Author’s Note: The latest and most detailed proposal is in a third post, “The Case for Lidding I-5 in Downtown Seattle”, dated December 4, 2015.

Earlier this year I proposed capping Interstate 5 in downtown Seattle. The original idea was radical, and it received a flood of positive and negative feedback. This entry looks at the project more critically, dials it back to a reasonable scope, and includes additional details. I’m tackling this again because I plan to pursue it as a thesis for my master’s degree over the coming year. I chose not to update the original post in order to preserve it as a record, but to accommodate new readers I have repeated (and edited) some of the original content here.

When President Eisenhower envisioned a great road system connecting the far-flung cities of the U.S. he didn’t quite expect the engineers to plow through entire neighborhoods. But that’s what happened, and the legacy is the concrete ribbons that divide our cities to this day. An example is right here in Seattle with Interstate 5 running north to Shoreline and south to Tukwila for about 16 miles, dividing Downtown from First Hill, Capitol Hill from South Lake Union, and the U-District from Wallingford. Amid rapid population growth and demand for residential and office development, now is the time to reclaim valuable urban land.

Demolition and construction through Downtown, circa 1960s.

When the the Federal Aid Highway Act of 1956 guaranteed covering 90 percent of construction costs, the State of Washington immediately began planning I-5’s route through Seattle. Planners said Downtown “…contains older and less desirable buildings…from Marion Street to Denny Way — this section will traverse an area that needs improvement in appearance” (ironic emphasis added). Due to the area’s hilly topography, much of I-5 in Seattle cuts through the land with large retaining walls and elevated viaducts (rather than running flat on the surface). Part of the downtown segment is a trench with concrete walls on both sides. Paul Thiry, chief architect for the 1962 Seattle World’s Fair, suggested capping the freeway with a lid. Other proposals included building rapid transit in the median or tunneling underneath First Hill.

Residents of First Hill organized the only protest, decrying the anticipated consequences of pollution, noise, decreased property values, and isolation from Downtown. But citizen opposition was ineffective and the route went unchanged.  Eminent domain and federal funding were used to acquire a 300 foot right-of-way by purchasing and relocating or demolishing hundreds of homes and businesses. The Seattle section of the freeway opened to traffic in 1967.

A decade later, Freeway Park became a concession to residents. It opened in 1976 with 5.2 acres of walkways, landscaping, and water features that cross over I-5 from 6th to 9th Avenues between Seneca and Union Streets. Its modernist form, in line with surrounding office buildings at the time, was designed by architect Angela Danadjieva from the landscape architecture firm Lawrence Halprin and Associates. Danadjieva went on to author a Freeway Air-Rights Development Plan for the area. The Washington State Convention Center, opened in 1988, is adjacent to the park and also stretches across I-5 with over 410,000 square feet of meeting and exhibit space. These examples provide clear precedents for bridging the freeway’s massive width, which has been further discussed and studied (PDF) in recent years. At one point a federal courthouse building was considered as a lid extension.

I-5 has already been spanned with Freeway Park and the Convention Center.

The Washington State Department of Transportation (WSDOT) owns the air rights above the freeway, so continued cooperation between WSDOT and the City of Seattle will be needed for future efforts. Indeed, original plans for Freeway Park had additional structures in the three blocks southward of the existing park. An expansion doesn’t need to strictly honor the previous design, though, as the park’s brutalist architecture has a variety of nooks and crannies that give it a reputation for crime, injuries, and homeless camps. But additions should follow its spirit as a momentary sanctuary from urban life.

The motivations for covering up I-5 are numerous. Land is extremely valuable in the Downtown and First Hill blocks around I-5; a rough calculation based on parcel data from King County, using a range of land uses and parcel sizes, revealed an average land value of $18.1 million per acre ($416 per square foot). Amid explosive growth and a growing affordability problem, such a wide strip of underused land is increasingly attractive for private development, civic functions, and park space. More broadly, I-5 is a gash through the heart of Seattle, which is a city that prides itself on alternative transportation and reducing single-occupancy-vehicle use. While the city is moving forward with rebuilding the waterfront, it can also capitalize on this opportunity to improve the image of the broader Downtown area. Other issues include the unpleasant noise and exhaust fumes, though these nuisances haven’t hindered adjacent residential development.

