New Armory Way Bridge Highly Rated in Faulty Magnolia Bridge Study


The Seattle Department of Transportation’s study on replacement options for the Magnolia Bridge, one of three routes into and out of Magnolia, came out last week. The current bridge is nearing the end of its operable life, and the study predicts dire consequences for vehicle travel times if the bridge is not replaced in some form and for not only Magnolia but the entire 15th Ave W corridor, including Ballard and Queen Anne. The study sets the stage for another large infrastructure project to get dropped on the city’s plate with no funding sources identified.

But what to make of the assumptions that form the basis for all of the modeling in the study? The picture painted of future traffic volumes does not actually appear to be realistic, and repeats the car-centric planning mistakes of the past that put level-of-service considerations in the center of the discussion. But let’s look at the alternatives that were considered and how those assumptions played into them.

Among the Alternatives, Armory Way Bridge Comes Out Ahead

According to the study, the course of action that ranked highest was construction of a new bridge connecting 15th Ave W and Thorndyke Avenue at W Armory Way, as well as reinforcing the eastern portion of the current Magnolia Bridge to continue to allow alternate access to Port of Seattle facilities and Expedia’s new campus at Smith Cove. It would also involve improvements to nearby streets (Thorndyke Ave W and 20th Ave W) and the creation of a new street bordering Port of Seattle property to connect the western end of the bridge to the Elliott Bay Marina and Smith Cove. The categories this set of projects ranked highest in were mobility, cost, and implementation (most of the construction could occur while the current bridge still stands). The Armory Way bridge package ranked highest among all for bike and pedestrian connectivity.

The new Armory Way bridge itself would cost only $45 million, with the bridge retrofit adding another $67 million onto the project cost. However, that doesn’t include right-of-way or design costs, so with the other project components the grand total is estimated at $266 million with a 30% contingency already included.

The Armory Way bridge concept actually came out ahead of the in-kind replacement of the bridge. (City of Seattle)
The Armory Way bridge concept actually came out ahead of the in-kind replacement of the bridge. (City of Seattle)

A full in-kind replacement of the bridge ranked just below the Armory Way bridge alternative. It scored well, unsurprisingly, on mobility (most of the traffic models extend current trends) and environmental impact (few new areas are impacted), and community support. But the full replacement scored lowest in terms of cost, with a top level estimate of $420 million.

A Magnolia bridge replacement could take up nearly half a levy the size of Move Seattle. It’s also over four times the cost of the Lander Street overpass, which was the recipient of the largest single grant ever awarded from the Federal government to a transportation project in Washington. The replacement also scored low on construction impacts, with high disruption occurring with a demolish-to-build schedule. It was, however, the highest ranked alternative for Sound Transit, the Port of Seattle, and the BNSF railroad company. The Armory Way bridge was ranked lowest by all three of those stakeholders.

The other two alternatives that were studied would not provide any new connection over the BNSF railway between 15th Ave W and Magnolia itself. Because the traffic models in the study project a large increase in traffic on 15th Ave W, the models predict a high amount of delays would be associated with utilizing 15th Ave W to get vehicles to the nearest bridge at W Dravus St.

Alternative 2 would create the connection between 20th Ave W and Smith Cove and retrofit the eastern portion of the bridge connecting 15th Ave W to Expedia as in Alternative 1. But rather than add another bridge at Armory Way, it would add additional lanes to W Dravus St.

Alternative 2, a widening of W Dravus St rated low by the study. (City of Seattle)
Alternative 2, a widening of W Dravus St rated low by the study. (City of Seattle)

Alternative 3 would do the same roadway widening on W Dravus St with the only other change being a reconstruction of the lower segment of the Magnolia Bridge only between 15th Ave W and Smith Cove, not Magnolia’s bluff.

Alternative 3 would only reconstruct the Magnolia Bridge's lower portion to access industrial areas. (City of Seattle)
Alternative 3 would only reconstruct the Magnolia Bridge’s lower portion to access industrial areas. (City of Seattle)

The main category that these alternatives scored low on was mobility, which brings us to the traffic volume projections that form the basis of the study.

Flawed Study Not Grounded in Reality

There are a number of assumptions made in the study that are not grounded in reality. First, it assumes an increase in vehicle volumes of 0.4% every year through 2035, an increase of high single digits after 18 years. This is not reflective of the actual trend for vehicle traffic in Seattle overall. Since 2006, average daily traffic in Seattle has declined.

Average daily traffic in Seattle 2006-2017. (City of Seattle)
Average daily traffic in Seattle 2006-2017. (City of Seattle)

Second, it adds trips associated with all nearby projects on top of that assumed growth: Expedia, cruise ships, a million square feet of new industrial space at Terminal 91, growth at Fishermen’s Terminal, and a large number of new housing projects in the development pipeline. If you’re thinking that Magnolia isn’t known as a hotbed of housing activity, you’re right–most of those projects are in Ballard or Uptown. Some are in Belltown. Magnolia isn’t even a designated urban village, meaning single-family zoning predominates.

