Ballard Spur and Metro 8 Subway Serve Seattle Better than Interbay Light Rail


Reading the tea leaves of the Sound Transit Board, serving Ballard is a clear priority for Sound Transit 3 (ST3). But the Board evidently favors a Ballard-Downtown line via Interbay (I’ll just refer to this as Interbay henceforth for brevity). Ballard-University District (Ballard Spur) light rail might offer greater benefits and also leave enough money for the Metro 8 Subway. In combination, the Ballard Spur and the Metro 8 Subway will serve Seattle better than the Interbay line on its own.

The Metro 8 Subway (Orange) and Ballard Spur (Green) would both intersect Central Link (Blue) and offer connections. Peruse map here.

I made the case for building a Lake Union Loop by completing the Ballard Spur, the Metro 8 Subway, and Ballard-Downtown (via Interbay) light rail. Building those three lines might be too expensive for ST3, especially if the scope isn’t increased beyond 15 or 20 years. Even if we did have enough money, we’d still have to determine the order in which to build the lines. I’d argue the Ballard Spur and the Metro 8 Subway should be first because they are more beneficial than Interbay light rail alone.

The grade-separated Interbay routings are not cheap. A Salmon Bay crossing is part of the high cost. (Sound Transit)

Cost Differential

The latest study indicates that the elevated Interbay option with a new movable ship canal bridge would cost between $4.4 billion and $4.7 billion while the tunnel option would cost between $5 billion and $5.3 billion. Either option would operate best with a new Downtown transit tunnel, which would add another estimated $2 billion to the total project cost. Meanwhile, Sound Transit has pegged the Ballard-University District line around $3 billion. As an added bonus, this line would also connect Ballard and the University District to neighborhoods in between.

Travel Time

Sound Transit’s modeling predicts that the travel time from Ballard to Westlake Station will take approximately 18 minutes for an elevated line, or 19 minutes for an underground line. The Ballard Spur will be competitive with those times, even with the transfer to the Link spine. Here’s why:
  • From Ballard, the Spur would reach the University District in 7 minutes.
  • From University District to Westlake Station, Link will have a travel time of approximately 8 minutes.
  • The transfer penalty should average about 2 minutes at peak and 5 minutes off (since Link is expected to have an operating frequency anywhere from 3 to 10 minutes depending on the time of day).

Thus, at peak, the Ballard Spur could average faster times to downtown at 17 minutes. Off peak, the average travel times will only be marginally slower than Interbay.

Note that the Downtown-Ballard trip might be slightly slower since the Ballard Spur is expected to have a peak headway of 6 minutes and therefore a 3-minute average transfer penalty at peak. On the other hand, operators could hold the Spur train for incoming Link spine trains to optimize connections.

Through-routing the Spur?

To avoid the transfer penalty for Ballard trips, the Ballard-University District line could also just continue south to downtown Seattle, sharing track with Link spine trains. Theoretically, this would allow Ballard to Westlake Center time of 16 minutes, the fastest of all options. Sound Transit doesn’t seem to keen on this idea, apparently citing rail congestion and concerns about station design. However, plenty of systems operate at higher frequencies than every 3 minutes without issue. It may not be necessary if the transfers are efficient enough, but through-routing the Spur would seem to be the gold standard on travel time from Ballard to Downtown.

Metro 8 Subway

One of the biggest things the Interbay alignment brings to the table is a station in Uptown and either Belltown or South Lake Union (perhaps both with a somewhat squiggly alignment). The Metro 8 Subway could do the same thing for less money. The most important stretch connecting Uptown to Madison Street would be about 3 miles. It takes another 0.8 miles to reach Cherry Street and 23rd Avenue closer to the heart of the Central District. Since the length is about the same as the Ballard Spur, the cost should be similar.

However, since I’m gunning for at least 6 stations (two more than in the Ballard Spur study), the price would rise. Underground stations cost an average of $175 million for the University Link project. Sound Transit seems to be estimating higher costs in ST3 studies; although, this may be a function of their over-designed stations with oversized mezzanines. So, let’s add about $400 million to the cost for two more hopefully modestly-designed stations. Thus, it would seem a $3 billion Ballard Spur and a $3.5 billion Metro 8 Subway would still cost about the same as a Ballard-Downtown line with a new transit tunnel.

Metro 8 Subway opt 2
This routing of the Metro 8 Subway would serve Belltown with a station farther west in the vicinity of 3rd Avenue and Vine Street. The light blue line indicates the Central Link while the faint orange line indicate possible extensions for the Metro 8 Subway.

Regional Significance

Given how much these two short lines would improve mobility in Seattle’s densest neighborhoods, it seems a more judicious use of funds than building an Interbay-routed light rail line alone. The Metro 8 Subway hasn’t been officially studied, so it seems that Sound Transit doubts its significance; the regional significance, however, is clear to me.

