By Jenny Neyman
Alaska’s bridges aren’t paved in gold, nor enhanced with any other such mythical-sounding properties, but they are reinforced with a powerful silver lining that, 50 years ago, was as nonexistent as leprechauns and unicorns.
When the 9.2-magnitude March 27, 1964, Good Friday earthquake struck Southcentral Alaska, earthquake engineering was given little thought, when it was thought of at all. Code requirements were lacking, research was even more lacking and as a result, even best practices weren’t very good. Those faults were dramatically demonstrated in the destruction caused when the fault between the Pacific and North American plates ruptured near College Fjord in Prince William Sound. Bridges, buildings and docks collapsed, roads cracked, pipes and other infrastructure ruptured. On the central Kenai Peninsula, bridge failure was the biggest damage sustained, with the collapse of the Kenai River bridge at Kenai Lake in Cooper Landing. Communities throughout Southcentral, and as far away as California and Hawaii, sustained damages.
But the resulting shockwaves generated in the engineering community were even farther reaching, and longer lasting.
The second-largest quake in recorded history, blamed for 139 deaths and $311 million in damages (equivalent to $2.28 billion today), was not a welcomed event, but presented a valuable boost to science. Today, better buildings, better bridges and better infrastructure have been built in large part because of the destruction of the Good Friday quake.
“There certainly have been vast improvements made to earthquake engineering, many of these changes initiated as a result of the 1964 earthquake,” said Elmer Marx, senior bridge design engineer for the Alaska Department of Transportation and Public Facilities. “This probably was the event that really kicked it off. There was earthquake engineering before that but it was an afterthought, at best.”
That changed as dramatically as did the landscape of downtown Anchorage in the quake, though not as immediately. In the aftermath, researchers flocked to Alaska to study the damages, using that information to revise engineering standards. Alaska has since become a leader in earthquake research and design, contributing to the standards now used throughout the country and elsewhere in the world, but the new thinking wasn’t always quick to be adopted.
“There was progress being made but they were guidelines, not a real, live code, but suggested codes that might someday become codes. Some states used those guidelines because it was good practice, but a lot of agencies didn’t want a code. More were like, ‘If we want to do it, we’ll do it. If we don’t, we won’t,’” Marx said.
It took shakeups in California to really get the ball rolling toward large-scale, consistent change, particularly the 1971, 6.6-magnitude San Fernando quake.
“We have more and more violent earthquakes than they have in California, but as they have orders of magnitude more people than we have, and they’re everywhere, the least little earthquake hits any part of their state, you’re bound to hit somebody,” Marx said.
By 1983 earthquakes became a mandatory design consideration for all bridges. In 2009 a significant update to the seismic provisions for earthquake engineering for bridges went into effect.
“Alaska helped prepare some of these provisions, that’s the document we currently use,” Marx said.
So what, exactly, has changed? Why are the 1,000 or so bridges built into Alaska’s highways much safer today than they were in 1964? Two reasons. One is a matter of philosophy, and where the rubber meets the road, it’s due to very detailed advances in engineering.
“In pre-1964 engineering, earthquakes were hardly even thought of. When it was it was treated in a very simplistic manner,” Marx said.
The wisdom of the day was to build a bridge to withstand the lateral — sideways — force exerted on a bridge during the shaking in an earthquake. Common practice was to build to withstand a mere 2 to 6 percent of a bridge’s weight in lateral load.
“Well, it turns out, as we collect more information, we know that those numbers are very, very, very low. We might be getting a value closer to 100 percent of the weight pushing sideways. We need to very greatly increase that sideway, lateral loading,” Marx said.
However, it is not easy to design a bridge, or any structure, to resist that kind of load.
“The bigger you make that component the more force it attracts from the earthquake, so you’re chasing your tail. You make it bigger, it pushes harder,” Marx said.
Therein comes the philosophy.
“After 1964, some engineers of the day that were cutting edge said the way to resolve that force is to give instead of trying to fight back — bend but don’t break. And that’s precisely the philosophy we use today. We design the columns that support the bridge to accommodate damage associated with these large lateral loads. They give. They’re weaker than the earthquake but they’re capable of accommodating the large deflections that the earthquake is going to impart on it,” he said.
Bridges built by this idea have fared better in subsequent seismic events than the widespread bridge collapses in the 1964 quake. The Denali quake in 2002 saw relatively little damage for its magnitude.
“It was a very large earthquake, a 7.9 earthquake. It’s not a trivial matter,” Marx said. “We ended up sort of fixing and replacing some bridges, maybe four, that were not critically damaged. It was just easier to replace them than it was to fix them back up.”
Contrast that with the 9.0-magnitude quake that struck Japan in 2011, which caused widespread, dramatic destruction, with bridges showing giant cracks or collapsing altogether.
Japan’s philosophy had been to build to fight force, not yield to it, Marx said. Japan since then has adopted the engineering approach used in the U.S. and much of the world, Marx said, but at the time their idea was bigger is better.
“The Japanese philosophy that developed about the same time as the U.S.’s was that they could outmuscle the earthquake. They could take on Mother Nature and they could design a column that’s big enough to fight back,” Marx said. “The first problem is we don’t really know how big the earthquake is going to be. We’re trying, and we’re getting certainly better now than we’ve ever been in the past, but we really don’t know. It’s a difficult target to hit on many fronts. So you might design for one earthquake only to find out the earthquake that showed up was bigger, and, consequently you didn’t provide enough strength and failure occurred.”
And it’s not just lateral load to worry about. Earthquakes present a violent variety of stressors. The shaking is the most obvious, but as research from the 1964 quake showed, that’s not the most destructive.