All of these wild assumptions combine to spell out armageddon for any scenario that fails to maintain or expand car capacity. That goes double for options that utilize 15th Ave W–the corridor every new resident from Belltown to Ballard will obviously be driving up and down every day during peak hours–even after the corridor gets Link light rail in 2035, apparently.

Volumes on 15th Ave W are projected by this study to increase by 33% over 2017 levels, reaching 4,800 vehicles per hour in both directions north of W Garfield St by 2035. That’s a similar volume level to the number of vehicles per hour that use the SR-99 tunnel currently and over a thousand more vehicles headed northbound during afternoon peak than currently use the street.

In other words, the traffic modeling concludes that even if another Magnolia Bridge is built and things remain exactly as they are now, it will take drivers 32 minutes to get from Elliott Ave W to the Ballard Bridge during the afternoon commute. Take out the bridge and do nothing? That will go to 66 minutes. Yes, that’s right, people will apparently spend an hour in traffic with a light rail line running between Ballard and Downtown literally right next to them. That amount of time goes up to 126 minutes to get out of Magnolia without the bridge.

Travel time alternatives as shown in the report, with the in-kind replacement maintaining the status quo. (City of Seattle)
Travel time alternatives as shown in the report, with the in-kind replacement maintaining the status quo. (City of Seattle)

The idea of “no build” is really not addressed much in the report, but the dire travel times spelled out by the flawed traffic data are tucked away in the report. Bizarrely, the model shows a nine-minute travel time heading to Ballard during the morning peak, heightening the incredulity of the numbers.

2035 travel times with a no-build option. (City of Seattle)

Since the Magnolia Bridge study turned out to be a study supporting adding four or five lanes of traffic to 15th Ave W, it’s hard to take its conclusions seriously. The mobility assumptions that went into rating the possible alternatives were weighted 75% on general vehicle access and only 25% on everything else, including 10% transit, 10% bike and pedestrian access, and 5% freight. That is not an accurate balance for the Seattle of 2035, or even the Seattle of 2019.

What is missing from the report is the environmental impact of not reducing vehicle travel dramatically by 2035, when even our very aggressive electric vehicle goals still predict most cars will run on fossil fuels. The carbon emissions associated with six-minute travel times between Magnolia and north downtown for motorists are left unaccounted for, with vehicle delay at the top of the list of worst-case outcomes. Also unaccounted for is the safety impact of streets designed to accommodate that much traffic during peak periods or the health impacts overall if the system continues to encourage that much driving.

With no funding source for any future Magnolia Bridge replacement identified yet, the context around how Magnolia and Seattle as a whole would adapt to the different scenarios is just not correctly accounted for in this long-anticipated report. The conclusions it reaches should carry very little weight when leaders are debating the next course of action for the bridge.

We hope you loved this article. If so, please consider subscribing or donating. The Urbanist is a 501(c)(4) nonprofit that depends on donations from readers like you.

Ryan Packer lives in the Summit Slope neighborhood of Capitol Hill and has been writing for the blog since 2015. They report on multimodal transportation issues, #VisionZero, preservation, and local politics. They believe in using Seattle's history to help attain the vibrant, diverse city that we all wish to inhabit. In December 2020, Ryan started a three-month stint as editor of Seattle Bike Blog.

Inline Feedbacks
View all comments
Skylar Thompson

How much to widen the bike/pedestrian bridge and trail in the Interbay rail yard to support more than one direction of travel at a time? That’s guaranteed to be lower cost and more worthwhile given that it’s the only safe way to cycle out of Magnolia to downtown short of going all the way around Queen Anne.

Magnolia has plenty of ways to drive in and out, and losing one won’t be the end of the world. What it lacks are good alternatives to driving.


That’s rich, “Magnolia has plenty of ways to drive in and out”, “What lacks are good alternatives to driving”. Have you ever been to Magnolia? Do you even know where it is? Magnolia is south of Ballard, North of West Seattle. It has only 3 roads to access on and off of Magnolia. The City has already screwed up 1 access point by Fishermans Terminal, and try accessing I-5 via Mercer from Magnolia or just going across the Ballard Bridge around 5:00. There is no light rail station for Magnolia. There is very limited bus service from Magnolia to other parts of the City. Magnolia consists of two good size hills, not inviting for bike riding on and off. [Condescension]


I live in Magnolia and strongly disagree with just about everything that you said. First of all, I bike with my 1 year old everyday on and off of Magnolia and it’s great. It’s connected by trails to and from downtown. The bus service is lacking and should be the priority. I own a car and a single family home too just in case you want to imagine that I’m “different’ than you.

Brian Nelson

[Ad hominem, off-topic]

Brian Nelson

Seriously? How was what anything I wrote an Ad hominem attack?