This potential subway line would connect South Lake Union, a major employment center in Seattle, with the dense residential neighborhood of Belltown and the residential/cultural area centered around Uptown and The Seattle Center. A station near Madison Street and 23rd Ave E would also allow for connections to/from the planned Madison Bus Rapid Transit (BRT) line and provide high-quality transit service to the underserved Central District and portions of the east slope of Capitol Hill. Adding a station in the Cherry Street area would link the Central District even better, improving Sound Transit’s equity credentials.
The Metro 8 Subway would be a massive transit service upgrade. The Denny corridor is one of the most congested in the city due to interactions with busy I-5 and SR-99 on-ramps. Sometimes taking the King County Metro Route 8 bus is effectively no better than walking. From Capitol Hill to Uptown, the bus is scheduled to take 29 minutes at peak; often it takes even longer. Meanwhile, Google Maps says walking those 2.3 miles takes 47 minutes. A subway could travel that distance in about 8 minutes and be impervious to weather conditions.

Long Term Plans

Building the Metro 8 Subway would set the stage for the Ballard-Downtown line since it would take care of one mile of tunnel work and two underground stations. I’d wager that would shave almost a billion dollars off the price tag if Sound Transit was to build the Interbay route later on. Building the Ballard station of the Ballard Spur would also knock off the cost of another underground station ($200 million give or take) on the Interbay light rail budget.

These alternatives, as presented here, provide an intriguing alternative to the current Interbay plans for approximately the same amount of financial investment. Furthermore, the Ballard Spur and the Metro 8 Subway would connect more diverse, transit dependent neighborhoods, remove a lot of inefficiencies present today, and create a more dynamic transit network.

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

Agreed about Metro 8 & UW-Ballard. Though when it comes to Ballard-Downtown my favorite by far is Option D on ST’s recent study of this corridor. Option D was Belltown-LQA-Queen Anne Hill-Fremont-Ballard, all underground while avoiding the nothingness in Interbay and also to a lesser degree along Westlake by the lake. Option D hits all the walkable places and urban villages like ducks in a row. I would hope this would be eventual route chosen for this corridor and tie it into Metro 8 & UW-Ballard.
Just a thought, but what about instead of 23rd Ave the subway tunnel run under 12th Ave? Yes there is 23/Union node but that’s about it. Just seems to me 23rd is much less dense and more surrounded by single family residences than a further-in route, under say, 12th. A route under 23rd is going to need a serious upzone that’s for sure.


I like the idea of supporting a Metro 8 subway because trying to go anywhere northwest from the CD (to SLU or Queen Anne, or Ballard for example) is very difficult. Seattle definitely needs rail lines with higher density of stops within the city to make rail more useful in reducing car usage. One minor suggestion is that the 23rd and Cherry stop center between Union and Cherry since Union is a substantial throughfare and will become more so in the future.

Chris Burke

This is a good proposal; better than the city’s. UW-Ballard is the best investment ST3 can make. Ballard to downtown in a second transit tunnel to West Seattle is hugely expensive…on the order of $10 billion. I doubt any of that would get approved if all of it wasn’t approved. ST board minutes indicate their first priority will be finishing the “spine.” If that holds true, there won’t be enough money left to build Ballard-West Seattle, but there probably would be enough for Ballard-UW.

Howard Metzenberg

The big problem with the Ballard to Downtown route is that it must pay for crossing the ship canal all by itself! Seattle’s Ballard Bridge is 100 years old and has many traffic and safety problems that need fixing.

Sound Transit should postpone the Ballard to Downtown route until a light rail crossing of the canal can be part of a comprehensive solution to the Ballard Bridge problem in general. It will be much less costly to cross as part of a new bridge that safely accommodates motorized, bicycle, and pedestrian traffic. When other partners and stakeholders (the City of Seattle, Washington State, the federal government) are ready to build such a bridge, we can hope they will share an interest in building a fixed bridge at higher elevation that does not have to close for leisure boat traffic.

A tunnel under the ship canal at Ballard must necessarily tunnel under adjacent neighborhoods to the south in Interbay and to the north in Ballard. Such stations will be awkward and expensive to build and will fit the needs and topography of these neighborhoods poorly. Imagine descending into and ascending from a deep tunnel just to cross the ship canal with a bicycle.

Perhaps the Metro 8 route could end as a stub with an elevated station at the south end of Interbay. Eventually, perhaps with ST4 in 20 years or so, Sound Transit could extend this line north across the ship canal as part of a comprehensive Ballard Bridge rebuild. In the meantime, Ballard, Fremont, and other neighborhoods north of the ship canal would have long-established connections to Central Seattle through Ballard-UW-Link, while Metro 8 would have long been serving populous neighborhoods south of the ship canal.


Thanks for raising some very good points. I agree completely with your first three paragraphs. Too often we build things in isolation around here. The stadiums are probably the best example. Two new stadiums to hold games and crowds of roughly the same size and then we lose out on our longest running sports franchise. Oops.

