Stuck in Seattle

The Aggravating Adventures of a Gigantic Tunnel Drill

About 20 workers wearing hard hats and reflective vests clump together on the edge of a chasm near Seattle’s waterfront, peering down a hole 120 feet deep and 83 feet wide. The last men have been craned out of the pit in a yellow metal cage. Gulls squawk. A TV news helicopter hovers overhead.

A dozen journalists stand nearby on the bed of a truck. We’re here to see Bertha, one of the world’s biggest tunneling machines. Or at least a piece of her. A 240-foot crane is about to haul a 540,000-pound steel shield out of the ground, 20 months after Bertha started digging a highway. Almost imperceptibly, the crane starts rising.

The event, on a Thursday in mid-March, is part of a massive rescue mission to fix the $80 million machine. She broke abruptly in December 2013 after boring through just 1,000 feet, one-ninth of her job. Her seals busted, and her teeth clogged with grit and pieces of an 8-inch steel pipe left over from old groundwater tests. She stopped entirely.

The tunnel, with a budget of $1.4 billion and originally scheduled to be finished in November 2015, is two years behind schedule. The state’s contractor, a joint venture called Seattle Tunnel Partners (STP), has spent months digging to reach Bertha and crane her to the surface, where a weary Seattle awaits.

Bertha’s job is to bury a highway that runs on a structurally unsound elevated road smack in the middle of an earthquake zone. The viaduct, as it’s called, follows the shoreline, effectively barricading downtown Seattle from what could be a beautiful waterfront. The tunnel will let most of the traffic travel deep underground; at street level an old freeway will be demolished, and in its place the city will build a boulevard and shoreline park created by the designers behind New York’s acclaimed High Line park. The $4.2 billion plan calls for the long-neglected waterfront to come to life; Seattleites can celebrate the glory of Puget Sound, where ferries dash across the bay and the jagged peaks of the Olympic Peninsula jut in the distance.

Everything about the project is gargantuan, starting with Bertha, who is as tall as a five-story building. She runs on a 25,000-horsepower motor and has a head weighing 1.7 million pounds, with 260 steel teeth designed specifically to chew through Seattle’s silty soil. She’s named after the city’s first and only female mayor, Bertha Knight Landes, who served in the 1920s. According to the machine’s official state biography, her role models include “whoever invented the shovel.” Bertha’s got 15,700 Twitter followers, has inspired Halloween costumes, and was once feted by thousands.

After Bertha got stuck, she couldn’t back up because she builds the concrete walls of the tunnel as she drills forward. That means the hole she leaves behind is narrower than she is. The contractor has devised a method—itself unprecedented—­to repair Bertha by craning her in sections to the surface. After almost a decade of debating the tunnel’s merits and three more years of construction, more than a few Seattleites argue that Bertha should be buried where she is, her last rites read, and another plan pursued.

Two Republican state senators proposed cutting off Bertha’s funding, declaring the project has failed. As part of an interactive exhibit, the Seattle Art Museum asked visitors to imagine what Bertha was thinking. Among the responses: “How in the hole did I get here!!!” and “Another billion dollars, please!”

The Seattle Times editorial board tried to calm the hysteria. Bertha may be a lemon, they wrote, but the “herculean engineering work” to repair her makes it “too soon for a Plan B.” The state and the contractor, united until the bill comes due and the widely expected legal battle over the price tag starts, say they’re not abandoning ship. Bertha has become too big to fail.

Megaprojects almost always fall short of their promises—costing too much, delivering underwhelming benefits, or both. Yet from the London-­Paris Chunnel to Boston’s Big Dig, cities still fall for them, seduced by new technologies and the lure of the perfect fix. A mix of factors has given Seattle a particularly acute sense of angst. The project depends on a singular piece of engineering. And Bertha’s building a highway for cars in a city where workers overcrowd buses and commuters wrap themselves in waterproof everything to bike in the rain.

At the construction site, onlookers fall silent as steel cables quietly creep out of the pit. Eventually the top of the arched piece peeks out. “Wow,” someone whispers, as camera shutters click. A worker from Japan’s Hitachi Zosen, which manufactured Bertha, puts her hands together and makes a small bow. Once the metal shield fully emerges, it hovers over the pit like a hulking rusty rainbow. The crane takes a half-hour to pivot 90 degrees, back up, and gently set the piece down on the ground.

