Building ST4: The Case For Upgrading RapidRide E To Rail

Three car light rail trains. (Credit: Sound Transit)

Seattle Subway’s Vision Map generated a great deal of attention and that’s great for growing the transit-advocacy movement as we seek to build our success in passing ST3. That said, some neighborhoods were conspicuously missing from the vision, such as First Hill and Belltown. Generally, the principle in building high capacity transit should be start with your highest transit-use neighborhoods and then work your way out. Almost invariably the highest transit use neighborhoods are dense in housing, jobs, and/or entertainment. First Hill and Belltown are definitely in that category–the densest census tract in Washington state is in Belltown–we should definitely be including them in our rail plans. One way to do that would be upgrading the RapidRide E to elevated rail and then extending it to First Hill and potentially on to the Central District or Capitol Hill.

An ambitious vision. (Seattle Subway)

High Ridership

Jane Jacobs advised transit planners to look to their network’s busiest bus routes when deciding where to build rail. The RapidRide E fits the bill as Metro Transit highest ridership bus route, hitting 17,000 rides in 2016. (Highest ridership even without a N 38th St stop to serve Lower Fremont, the busiest neighborhood it crosses before reaching downtown.) The Aurora Avenue corridor carried 29,500 rides when Routes 5, 26, and 28 are included since they use it to and from downtown. For perspective, South Kirkland to Issaquah light rail is projected to get daily ridership of 12,000 to 15,000 in 2041, not even surpassing today’s RapidRide E.


Seattle Subway sites engineering challenges when explaining why Aurora Avenue wasn’t included in their plans. However, building elevated would seem to minimize many of those issues, such as tunneling under the Fremont Cut and avoiding the Bertha tunnel (by going over it rather than deep under it.) Building elevated rail rather than a subway on Aurora Avenue makes the problem more clearly political; we may have to take a lane away from general traffic to build the support structure for elevated rail and the associated stations. Sounds far-fetched. But what if Sound Transit did WSDOT a favor and purchased SR-99, or at least the center lanes, reducing their maintenance load? Controlling SR-99 right-of-way, Sound Transit would have the power to realize an elevated light rail plan. The pinch point would likely be the George Washington Bridge, which is aging and not necessarily light rail ready but perhaps could be reinforced and retrofitted. If not, perhaps we have to assess whether we’re approaching the time to plan a replacement for the bridge, which was dedicated in 1932.

The obstacle E rail still faces if it ventures toward First Hill is that old Seattle nemesis–a steep grade–and its newer nemesis: crossing I-5, but those challenges may not be so insurmountable. Compared to engineering deep tunnels and stations, smoothing out the grading of the elevated track toward Capitol Hill and First Hill doesn’t sound so bad. That routing would require gaining about 250 feet in half a mile. Making a 8% or 9% grade out of that is conceivable.

Another obstacle is overcoming aesthetic objections to elevated rail in dense urban areas. It’s clearly not out of the realm of possibility since Sound Transit prefers to use elevated rail in Ballard–although, side note: we may change their mind when the difficulties and expense of a second Salmon Bay bridge crossing become more apparent. People may associate elevated rail with the failed monorail expansion measures; however, when King County voters realize they can get more transit for their money, it may prove popular again.

Despite the obstacles, we must get our priorities right; Seattle’s rapid transit network must work for our densest neighborhoods first before seeking out new car-dependent hinterlands to tame with transit-oriented development. Before we tackle another Lake Washington crossing and build two lines to distant Woodinville, we need to get our central Seattle neighborhoods integrated into the network. Building elevated rail on Broad Street, Way, Second Avenue, Spring Street and Broadway could accomplish that goal. The improvement in service would be enormous since downtown bus routes are often stuck in debilitating traffic at peak hours, even before One Center City attempts to manage controlled decline of downtown bus service rather than a visionary transit plan.

Serving The Central Area With Rail

The E alignment is highlighted in hot pink. (Doug Trumm / Seattle Subway)
An alternate E alignment, with a diversion to Queen Anne, is highlighted in hot pink. (The Urbanist/ Seattle Subway)

Another benefit of this E alignment is it could interline with other lines like Seattle Subway’s Gold Line with service to the Central Area, or alternatively, the Gold Line would use elevated track to reach Downtown instead of tunneling up and around, shortening the distance of new tunnels to dig and track to lay. Instead of meandering up to Roy St and 15th Ave E, the elevated route could either take a direct line toward First Hill on Union or stop near Denny Way and Broadway to offer transfer to Central Link at Capitol Hill Station. An elevated Central Area line could potentially stop at Boren Ave and 16th Ave and 23rd Ave on E Cherry St before turning toward 23rd and Union or Mount Baker. That’s somewhat close stop spacing, but elevated stations are much cheaper than subway stations to build, opening up the possibility.