Pictured: prime real estate.

One option is puting I-5 in a tunnel. The under-construction Highway 99 tunnel just a few blocks away is a convenient example for a cost comparison. That project has a number of components, including new surface-level paving and demolition of the Alaskan Way viaduct; the underground activities have been allocated about $2.05 billion. At 2 miles long and 4 lanes, that comes out to about $1 billion/mile or $.25 billion/lane-mile.

I-5 through Downtown is 14 lanes at its widest, including the reversible express lanes. Bertha, the tunneling machine working on Highway 99, is already the world’s largest with a diameter of 54 feet. To keep the same amount of lanes, an even larger machine would be needed for an I-5 tunnel, or four Bertha-sized tunnels would need to be bored parallel. The latter idea would have an estimated cost of $3.5 billion/mile. That is absurdly expensive, and does not even begin to consider the other costs of design, administration, and operation. And, of course, Bertha is expected to be inoperable for over a year, so this wouldn’t be a popular option.

A non-boring option is cut-and-cover, a method in which a trench is dug and a roof built overhead. But long term lane closures, which would be necessitated by this method as excavation is done in the right-of-way, could be disastrous; I-5 carries some 250,000 vehicles and 50,000 transit riders per day (PDF) in Seattle and thousands of tons of freight up and down the west coast.

The proposal’s scale has been significantly reduced. Click to enlarge.

The best alternative, then, is to construct a lid over the existing lanes. The original proposal was ill-conceived and had the lid stretching from Yesler Street in Pioneer Square to Galer Street in Eastlake, a distance of 2.5 miles. Much of this route only has a retaining wall on one side or is slightly elevated, meaning the downhill side of the lid would require massive columns that would add cost and unsightliness.

The preferred area is Downtown where the freeway is in a trench. The adjusted proposal is about three-fifths of a mile from end to end with eight sections, from Madison Street to Olive Way. South of Madison Street is a small section that could also be lidded, but is not included in this proposal. Future considerations could possibly extend south to Yesler Street and north to Denny Way, where the surroundings have dense conditions, though these sections would require the aforementioned additional engineering.

Study areas of the project. Total area (in orange) of the eight sections is about 9.6 acres. Click to enlarge.

The lid will be built with concrete and steel, with large beams spanning the trench and supporting the superstructure. Center columns will probably be needed in the medians, where existing overpasses are supported. The elevation changes from one side of the trench to the other, so the lid could either be sloped or terraced. Considering the length of enclosed roadway, the lid will create a tunnel that requires exhaust ventilation, fire suppression systems, lighting, and round-the-clock monitoring from WSDOT’s control centers. Height clearance, on- and off-ramps, and number of lanes can be preserved to avoid traffic impacts. The lid itself will have several layers, including a structural slab, drainage beds, and engineered soil.

The bottom of Freeway Park. New lid sections would use similar construction.

The soil needs to be lightweight but able to support plants, because the primary function of the lid will be a park that creates much-needed open green space for the dense Downtown. Like other parks, features can include playgrounds, sportsfields, walking and biking paths, pools, pavilions, and community centers. Plazas could be used for food vendors and public events like concerts, markets, and festivals. The presence of quality park space could also help draw more families to the Downtown as calls go up for bringing an elementary school to the neighborhood. Indeed, if no other space is found, which seems likely, a school can also be built on the lid.

Conceptual view of a park on Section 4, looking south from Pike Street. Click to enlarge and see comparison.

The original proposal made the case that the lid could support high-density private development. There are a few examples of towers being built in such conditions, such as New York’s ongoing Hudson Yards development, but it would drastically increase the costs of the lid structure. That said, it’s probably feasible for low-rise steel or wood-framed buildings, perhaps no more than three or four stories, to be built on the lid without adding cost. These buildings could contain a mix of retail, office, and residential space. The lid itself will be a place-maker, but putting people on top could create an even more unique destination and amenity.

Conceptual view of a school on Section 8, looking north from Madison Street. Click to enlarge and see comparison.