Skylar Thompson

Perry, yes, I’ve been to Magnolia plenty of times. There’s a good reason that Magnolia won’t be getting a light rail station (though one will be pretty darn close in Interbay) and shouldn’t justify a $400 million bridge: it’s not on the way to anything. The city has plenty of transportation needs that would benefit much more of the city than a neighborhood that has resisted anything other than single-occupant vehicle infrastructure.

Now, that doesn’t mean we shouldn’t invest in Magnolia’s transportation needs. Like I mentioned before, improving cycling accessibility is cheap and, while Magnolia does have a couple big hills, the dense parts are towards the bottom of these hills and don’t involve big climbs.

You complain about driving out of Magnolia, but the fix isn’t improving driving conditions, but to give alternatives to driving.


I live in Magnolia and I’m completely in favor of making driving downtown as difficult as possible in my neighborhood (I try to bike everywhere and yes I own a car and I have a baby). The one thing that seriously needs consideration is the 3 bus routes that use the Magnolia bridge. People are rational and will choose the fastest mode of transit–make it the bus or the bike in this situation. There was also talk about adding a bike trail that would connect the Elliott Bay Marina to 32nd Ave W to make biking more doable for the western side of the neighborhood. Here is an excellent opportunity to make a neighborhood more public transit and bike friendly and spend significantly less $.


Magnolia is a wealthy, single-family zoned, non-urban village demanding that we spend an arm and a leg on their non-mobility while the rest of city is still in dire need. Not sure why you’d expect The Urbanist to support that.




Magnolia is full of multi-family apartments and condos with a wide range of families and people across all income levels. It represents a cross section of what makes up Seattle. The Urbansit seems to represent a very small and single minded group of people concentrated in a part of Seattle which has issues and concerns which they care about but do not represent the City and most neighborhoods.


“Bizarrely, the model shows a nine-minute travel time heading to Ballard during the morning peak, heightening the incredulity of the numbers.” That’s not bizarre at all. Morning traffic flows south to downtown/I-5, not north to Ballard. It reverses in the afternoon.

Car heavy Magnolia is a thing and will stay a thing until ST3 is built (15 years from now) and even then only if Metro seriously redesigns how the buses work on Magnolia which currently work as a slow downtown only ride. Even then it will remain car heavy.

Once Expedia opens up there will be a serious increase in the number of cars during peak hours. It will be the largest single employer the Interbay area has ever seen, around 4,500 people. Many will use transit, but even if only 10% drive that still represents 400-500 more cars. If the port gets its way there should also be a serious increase in freight truck traffic as well.

Frankly the best design, if cost is a concern, is a mix of 2&3. Keep the overpass so Magnolia drivers don’t have to execute a left hand turn across 15th and then dump them onto the road connecting back to Thorndyke. Though probably a better idea would be to route it to whatever new access road the port wants to build to support its new warehouse proposal. That way freight trucks can also use the overpass.

The idea of somehow adding lanes to Dravus street is so ridiculous that’d I’d go so far as to call it an outright lie by the city. The only way to improve Dravus street would be to install some sort of massive clover shape interchange between it and 15th. That’s the choke point. Adding more lanes doesn’t fix the choke point. So the only real option is to add or in this case maintain a different access point.

Frankly this all should have been fixed a decade ago. The problem with Magnolia bridge was the same as the Viaduct.


I feel like the only thing that matters is the funding. As long as a federal grant is secured, it’ll be replaced.

Hopefully it’s at least closed before it fails. I know in America we usually like to wait until there’s a body count before we do road maintenance, which is in my opinion a sub-optimal strategy.


SDOT and the city cannot continue to ignore Magnolia. Replacing the Magnolia Bridge is the right choice, better to get it done before it fails. Hard to understand the continued bashing of Magnolia from the Urbanist writers.

Just Some Guy

“Hard to understand the continued bashing of Magnolia from the Urbanist writers.”
It’s not hard at all. Magnolia might as well be the suburbs. I can’t fathom why it should be a higher priority than projects that serve more people (aka more density).

Brendan Miller

Well, Magnolia only has a population of 20,000, and this is a bridge replacement that costs hundreds of millions of dollars.

For comparison, the Southpark neighborhood struggled to get its bridge replaced. Southpark has a slightly higher population than Magnolia, a bit higher ridership over the bridge, replacing the bridge cost much less, and they have fewer alternative routes to downtown.

They had no bridge for 4 years before it finally got replaced.

Southpark is one of the poorest neighborhoods in the city, and Magnolia is one of the richest. This is probably why the city is doing this study and pushing for an expensive option. However, this is expensive enough that they can’t actually build it without a voter approved levy. Ask yourself, who is going to vote for this levy outside of Magnolia?

Also, to be clear, there is not a lot of money for this kind of project. The Southpark bridge only got $34 million, and that was during the recession when there was more federal infrastructure spending being awarded by the Obama administration. Trump is actually clawing back infrastructure grants.