“THE room flew apart in a split second. Everything came out of the ceiling, and concrete slabs popped out of the floor as the building torqued.” Brian Clark is recalling his experience of a plane hitting the south-east corner of the World Trade Center’s south tower just three floors below where he was standing. The impact twisted the 110-storey building anticlockwise – and then, for 10 seconds, Clark recalls, things got really frightening.
“I sensed that the building was swaying one way – a long way – towards the Hudson River,” he says. “The sensation was that the building went 6 to 8 feet, stopped, and then came back to vertical.” From then on, he says, he felt in control. What he didn’t know at the time was that he only had 56 minutes to find a passage down from the 84th floor before the tower collapsed.
Clark found the only stairwell that was not destroyed by the impact and got out alive. He was one of the lucky ones: more than 2800 people died at the World Trade Center on 9/11, including more than 350 firefighters and other emergency workers. Most were killed when the structures collapsed before they could escape or be rescued. Which makes it all the more astonishing that four-and-a-half years after the event, and with new skyscrapers springing up in many of the world’s cities, controversy is still raging over why the towers collapsed.
Advertisement
On one side of the debate is the US National Institute of Standards and Technology, which was charged with investigating the tragedy. Its final report, published last October, is based on the work of 200 specialist investigators. NIST reviewed tens of thousands of pages of documents, conducted more than 1000 confidential interviews, analysed wreckage and used three mathematical models to simulate what happened between impact and collapse. It also had access to more than 150 hours of video and nearly 7000 photographs.
NIST calls its report “the most exhaustive examination of a building failure in history”. It says that the initial collisions damaged exterior walls, floors and columns at the core of each building. The fires that then spread through the towers caused the floors to bow, where the fireproofing had been damaged. This eventually led to collapse as the columns were pulled inward by the sagging floor, concludes Shyam Sunder, who led the NIST investigation.
NIST insists that the twin towers would probably have stayed up if the aircraft impacts had not dislodged the fireproof coating that protected the steel structure. Yet experts at leading engineering firm Arup challenge that view. They worry that since 9/11, engineers and architects have been concentrating too much on improving the fireproofing, whereas Arup believes there are more pressing issues to deal with. “We do support the concept of fireproofing,” says Barbara Lane, a fire specialist at Arup who has helped to review seven planned high-rise buildings in London, and others in China. “But there are more robust changes you can make to a structure that will last much longer and be more resilient than just making fireproofing thicker.”
Those more robust changes are concerned with the joints between beams and columns which expand when heated by flames and can spell disaster. Arup points out that engineers have traditionally ignored the complex ways that heat causes steel to expand and that this can create distortions in the structure. Instead they have concentrated on protecting the structure by insulating steel with fireproof coatings.
Arup’s concerns about the NIST report arose from the findings of a panel of its in-house experts, set up to address concerns about other potential high-rise terrorist targets four months after 9/11. According to Arup, steel beams longer than 10 metres can generate significant strains on a structure unless the joints between the beams are designed to accommodate thermal expansion. In the twin towers, the longest beams stretched nearly 20 metres. “You need to understand what heat does to your structure and design,” Lane says. “Long spans expand more and your structure has to withstand that expansion.”
When NIST invited comments on its draft final report in July 2005, Arup criticised the investigation for underplaying the role of thermal expansion. NIST only considered its effects on exposed beams, rather than all the building materials. And, Arup says, the institute did not pay enough attention to the fact that significant thermal expansion can occur even at modest temperature increases.
Arup is adamant that the fires on 9/11 were so intense that the towers would probably have fallen down irrespective of fireproofing. It’s not a criticism of the building’s engineering, Lane insists; this understanding of how to deal with the structural effects of fires has only developed over the past 10 years, well after the completion of the twin towers. “The World Trade Center was designed to the state-of-the-art at the time,” she says.
Nevertheless, NIST is sticking by its conclusions. In a prepared statement, Sunder says: “NIST is confident in its determination of the probable collapse sequences for the towers.” The institute’s final report re-emphasises this view, though it does say that engineers should begin to consider the influence of heat on their new designs and re-examine their old ones.
