Survivors of Minneapolis’s bridge collapse said there was little warning. But the warning signs were in place years before.
Long before Wednesday’s accident, which killed at least four and injured around 80, the Minnesota Department of Transportation (MDOT) had warned about problems with the bridge. The structure had a "structurally deficient" rating since 1990 and has been inspected every year since, according to MDOT. Don Dorgan, an MDOT engineer, says, "We thought we had done all we could—obviously something went very wrong." In a 2001 report, engineers wrote that the bridge’s deck truss had many fatigue details on the main truss and floor truss system. Still, the report concluded the bridge “should not have any problems with fatigue cracking in the foreseeable future.” Then, in 2005 and 2006 the bridge was deemed “structurally deficient,” meaning that the bridge doesn’t need to be closed but is in poor condition and ill-equipped to handle current traffic loads. Apparently that still wasn’t reason for urgency. At the time of the dramatic collapse, road crews were working on the bridge, fixing the joints, guardrails and lights. Further inspections of the bridge were ordered, but were only due to be completed some time in the fall. Meanwhile, more than 140,000 drivers commuted across the bridge every day.
A final death toll hasn’t come in yet and dozens of people are still reported missing, but people across the country are already questioning the safety of bridges and highway infrastructure in their own states. Minnesotans have been calling into local radio shows and demanding that the government name the other 460 bridges in the state that local officials deem “structurally deficient.”
But Minnesotans aren’t the only ones worried. On Thursday, U.S. Transportation Secretary Mary E. Peters called on all states to immediately inspect any steel-deck truss bridges similar to the I-35 structure (there are 756 structures in that category nationwide). And the “structurally deficient” or “functionally obsolete” status that the now collapsed bridge carried actually applies to more than 27 percent of all U.S. bridges, according to a 2003 report by the American Society of Civil Engineers. That’s one in three urban bridges and one in every four rural bridges across the country (though the figure is a slight improvement from 28.5 percent in 2000). Still, in 2005 the organization gave the nation a “D” after looking at 12 categories of national infrastructure including rails, bridges, roads and dams. Some recent examples: last month’s eruption of a 100-year-old steam pipe in Manhattan, the failed levees of New Orleans and the 2003 failure of the Silver Lake Dam in Michigan.
The country’s more than 577,000 bridges are of particular concern because they are especially subject to corrosion and because many of them were first built more than 40 or 50 years ago, much like the Interstate 35W bridge, which was constructed in 1967. This was long before Americans became obsessed with SUVs and truckers started carrying heavier loads. “There was huge boom in bridge building after Eisenhower created the Interstate system, the boom lasting from the late '50s to the early '70s,” says Edwin Rossow, professor of civil and environmental engineering at Northwestern University. “Being ‘structurally deficient’ doesn’t mean the bridge is no good. It means many things, including the fact that because of changes in the allowable loads a vehicle can carry and the fact that cars are bigger these days these can all strain an older structure. But all bridges deteriorate over time.”
Tripp Shenton, a professor at the department of civil and environmental engineering at the University of Delaware, explains that a “functionally obsolete” bridge can still work, but it’s layout isn’t effective. “It’s like trying to feed a four-lane highway into a two-lane bridge.” Functionally obsolete can also mean that it fails to meet other traffic standards, such as not leaving enough clearance for trucks on roads that pass below it. Shenton adds that steel bridges, like the Minnesota bridge, are more likely to corrode than other types of bridges.
Certain states have more reason to worry than others. “Clearly the Northeast and a place like Minnesota with severe winters are more susceptible,” says Gregory G. Deierlein, a professor of civil and environmental engineering at Stanford University. “It’s a combination of things—de-icing and just the harsh, freeze-thaw conditions. Water will infiltrate a small crack in the highway; it will then freeze and expand.”
The District of Columbia has the country’s least safe bridges, according to the Federal Highway Administration, with 60 percent rated either structurally deficient or functionally obsolete. Bridges in Massachusetts are also worrisome, with 55 percent falling into those two categories. Rhode Island has problems with 49 percent of its bridges, Hawaii 43 percent and New York 37 percent. Jerry Williams, director of the Rhode Island Department of Transportation, agrees that weather may be a factor. “Obviously, in New England we have the change of season, the salt, the weather, the extreme cold and the extreme heat. We also have infrastructure that is decades old, versus the Southeast, which may be not as old.” Williams says that on Wednesday Rhode Island officials closed a small, two-lane bridge within hours of an inspection that found deterioration.
Minnesota’s governor ordered the inspection of all similar bridges in the state, and federal authorities quickly followed suit, calling on all states to inspect their questionable bridges. Ironically, Minnesota actually has a better track record with bridges than other states, with only 8 percent of its bridges categorized as troublesome. “The science is there to fix these bridges and has been for some time,” says corrosion engineer William Schutt. “But the bridge lasted for 40 years; age is not the factor. Maintenance and the environment are the most obvious culprits.” Schutt says that the use of de-icing salts could play a pivotal role in the safety of a bridge. “If those salts get into the steel or the concrete, the corrosion will be exponential.” If a bridge is only checked every two years, that might not be often enough. “When we test a bridge, one of the things we do is study how much salt has been absorbed into the concrete or steel. One pound of salt per cubic yard of concrete, about 3,500 pounds, can cause serious problems.”
Zdenek Bazant, a professor of civil and environmental engineering at Northwestern University, says he suspects “fatigue” caused the collapse, fatigue simply meaning that repeated loads will weaken structures over time. Because the bridge was older, there were fewer redundancies in the design that would have protected the entire structure from a small crack. “From the video I saw, it looks as if the collapse started near the supports, there may have been a crack at the joint of the first diagonal strut at the support—that’s my biggest suspicion.” But Bazant stresses that officials were not negligent in not closing the bridge. “We would have to close a huge number of bridges in the U.S. if you go by those standards. Many inspections are done visually and you simply can’t see some cracks, especially if they are inside the structure.”
Other famous bridge collapses due to fatigue are rare, he says, pointing to the last big fatigue failure at the Mianus River Bridge in 1983 in Connecticut and, before that, the Point Pleasant Bridge in West Virginia in 1967. “Now, after 24 years we’ve had probably about 50 people killed,” Bazant says. “Yes, we should invest in our infrastructure, but if we’re going to pour tons of money into something to save lives, we should notice that millions die in car accidents, not bridge accidents.”
Yet the Interstate 35W—a steel, arch-truss style bridge with a concrete deck—was deemed safe despite its status as structurally deficient. Northwestern’s Rossow suggests that a variety of factors may have led to its collapse. “It could be a combination of things—the change in temperature due to the summer heat, a scouring of some of the foundation below the water line, or even possibly the work being done on the bridge at the time. They were not making structural changes, but by shifting traffic to one side of the bridge, that changes the load pattern. And there were signs of fatigue from earlier inspections. These could have all come together in some unfortunate way.” One obvious question in the wake of the collapse is whether the definition of “structurally deficient” should be altered to better reflect the danger at hand. “‘Structurally deficient’ can mean a lot of different things,” says civil engineering specialist Shenton. “Perhaps a more specific term might be needed.” Since the term didn’t include the threat of imminent collapse, perhaps he’s right.