“The ground shaking will get a bridge or a building or a car, whatever’s on it, shaking and that shaking might damage it,” Marx said. “What might be a better way to look at it is earthquakes cause all kinds of different damage.”
Particularly in Southcentral Alaska, with its granular soils, soil liquefaction is a bigger threat.
“It puts the soil particles into suspension, meaning they’re now floating in the soil water matrix. Because they’re just sort of floating around there they don’t have any strength, so they flow like syrup or mud. A lot of the bridges were destroyed because the soil those bridges rested on was liquefied,” he said.
Without a sturdy base to support a structure’s load, it can sink, rotate, crack and collapse.
“In Alaska we’re also concerned about liquefaction and temperature, as it’s often very cold, and as it gets cold these materials become more brittle, whether they be steel or concrete. Also the soils can freeze, and that can be a problem,” Marx said.
As if that weren’t enough to worry about, in coastal communities there’s also the wild card of tsunamis.
“Well, I certainly didn’t design for a boat to smash into the side of a bridge,” Marx said, of the old style of engineering pre-1964.
All of these possibilities are mitigated under the idea of building to yield, not break. For example, old-style bridges were supported with traditional, steel-reinforced concrete columns. The bigger the predicted load, the beefier the column. After 1964 the idea became to focus on confinement of the concrete, to use more transverse (horizontal) steel to create a sort of cage which would keep the concrete from disintegrating into weak spots. That can take a lot of steel, though, so today in Alaska the standard is to use concrete-filled steel pipes as bridge supports. They’re pounded into the ground, some of the dirt is augured out and the pipes are filled with concrete.
“It’s an infinitely tight weave, there’s no concrete getting out of that section. The steel shell is very strong, it’s very stiff and it’s capable of deforming a lot — lots of bend to it before it breaks, more so than traditional concrete columns. They’re providing that confinement along their entire length. So if we are mistaken about where damage is going to occur that’s OK because this pipe has got that capability of accommodating damage along its full length, not just in a few spots,” Marx said.
“Now, I’m not saying there won’t be failures — we still have a lot to learn — but this is certainly steps in the right direction, and a great deal of it is based on research, and Alaska DOT is one of the agencies that’s sponsoring some of that research,” Marx said.
Structural engineering is a macabre art. The goal of an engineer, says Henry Petroski, a specialist in failure analysis and the guru for American engineering, is to obviate failure.
“Think about everything that could happen and make it not happen,” Marx said.
With earthquakes, though, even with all that has been learned since the Good Friday quake, there are still potentially many more lessons. And engineers don’t know for sure whether they’ve got the answers right until they’re tested.
“There’s all these other phenomena that come with the earthquake — tsunami and liquefaction and so forth. And then there’s the stuff we don’t know about yet, we just haven’t had enough earthquakes to find and expose all of our vulnerabilities,” Marx said. “The problem is, how do I know what’s going to fail until after I have some tests or experiments to see what’s going to fail? And, unfortunately, we need a real live earthquake as our experimental testing ground. On one hand I’d rather not have any earthquakes, that solves the problem quite nicely. On the other hand, without the earthquakes, how are we going to expose our vulnerabilities so we can address them?”
It’s a daunting task, to say the least, especially for some of the old-school engineers at the time of the 1964 quake.
“They came back from that event and said, ‘If that’s what we’re supposed to design for, don’t bother. Because you can’t do it,’” Marx said. “They’re all gone now, and I’m naïve enough to think that we can design for it. But these were not ignorant, dumb engineers, these are guys who were every bit as bright as the folks practicing today, and if they say that (the ’64 quake) was that devastating and the folks who lived through it agree, then maybe we better not get too excited and say, ‘No, it won’t collapse.’ The next great earthquake might identify something that we didn’t even know was a problem and the bridges and buildings and everything else suffer because of it. What I can say is we’re employing the best technologies and techniques, we’re researching these things, we’re trying to obviate these modes of failure.”
Any bridge DOT has built in Alaska since the mid-2000s is designed and constructed using the newest earthquake techniques, like the steel-filled pipes.
“But as many bridges as we replace a year, about 12 a year, we’ve got about 1,000. It’s going to take a long time to replace all these bridges to the high standard that we design to now,” Marx said.
So for older bridges, DOT has instituted a seismic retrofit project to reinforce existing structures.
“These are relatively inexpensive but usually quite successful techniques at helping a bridge better survive an earthquake,” he said.
On the Kenai Peninsula’s highways, the recently built bridges in Seward are of the latest and greatest design, while the others, such as the Kenai River bridges, have been retrofit and should fare better when — not really a matter of if — the next big quake hits.
Most damages to a bridge would be visually obvious — surface cracks, chunks missing, rails bent, etc. But the severity of the damage might not be assessable with a glance, and if a quake happens at night, it’ll be that much harder to see. So people are recommended to stay put in the event of a quake as much as possible until crews have a chance to investigate and remediate any road and bridge damages that might have occurred. Locally based DOT highway crews are trained to assess bridge and roadway damages. And in anything over a 6.0-magnitude quake, DOT inspectors are dispatched to assess whether bridges are safe to travel on. In the 2002 Denali quake, affected bridges were deemed safe to support repair crews and equipment within 24 hours.
That’s the ultimate goal, to take a licking and keep on trucking.
“The best compliment we could probably receive after a large earthquake is they didn’t know there were any bridges on that route anyway. If we go unseen, that probably means we’re doing our job well,” Marx said. “There are no absolutes in earthquake engineering, it’s just the best we can do today, and we’re certainly doing more and more all the time.”