A Rail Plan For Seattle

With core downtown neighborhoods served and the eastern Capitol Hill, Central Area, Lower Fremont, Phinney Ridge, Greenwood, Aurora-Licton Springs, and Bitter Lake integrated into the rail transit network–in addition to the Ballard to Sand Point, Junction to White Center, and Ballard to Lake City lines already in Seattle Subway’s vision map–this ST4 package would have a lot to offer Seattle. Seattle Subway’s approach of focusing on King County may ultimately be the pragmatic path since the ST3 ballot measure had the most success there, while losing in Pierce County and breaking even in Snohomish County. Those counties may end up passing their own packages but scale of what’s needed is much larger in King County and rail is more clearly the solution for many King County corridors.

Seattle Subway helped ensure the ST3 ballot measure’s victory and their vision map will help inspire the grassroots movement to keep fighting for more transit in the Puget Sound region. A few improvements could help the vision map do even more to unite Seattle. To me the biggest improvement is turning the RapidRide E into an elevated rail line and extending it to First Hill.

Author’s Note: The earlier map created some confusion, so here is a map with a street grid for clarity. Alternate routes are possible, but this proposal centered on using Aurora Avenue right-of-way, until reaching the future (ST3) station at SR-99 and Harrison St, which will be underground and connect to Ballard.

The alignment uses Aurora Avenue until veering off on Broad St. (Doug Trumm/Google Maps)

Map Of The Week: Seattle Subway Vision Map

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.

Doug Trumm is The Urbanist's Executive Director. An Urbanist writer since 2015, he dreams of pedestrianizing streets, blanketing the city in bus lanes, and unleashing a mass timber building spree to end the affordable housing shortage and avert our coming climate catastrophe. He graduated from the Evans School of Public Policy and Governance at the University of Washington. He lives in East Fremont and loves to explore the city on his bike.

Inline Feedbacks
View all comments

A few thoughts:

“Jane Jacobs advised transit planners to look to their network’s busiest bus routes when deciding where to build rail.”

Sure, but we also have to keep in mind the speed of those bus routes. I think it is quite possible that you could come up with a formula that works really well for predicting ridership. Current bus ridership divided by average speed is a good starting point. The 44 and the 8 suddenly come to mind. They have top ten ridership, despite the fact that the buses are stuck in traffic. This makes them great candidates for grade separation. Once you have that grade separation, you might have ridership that can justify rail.

I think it also important to look at where people are actually taking the bus. Is everyone headed downtown, or are people going a mile or so down the road? Light rail stations tend to be expensive. Having fewer stops makes the long distance trip a lot faster. If everyone is headed to downtown, then this means people are willing to travel half a mile for that faster run. If, on the other hand, people are just going down the road, then they will want more urban stop spacing, similar to what NYC (and the E) has.

All of this suggests that while the E is a fine corridor, it might not be the best for transitioning to rail. Ridership is high because the bus is very fast and very frequent. It also makes lots of stops (although, as you wrote, it should stop around 38th, an idea many people — including me — have promoted for years). The lack of a stop there actually becomes less of an issue if you build a train that serves lower Fremont (just make a transfer to the train).

There is also the matter of cost. The main reason that Link travels next to the freeway once it gets to the suburbs is because it is cheap. But if fixing up the Aurora Bridge is expensive, then what have you gained? You have simply built a very expensive rail line that really isn’t much faster than the bus line it replaced. For some riders, the trip actually takes longer, as they spend more time walking to a station than they did before.

This is why I actually like the first map better (the one that confused people). This is the one that Bruce used, when he came up with his map. It actually complements the bus system really well, which is always a good idea. The best transit systems have fast trains *and* fast buses. The rail line doesn’t have to serve many places on Aurora. That is the job of the E. So, for intra-Aurora trips, you take the (frequent, fast) E bus. Same with getting to the east side of Queen Anne or a direct connection to South Lake Union. For longer distance trips to Belltown, you take the train. The train also connects you to neighborhoods that are extremely difficult to get to otherwise (by bus or car). With absolutely no traffic, a trip from 85th and Aurora to the top of Queen Anne takes 15 minutes ( With a tunnel, it would probably take half that. Saving that much time is what gets people out of their cars (or remove their Uber apps). It also means that transfers aren’t that big of a deal, which means you can draw from a much bigger set of people.

Of course, such a line is very expensive. But it is also worth it. Belltown is the most densely populated area in the state. Aurora is a major corridor, and the deviations from it are major, complementary destinations that have higher potential ridership. SPU, Upper Queen Anne and Lower Queen Anne are all great destinations. Extending to First Hill, Cherry Hill and Garfield High School are arguably a better value than just about anything we have built (other than UW to downtown). It is both urban and short, which means that once you spend the money, at least you have the ridership (and low maintenance cost) to justify it.