Deep at the bottom of the rescue pit, the guts of the machine—the source of so much trouble—are finally exposed. “So THAT’S what my insides look like,” tweets @BerthaDeBlues, one of several parody Twitter accounts. “I’m gonna hurl.”

In 2001 the 6.8-magnitude Nisqually earthquake shook Seattle, causing cracks along State Route 99’s Alaskan Way Viaduct, the elevated highway that skirts downtown. Everyone agreed the viaduct wasn’t safe to keep open much longer. But true to the so-called Seattle Process of inclusive and abundant dialogue, what should replace the roadway was ferociously debated.

At first there were two main options: a new elevated structure or a shallow tunnel roughly 50 feet below ground. The elevated option was simpler, but it meant there’d still be an unsightly highway running along the waterfront. The tunnel would free up the landscape, but it would be built using a “cut and cover” technique, which tears up the streets and, as anyone familiar with the Big Dig knows, is tremendously disruptive.

Seattle voters roundly rejected both options in 2007 in what Cary Moon, a local urban designer, says amounted to a “No, and Hell No!” vote. She, along with other artists, environmentalists, and community activists, pushed another idea to the fore: tear down the highway and don’t replace it. They pointed to evidence that new roads increase traffic rather than ease the crush, a phenomenon known as induced demand. “When roads are built you just think, Oh, I can now go to that mall, or I can sign up for that soccer camp,” Moon says. “In aggregate, those decisions add up to a lot of car traffic.”

Moon’s coalition said Seattle should follow New Urbanist examples, such as Seoul, Milwaukee, and Portland, Ore., that replaced highways with smaller surface streets, public parks, and dedicated lanes for mass transit and biking. Instead of seeing gridlock, those places found car trips declined as people opted for other means of transport or changed their plans and didn’t travel as far. San Francisco’s Embarcadero was an oft-­cited example. Like Seattle’s viaduct, the double-­decker roadway carried more than 100,000 vehicles a day. After the 1989 Loma Prieta earthquake, San Francisco tore down the heavily damaged elevated road and constructed a boulevard with a streetcar and waterfront promenade. It helped revitalize the South of Market neighborhood, now beloved by tech startups, and trips on mass transit in the area increased 75 percent in a decade, according to a 2007 study.

Seattle’s business interests dismissed the idea. “It was all about a pretty waterfront park,” says Bruce Agnew, director for the Cascadia Academy, a regional think tank. He and others feared cars would clog downtown streets and nearby Interstate 5, a major thoroughfare. They worried about the 4,000 freight trucks that run on the viaduct each day. Agnew helped revive an idea the state had previously deemed too pricey: a deep-bore tunnel, mined by a custom-built machine that digs without tearing up surface streets.

Cascadia, funded with $9 million from the Bill & Melinda Gates Foundation, is part of a local conservative group known for promoting “intelligent design.” It joined with global engineering company Arup to develop the tunnel idea. Arup produced a study that showed that boring costs had come down, though the study said it didn’t adjust for inflation and included estimates for tunnels not yet built. Cascadia hosted a conference with 100 guests, largely from companies that design and build deep-bore tunnels around the world. “They basically said that the technology had advanced so much in the past few years,” says Agnew.

In late 2008, as a committee appointed by the governor was closing in on supporting the Embarcadero approach, Cascadia’s analysis provided the groundwork for the chamber of commerce and other business interests to push for a deep-bore tunnel. They trumpeted a “Tunnel + Transit” plan, which included $190 million in additional transportation investments, such as new rapid bus lines. Because the tunnel would be so deep, it couldn’t have any local exits in downtown, which accounts for about 40 percent of the viaduct’s traffic.

Inside the tunnel in September

Photographer: Ted S. Warren / AP Photo

Bob Donegan, president of Ivar’s, a local chain of restaurants with a flagship location on a pier where the future waterfront park will be, helped build political support. “We brought in the agriculture interests from Eastern Washington who said we need to get our wine and apples to the port,” he says. The American Automobile Association of Washington, labor unions, and real estate developers backed the tunnel, he says, as did Boeing, which has major facilities north and south of Seattle. “They made it very clear that if all of that traffic goes through I-5, we will do commerce somewhere else,” he says. Boeing declined to comment.

The tunnel’s proposed path showed just how many constituencies the project had to please. Its southern portal dives below ground by an industrial zone that includes the fourth-largest port in the country. It then runs under downtown and reemerges in the South Lake Union neighborhood, where Amazon.com has snapped up block after block.