Last October, NIST presented its report into the twin towers’ collapse to the US Congress along with recommendations to improve the safety of high-rise buildings. Among them are improvements to the way buildings are evacuated. In the past, building designers have tended to assume that you would only need to evacuate one floor at a time. If a fire breaks out, for instance, you would evacuate that floor and perhaps those immediately above and below. Today, full evacuation of all floors is the order of the day. “There might be several instances such as power outages and earthquakes where full-building evacuation might be required,” says Sunder.
Evacuating an entire skyscraper is no mean feat, however. When the terrorist attacks on the twin towers began, there were an estimated 17,400 people working in the buildings, about one-third of their capacity. Accounts from survivors such as Brian Clark offer a unique opportunity to understand how people behave in an emergency. “You can’t really simulate this stuff in the laboratory,” says David Canter, a psychologist at the University of Liverpool in the UK and one of the leaders of a three-year project, starting this month, to interview survivors (see “Model behaviour”). “The World Trade Center evacuation is so important because it is, in fact, hundreds of different case studies.”
About 87 per cent of people in the twin towers survived. They escaped from many levels, each with their own unique layouts housing a variety of different organisations that had their own procedures for dealing with an emergency. “It gives us the basis for exploring in much more detail all the patterns of behaviour,” says Canter.
Whatever the evacuation procedure, researchers agree that the height of the building makes a difference to your chance of survival – a difference that has so far been ignored, Sunder points out. “We do believe that tall buildings pose a greater risk to occupants than short ones,” he says. And for the most part our code requirements have been indifferent to the height of the building.”
It is certainly Arup’s belief that not enough has been done to reassure occupants of high-rise buildings in the wake of 9/11. Since the event, there has been a “general unease” among property developers, according to Lane, who has found property developers increasingly wanting to know the risks of living and working in high-rise buildings. Such reviews aim to make the worst conceivable emergencies survivable.
In practice, this means measures such as better CCTV, audio systems in stairwells, more robust elevators that can work even during a fire, and a greater appreciation of the physical effects of fire on a structure (see Graphic).
It’s likely, however, that the unease will persist for some time yet. Engineers in the US will only start revising minimum building standards next month at the earliest – four-and-a-half years after 9/11. Lane welcomes the promise of better minimum building standards as far as they go. But that’s not enough, she believes. “It’s not about changing the fundamental way buildings are designed,” she says. “It’s about using science and some logic to show that making emergencies survivable is a reasonably easy thing to do with some proper management and proper systems.”
Model behaviour
Brian Clark’s first-hand testimony of his escape from the blazing tower is about to become the focus of unprecedented research on the way people behave in emergencies. Later this month, Clark and up to 2000 fellow survivors, all with vivid accounts of what went on inside the twin towers, will begin to recall afresh their experiences of the disaster for the benefit of fire investigators and psychologists.
Their accounts will help refine software that models the evacuation of buildings. By adding more realistic human behaviour, mathematical modellers hope to show architects and engineers how they can make escaping from buildings safer and faster. With no consensus yet on what caused the towers to collapse, and hence no sure way to prevent a repeat, evacuation modelling is crucial.
The three-year project to interview survivors is being coordinated by Ed Galea, a mathematical modeller at the University of Greenwich in London. His software for modelling the evacuation of buildings and other places, such as ships and planes, is called Exodus and was one of the models the US National Institute of Standards and Technology used to simulate events inside the twin towers. “My attitude is that we actually owe it to the people who died in the building to learn as much as possible from the evacuation,” says David Canter, director for the Centre of Investigative Psychology at the University of Liverpool and the project’s lead psychologist.
Exodus includes not just peoples’ movements, but also certain human behavioural elements, such as deciding whether it is safe to walk through smoke. But Galea admits the model has its primitive side. “In the World Trade Center evacuation, a lot of people appeared to react and move in groups,” says Galea. “We don’t really understand enough about what motivates people to form groups in the first place and what causes groups to fragment.”
Galea hopes that the survivor interviews will allow him to understand human behaviour better and thereby improve his model. But his ultimate vision is for computers to monitor emergencies as they unfold and provide real-time advice. Buildings could be wired with heat sensors and strain gauges to monitor and signal changes, he suggests. Sending such information into a computer model could prove invaluable to security staff working in a dedicated control centre nearby. “They could receive live advice on how to manage an evacuation,” he predicts. Eventually with the help of supercomputers the models could help to manage evacuations across huge areas, such as the London underground or even entire cities.