As far as elevated versus underground, I think you raise a very good point. Thus far, we have built elevated rail next to major freeways (so called car sewers). An attempt to go elevated in Rainier Valley failed because people didn’t want a noisy train cutting through the various neighborhoods. However, that will all change very soon, unless ST finds a lot of money. I’m not taking about Ballard — which makes sense for elevated, given the noisy trains and what is essentially a freeway nearby — I’m talking about West Seattle. The most culturally important and interesting part of the entire peninsula will have an elevated rail line running right through it. There will be complaints (justified, in my opinion). But when the dust settles, people will be more willing to accept elevated rail in the rest of the city, which means it is a possibility here. I’m not sure if it would actually matter, though. It might make sense to go under the ship canal, or over it. Either way, at some point you would go into Queen Anne, which means going underground.


This question is not specifically about this proposal, but has been on my mind. Wondering if anyone can fill me in?

I’m always curious why Greenwood is never squarely in the plans to have high capacity, high reliability, high speed transit. It it seems like it’s getting much denser really quickly, has good commercial infrastructure and would be a good fit for this kind of connectivity.

Ballard lines always seem to go north up 15th and then cut across to Northgate via Holman Road, the Ballard/UW line would be south on 45th. Link runs where it does east of I-5.

All the lines seem to draw a box around Greenlake and Phinney Ridge. This line would be a big leg up, and a one seat trip to Capitol hill would be amazing.

I admit that the area is well supported by the 5 and the E lines.

I live nearby, so clearly biased.

Is it an issue of location? Too close to Aurora to justify going out of the way? Easily enough connected to other transit corridors by foot/bus? Just curious why 85th/Greenwood has never been a target of any rail plans.


As an E rider I like this in theory, but I don’t like the way your proposed ROW pulls away from SLU, a much bigger destination than LQA or Seattle Center. Currently about 1/3 of the E line gets off the bus at Denny Way. If you wanted to provide some of the same benefits on the cheap, I would guess its possible to build true BRT (i.e., center running / barrier separated) on Aurora north of 46th and then force a transfer to the future Downtown-Ballard-U District line…which would avoid messing w/the Aurora Bridge.


North of the park, I’m not sure why you’d need elevated segments – the ROW looks to be about 75′ or more (6 lanes plus a center lane) so you might be able to cram in a center running 2-lane roadway while maintaining 4 through lanes…maybe…to be sure, lots of complexity in how you’d manage traffic operations and maintain LOS. South of ~65th, I’m not sure you can do much more than the existing curb lanes, which work OK but not great due to the narrowing of the roadway at the Fremont Bridge.

I saw the transfer you are referring to…the transfer penalty could potentially wipe out the travel savings for people headed to SLU who currently have a one-seat ride via the existing stops at Aurora / Denny, though maybe that is made up for by a more central stop to the area.


A true BRT option would likely involve more than just center running, but also a tunnel (to Belltown). At that point, Doug is right — you might as well add rail. For a branch and trunk system, BRT makes sense. But for a single, solitary line, rail makes more sense.

As I said above though, if you are going to spend all that money on a new tunnel, then you might as well add significant value to the line, and to do that, you deviate from Aurora north of the canal. Let the E do what it does well (go along Aurora) and build a different train line that connects lower Fremont, upper Queen Anne, lower Queen Anne and Belltown.

Ian Barrere

Where are the proposed lower and upper Queen Anne stops? The Google Maps version shows the alignment along Dexter (?).


Well, my suspicion us the difficulty of crossing bertha is overstated. Subway lines cross each other all the time. Our existing downtown transit tunnel crosses the freight rail tunnel. In 3d you have a lot of space to work with.

However, if you were going the elevated route through downtown, I think a monorail would make more sense. They can climb hills and make less noise. Also the monorail pylons only use one traffic lane for two monorail tracks. Not sure if that’s possible with elecated light rail.


pretty much anything monorail can do can be done with rubber-tired rail (see many metros in France, for example) which are more common and thus cheaper to buy than monorail. Monorail’s guideway is nicer to look at from underneath — that is pretty much its only advantage.

If you address the grade to First Hill, you can even revert back to steel-wheeled rail.

The advantage of a fully grade separated line is that we could deploy a fully automated system such as the one in Copenhagen.


Monorail is more or less rubber tired rail, so we are not in disagreement 🙂

I think the important point is that the existing light rail technology isn’t suited for an elevated route through parts of downtown that have extreme grades. However, we shouldn’t be tied to that technology for completely new lines that


Extend the Ballard line to Lake City, the West Seattle line to Renton, then build the 8 Metro, 99 Metro, and the Ballard/UW line and suddenly King County has five new grade separated east-west connections and another north-south spine. Also, almost every urban village would be connected or directly served by a stop. As Seattle exists now this would be a more or less complete rail network.

Can anyone speak in more detail about the technical challenges of building below grade through Belltown and First Hill?