In January 2009, then-Governor Christine Gregoire, then-King County Executive Ron Sims, and then-Seattle Mayor Gregory Nickels publicly threw their support behind the deep-bore tunnel. Republican legislators, largely from suburban and rural districts, had little interest in funding a risky tunnel for the city, so Ed Murray, now ­Seattle’s mayor but in the legislature at the time, brokered a compromise: The state would pay $2.8 billion. The rest would be covered largely by the city, the port, and the federal government. If the tunnel ran over budget, the bill read, extra costs “shall be borne by property owners in the Seattle area who benefit” from the tunnel. Clark Williams-Derry, a researcher at Sightline Institute, an environmental think tank, says it amounted to the statehouse saying: “If Seattle wants its latte-filled tunnel, they can build it themselves.”

If Bertha was now conceived, her gestation period would last four more years through environmental reviews, protest-filled hearings, and two new mayors, one elected largely on an anti-tunnel platform. In the public airing, the challenges and risks of a deep-bore tunnel were becoming clearer. The Stranger, a local alt-weekly, summed up the anxiety in a 2010 cover story with a memorable opening line: “You’re about to get f-----, Seattle.”

Tunneling machines like consistently firm conditions across the drilling face. “The risk is very much associated with how you expect the ground conditions to vary,” says Andrew Whittle, a soils expert who heads Massachusetts Institute of Technology’s civil engineering department. The wider the tunnel, the more likely it is for one part of the drill to encounter little resistance while another meets tougher stuff. Initially the Seattle plan called for two smaller bores, one in each direction, but that was too costly. Making one tunnel with two levels of traffic was cheaper but meant pushing the diameter to an unprecedented 57.5 feet.

The Washington State Department of Transportation’s consultants identified eight different ground conditions in the tunnel’s path, including abrasive soil that can wear down a machine’s teeth; clay-like silt; and boulders up to 8 feet in diameter, deposited by glaciers more than 10,000 years ago. The southernmost stretch of the tunnel runs along Elliott Bay, where water ebbs and flows through the soil with the tides, in a consistency Donegan calls “softer than a milkshake.”

The tunnel’s path also runs so deep, up to 215 feet below ground, that it faces groundwater pressures about four times greater than at sea level. There are even pressure variations between the top and bottom of the boring machine. At five stories tall, it’s like the difference between being able to snorkel or needing scuba gear. The state, guided by an army of consulting firms, said it was aware of and could prepare for the risks.

WSDOT awarded a $1.4 billion design-build contract to STP, a joint venture between Tutor Perini, a California-based construction company with $4.5 billion in annual revenue, and Dragados USA, the local division of a Spanish company with an expertise in tunneling. The thousand-page contract outlines the exact requirements for the tunnel but leaves STP responsible for most of the design and construction, an arrangement intended to shift risk away from the state. STP beat out another consortium in part by estimating it could finish the project by December 2015, 11 months ahead of the state’s schedule. It was time to stop hyperventilating, the state said. “With this contract, we are confident that the tunnel will be built within budget and delivered on time,” declared Paula Hammond in 2011, then the head of WSDOT. (She left in 2013 for the engineering firm that led the reviews.)

STP decided to buy the tunnel-­boring machine from Hitachi Zosen, and in December 2012 school kids christened her Bertha. She uses pressure to keep soft earth from seeping into spaces where humans work and sensitive machinery is located. Her cutter face rotates like a cheese grater, chewing up and pushing soil into a cavity, where it’s mixed with an additive to make it the consistency of toothpaste. A massive corkscrew acts like her digestive tract, channeling the muck to a conveyor belt that carries it out the back of the tunnel to be barged away.

As Bertha drills, her robotic arms install precast concrete rings that create the tunnel’s permanent walls. Once she installs a ring, she pushes off it with hydraulic legs to continue her forward thrust. She drags about 300 feet of equipment behind, including lunchrooms and hypobaric chambers so workers can make repairs in pressurized areas and not get the bends.

Bertha arrived by sea from Osaka in 41 pieces and was reassembled in Seattle, then lowered into a launch pit 80 feet deep and 400 feet long. In July 2013, 5,000 people went to Bertha’s send-off, complete with confetti and food trucks.

