incoming hurricane surges.
But here we’re back to B — hubris and shortsightedness. People think because they’ve never seen a storm like what happened in New Jersey or they’ve never seen the tunnels flooded in New York City that it can’t happen, or that they need to think about building a levee.
When I lived in New Orleans, we lost everything in Hurricane Betsy [in 1965]: our house, wedding photographs, marriage license, birth certificates. Yet 40 years later, after Katrina, I go back to the same place. There’s a new home built on the foundation, and the owners are dragging wet, oily mattresses out the front door.
Luckily I had no one with me that morning, but I broke down and cried. It wasn’t tears of sadness. It was tears of frustration at such a miserable, despicable mess. While we can’t prevent disaster, we can do things that are more sensible to mitigate risks, like maybe not building homes in floodplains.
But the cities are already there. Are you going to move entire cities?
In some cases, yes. We did it in the Mississippi River Valley after the 1993 floods. We actually moved entire towns to higher ground, like Valmeyer, Ill., and Rhineland, Mo., because we suddenly recognized they’d rebuilt them five times in the same damn place. Doing it six times doesn’t quite make sense. But there is not a “one size fits all” answer.
In other cases, there are intermediate solutions that can work, such as occupying only what you can defend properly and in a sustainable manner. An example is the “new New Orleans,” where parts of the city outside of the defended perimeter of the levee system can be expected to flood severely and frequently. Individuals there are building structures on higher ground and making them stronger, and preparing to take care of themselves in future storms.
But even after Superstorm Sandy’s devastation, you can’t just completely rebuild Lower Manhattan.
No, you can’t. But you can follow the example of the massive Thames Barrier that has been built in London [10 steel gates that prevent the city from being flooded by tidal surges].
But that cost over 500 million pounds — about $850 million — when it was completed in 1982. Doing the same in Manhattan would cost up to $17 billion and an additional $10 billion to $12 billion to shore up areas next to the barriers, an astonishing amount of money.
Well, do you want to fix it now or fix it later, when it will cost 100 times as much? Damages from Superstorm Sandy in New York and New Jersey are estimated at $60 billion to $70 billion. The key question I always ask is, “Can we work this problem out in a responsible way, or do we wait until it fails — in other words, will we fix it now versus pick up the pieces later?”
We looked into this “pay me now or pay me later,” and in many cases, it’s more than 100 times the cost to fix afterward. We economically documented several major accidents where the factor is bigger than 1,000, like Katrina.
People regularly ignore risks, but isn’t it the extreme scenario — the thing that has the 1 percent chance of happening, the so-called long tail at each end of the bell curve — that causes all the trouble? What about events beyond the realm of normal expectations, like Superstorm Sandy?
Sandy is better classified as a “predictable surprise.” There were some groups in the greater New York area that had a clear understanding of the potential for significant flooding due to an intense storm. But there were other groups who had no idea and made no significant efforts to learn how vulnerable the city was to storm surges from the Atlantic.
What do you do when it becomes more costly to prepare for every disaster than just to take the risk? How safe is safe enough?
In many cases, you can’t prepare because no one’s willing to spend money to do this. We’ve rebuilt the levees around New Orleans, for example, but they’re right back on the same slabs. And it’s going to cost about 10 percent of the construction costs per year to maintain them.
Now where are you going to get $1.5 billion a year? So they can’t maintain it, and the next time it is challenged — and there is no doubt in my mind it will be challenged severely — it’s going to fail again, and the consequences are going to be worse because we’ve allowed more population and more infrastructure.
In aviation, which has unequaled safety records, they do predict and plan for the worst case because they can’t afford accidents. But Sully Sullenberger’s “Miracle on the Hudson,” when the pilot landed the plane on the Hudson River after flying into a flock of geese, does seem like a miracle.
It was a miracle, but it was not an accident. It was, in fact, rehearsed. That’s how the FAA was able to clear air traffic control, how he and his co-pilot, together with the flight crew, were able to get almost everything right, how the French Airbus had those backflow valves already designed in the fuselage.
We don’t normally land airplanes in water. They’re supposed to be on land, but they designed it to land in water as well, and the crew had rehearsed and planned what they would do in the event of a water landing. They are thinking about the impossible.
What warnings are going unheeded now? The antiquated levee system protecting California’s Sacramento Delta, which is the source of fresh water for 28 million of the state’s residents, comes to mind. I know you’ve been studying this.
An earthquake, a megaflood. Any of these natural disasters could rupture the delta levees and take a lot of the infrastructure — power lines, communication networks, gas pipelines, hydroelectric power systems — with it. Millions could be without power or fresh water for months. It isn’t going to be pretty, and it will knock out of commission the ninth-largest economy in the world.
And it could happen at any time.
Yeah. Tick, tick, tick.
Bea goes back in time to explain how we might have avoided notable catastrophes of the past.
Deepwater Horizon, 2010
The Deepwater Horizon oil-drilling rig exploded in the Gulf of Mexico, killing 11 crewmen and igniting a fire that could not be extinguished. Two days later the rig sank, leaving the well gushing in the seabed. It ultimately leaked 210 million gallons of oil in the largest offshore spill in U.S. history.
» What was ignored? BP, the industry and federal regulators understood the potential for an uncontrolled blowout in Deepwater drilling, but they failed to heed warnings about the structural weaknesses in the cement casings that protect the well pipes. Plus, the blowout preventer hadn’t been working properly for several days.
» What would have improved the outcome?Delivering the required degree of safety consistently and ensuring that protocols were followed. Understanding that wells pumping 162,000 barrels of oil a day are under much higher pressure, and therefore are much more dangerous, than wells delivering 500 barrels a day. Building stronger protective structures to withstand these intense pressures could have prevented the disaster.
Midwest Floods, 2008
Months of rainstorms led to heavy flooding in seven Midwestern states, including Illinois, Minnesota, Indiana and Missouri, resulting in 24 deaths and more than $6 billion in damage.
» What was ignored? The lessons from floods in 1993. The increasing fragility of the aging levee system, and more people building in low-lying areas susceptible to flooding.
» What would have improved the outcome?Giving the water the room it needed to flood and not build in those areas. Building and maintaining high-quality flood protection systems in the areas that could have been protected.
Columbia Disaster, 2003
The space shuttle disintegrated upon re-entry into the Earth’s atmosphere due to missing heat shield tiles, killing all seven astronauts on board.
» What was ignored?The mantra “better, faster, cheaper” drove the management decision to “bring the bird back,” even though the U.S. Air Force had photographs showing the leading edge of the Columbia was missing heat shield tiles, which were damaged during the launch. Problems with heat shield tiles during earlier missions hadn’t been adequately addressed, and NASA ignored engineers’ requests to ground the mission until these problems were solved.
» What would have improved the outcome? Fixing heat shield tile problems uncovered during earlier missions. Develop a backup plan for fixing the shuttle in space if it is damaged and ensuring the crew’s safety. They were lucky with previous missions that had missing tiles, but they took a chance, and their luck ran out.
Exxon Valdez Crash, 1989
The Exxon Valdez oil tanker ran aground in Prince William Sound, spilling an estimated 11 million gallons of crude oil along Alaskan shores. Numerous factors complicated cleanup, including the size of the spill, the remote location and a lack of readily available equipment and effective chemical dispersants to dissolve the thick oil.
» What was ignored?The dangers of Bligh Reef, where the ship hit the rocks. Taking shortcuts to save time to get the oil to Southern California refineries. Not traveling in regulated shipping lanes.
» What would have improved the outcome? Better communication between vessel captains and traffic control centers to avoid treacherous conditions. True pollution control to improve visibility, and better preparation for cleanup.