The primary issue with any large public works project is cost. Infrastructure costs are chronically underestimated due a variety of reasons (this paper succinctly discusses them), so what follows are educated estimates using previous projects.  Nearby on Mercer Island, a half mile long lid over Interstate 90 was completed in the early 1990s in response to citizen pushback to that freeway’s construction. On top is the 20 acre Aubrey Davis Park, which has open space, playgrounds, trails, and sportsfields. The structure, ventilation, and soil cost $146 million. Converting that $242 million in today’s dollars and estimating the lid is approximately a half-mile long and 200 feet wide, an estimated cost is $20 million/acre. This is similar to the 5.2 acre Klyde Warren Park in Dallas, which was completed in 2012 at cost of $110 million, or about $21 million/acre. Freeway Park, in today’s dollars, cost $18.5 million/acre.

Conceptual plans for each lid section. There is an opportunity to create a variety of designs within a small area. Click to enlarge.

The eight sections of this proposal are about 9.6 acres total, bringing the cost estimate to $192 million. Conservatively rounding up pegs the cost at $200 million, which is 10 percent the cost of the original proposal. For comparison, $200 million can buy half the Columbia Center, two-thirds of a Boeing 747, 2.5 Jumbo Mark-II class passenger ferries, or 44 Space Needles. This cost is also more politically feasible and entirely within the city’s means to fund, as seen with the waterfront redevelopment ($420 million) and seawall replacement ($385 million). But, again, it is only an estimate.

Funding is available from multiple sources. The City of Seattle can start with a “local improvement district” (perhaps call it a lid LID), which taxes surrounding private properties that benefit from a large public investment. This tool is expected to raise $200-300 million for the waterfront project. The waterfront is also being partly funded with money from the State of Washington, which will probably be willing to chip in on an I-5 lid like it did with Freeway Park. Construction bonds (borrowed money) or a citywide levy could also be used with voter approval. Also, like the freeway cap in Dallas, the City could partner with private interests to create sponsorships and solicit donations (Amazon Park and Starbucks Plaza, for instance). Parcels or buildings on the lid could be managed by a non-profit and leased out to fund ongoing maintenance and programming. Within a few years, the project will also support the increase of surrounding land values, and therefore property tax revenue for the City. The $115 million High Line in New York famously generated $2 billion in private development, for example. Like the High Line, the lid may also be a big tourist draw.

Along with economic gains, the lid will have environmental and social benefits. Noise and air pollution will be significantly reduced, while new vegetation can support wildlife habitat and reduce the local heat island. People will have improved connections between neighborhoods and gain a new area to live, work, and play.

After planning and design is complete, it might take two or three years to complete all eight sections of the lid. Construction will have traffic impacts, of course, but delays and congestion are nothing new for this stretch of I-5 (Freeway Park took about two years to build) and will be negligible considering the citywide benefit. Working on nights and weekends can limit impacts. Interstate freight and through-traffic can also be detoured to I-405, which is being widened. The lid also doesn’t have to be built all at once, as each section is unique.

Locally, WSDOT is also planning and constructing two lids in the Montlake neighborhood and three in Medina over the reconstructed State Route 520. Other cities across North America have recognized the merits of covering up or tearing down freeways in their urban cores. In San Francisco, planners are studying removing part of I-280 and relocating rail yards to open up some 27 acres for housing and employment. Philadelphia is developing concepts for capping I-95 with an 11-acre park (costs estimated at about $18.6 million/acre) and bringing people back to its riverfront with residential and commercial development. And a recent study in Toronto recommends tearing down a three-level expressway to make room for new development. Chicago’s Millenium Park was built over rail sidings and parking lots.

Freeways are vital for moving people and goods in a country as large as the U.S., but there is no reason for them to disrupt our downtowns and create barriers between people. The proposed I-5 lid will create new space for parks, residences, and commercial development while improving connections between dense neighborhoods and drawing private investment to surrounding properties. And perhaps most importantly, the lid will redefine the character of Seattle’s downtown.

If this interests you, subscribe by email and the Facebook page for future updates. You can also get involved with the local advocacy group Lid I-5, which is pursuing the same idea, or contact me.