In early December 2013, after she’d mined about 1,000 feet, Bertha started overheating. Why exactly Bertha wasn’t up to the task isn’t clear, at least to the public. WSDOT has said it’s STP’s problem if Bertha couldn’t handle the ground conditions it laid out in the contract. But STP has already submitted a change order saying the conditions differed from what WSDOT told them. Hitachi Zosen declined to comment,citing a confidentiality agreement. The question of Bertha’s breakdown is as much a financial problem as an engineering one, to be challenged through a dispute review board and, potentially, in court. “The lawyers will have a field day with this,” Donegan says.

Bent Flyvbjerg, a professor at Oxford’s Saïd School of Business, has followed Bertha from afar. His research on megaprojects has been cited by both backers and critics of the tunnel. Nine times out of 10, massive infrastructure jobs go over budget, he says. Tunnels on average cost 34 percent more than anticipated. No region is better at predicting costs, and estimates over the past century haven’t become more accurate, his data show.

Flyvbjerg recalls reading how WSDOT said taxpayers won’t end up having to pay for any cost overruns. “When I saw that statement, I said, ‘Oooh, that is really a risky thing to say,’ ” he says. “It is very common that taxpayers’ money ends up covering it, even though people think that they have a contract that prohibits it. The difference between the Seattle project and other projects was that the optimism was so pure. That was hugely deluded.”

A year ago, STP estimated the repairs could total $125 million, but that was when the rescue was supposed to be done around Labor Day 2014. STP’s latest timeline is almost a year later. In a joint interview, Matt Preedy, the state’s deputy program administrator, and Chris Dixon, STP’s project manager, deflect questions about the costs. What’s the latest repair estimate? “Costs are being incurred; we don’t have a projection as to what the total costs will be,” Dixon says. Has the work to date already exceeded the original $125 million estimate? “We’re not in a position right now to further elaborate on costs,” Preedy says. Can you understand why that may frustrate taxpayers, given the uncertainty over who’ll pay? Six seconds of silence, finally broken by a press officer: “I don’t know how you expect them to answer that question.”

Ultimately, Preedy stays on message, saying, “Those contractual issues will be worked out in time. Right now the focus is on completing what we started, getting the project done, safely.”

As parts break, delays mount, and costs rise, government mistrust has set in. “People are really dubious about all of the claims,” says Moon, the landscape designer. “Claims that they can fix it. Claims that we won’t have to pay for it. Claims that it is going to have no trouble boring again. They have lost credibility on all of those fronts.”

The repair site in February

Source: Washington State Department of Transportation

In mid-March, as journalists wait for Bertha’s shield to come out of the ground, Governor Jay Inslee is on Reddit, holding an Ask Me Anything session. “Hi Governor Inslee!” writes holierthanmao. “As a Seattle resident, I have to ask, what is plan B if this whole 99/Big Bertha project becomes a loss? There has to be a contingency plan, right?” Inslee responds that there’s no backup plan—yet. He says they’re focusing on holding STP to its contract to deliver the tunnel: “We have to insist they finish the job. Period.”

It’s also unclear whether the project will fulfill its original promise. Seattle has scaled back the waterfront park, saying the delay raised costs. And the $190 million in transit investments still aren’t funded. Chris Arkills, the transportation policy adviser for King County, which includes Seattle, says the legislature hasn’t given the county the taxing authority it needs.

On March 30, the TV chopper is in the air again, watching as one of the largest cranes on the West Coast slowly hauled Bertha’s 4 million-pound cutterhead and drivetrain up to the surface. Next, STP will install a new system of seals, replace Bertha’s main bearing, and reinforce her with 160,000 pounds of steel. Assuming repairs go well and Bertha starts drilling again, her path will soon turn east, directly under the unstable viaduct. She’ll then veer north, under downtown. While STP can dig a rescue pit now with minimal disruption to the city, that won’t be the case if she breaks down again farther along the route.

Corrects the size of the tunnel's contract and budget.

Corrects Cary Moon's profession in the 15th paragraph.

Weise is a reporter for Bloomberg Businessweek in Seattle. Follow her on Twitter @kyweise.

Kiyotaka Matsuda contributed reporting from Tokyo.

Illustrations: Steph Davidson

Design & Development: Sheryl Sulistiawan

Graphics: Evan Applegate & Chloe Whiteaker

Editor: Emily Biuso

Digital Editor: Aaron Rutkoff

Digital Producer: Emily Engelman

More features