I’m not so sure about what to do with the western end of a Metro 8 subway. A lot depends on what else we build. If we build a WSTT, then it would certainly include Belltown, which means that a Metro 8 subway could head out towards Interbay. But if not, then you are barely skirting Belltown, or you are barely skirting South Lake Union, which is what the author does. Unless you save a bunch of money with cut and cover, going along Denny is not very good — it is too close to the Westlake Station to add full value. So that means you want to go a few blocks north of there, which means that if you serve Belltown, you are looping around towards downtown. Perhaps the best way to handle things is just to end in South Lake Union. Make the tough decision after that.

I personally would build the WSTT and run the Metro 8 subway so that it connects to it. This means it ends at Aurora and roughly Thomas. The WSTT, meanwhile provides the help where it is needed most — getting through the city. When you fix the bridge is when you decide what to do. Either add dedicated bus lanes (along with bike lanes and a decent pedestrian crossing) or a lane for the train. Until then you still have much better transit for this city at much less cost than Interbay light rail.


Thank you Doug for a really excellent proposal.

From ST planning documents, it’s clear that much of the cost of connecting Ballard to Downtown is in crossing the ship canal. What’s more, potential crossing solutions (like an elevated bridge or a deep bore tunnel) often interfere with and limit possible connections north of the ship canal. This has lead to several inferior crossing solutions as options. Only a tunnel under the ship canal can be fast and reliable. What to do about this? Don’t build it!

Connecting Ballard to Downtown via LRT through Interbay is a waste. The Interbay corridor is full of dispersed light industry, and very little housing. Since this industry is decentralized, few employees will want to walk from a light rail station to their job sites. Nor will nearby residential neighborhoods use the station. Most of Magnolia is out of walking range and Queen Anne residents aren’t going to walk to the back side of the hill to catch a train Downtown. Magnolia and Queen Anne riders would probably rather board one bus and sit in the same seat all the way to Downtown.

Ballard would be better served by having a crosstown train that also serves Fremont and Wallingford, with all riders transferring at the U-District. This line could be extended north from Ballard to Greenwood and beyond, or east from the U-District to University Village, where it could emerge and run aboveground or remain underground through a less expensive cut-and-cover tunnel and continue to Children’s Hospital, a major employment center. Ballard riders would reach Downtown by transferring at U-District, and many other transit-friendly neighborhoods along the line could also be served. Ballard and all of the densely populated neighborhoods of north Seattle would be richly connected.

A crosstown “Metro 8” line can serve densely populated Belltown, Lower Queen Anne, and South Lake Union much better than a few stations along a dedicated Ballard-Downtown route. At the same time, transfers from North Seattle, the East Side, and Southeast Seattle are shifted from Westlake to other stations (Capitol Hill, Judkins Park, Mount Baker).

Downtown probably still needs another transit tunnel. Since the crosstown routes would create neighborhood connections, a second downtown transit tunnel (under 5th?) could be used for buses only, especially regional buses. Two downtown transit tunnels would operate much more efficiently if one was dedicated to buses and the other was dedicated to trains. Buses and trains, with differences in how they load and how long they should wait, mix poorly in transit tunnels.


I agree with all of your points. I would add that there are a couple reasons why UW to Ballard is a much better value than Interbay light rail: First, It is just about as fast to get downtown that way (

Second, Ballard to UW complements the bus service much better. It is relatively fast to ride a north-south bus on the corridors in the area. This means ridership would extend both directions along the route. Someone along 15th NW, 8th NW, Greenwood/Aurora or Wallingford/Meridian would take a bus, then transfer to the train if headed to UW, Capitol Hill or downtown. The combination of Ballard to UW light rail and a second bus tunnel (WSTT) would provide for a much better transit network for the area than the alternative ( You would be able to get to almost anywhere in Seattle very quickly using a combination of buses or trains. We would jump from a half-ass combination of services to a real transit network (similar in scope and functionality to Vancouver BC — which would be a huge leap). The one area left out would be the area covered quite nicely with a Metro 8 Subway.

As far as extensions go, the obvious one is to the west, to 24th. This gives you a station much closer to the cultural center of Ballard (old Ballard) and provides more direct, fast service along 24th. This is an area that is already reasonable dense, and is growing rapidly.


I completely agree with this plan. The Metro 8 Subway line is critical and should be a priority.


Excellent post. I agree.

Several people have come to a similar conclusion. They also add the WSTT. Basically, a lot of us believe that there are only a handful of major projects that we need to build in Seattle:

2) Ballard to UW subway
3) Metro 8 subway

Other improvements are also needed, but do not cost a huge amount (e. g. making West Seattle buses fast would not cost a fortune). Here are some links that talk about the first two projects:

The second one (the one I wrote) goes into great (if not tedious) detail as to why these projects are better. When you do the math, Ballard to the UW only takes a couple minutes longer than Interbay (assuming the stations that SDOT specified). That leads to several conclusions. Basically, the only trip that is much better with Interbay light rail is Ballard to Queen Anne. The number of other trips that are much faster (e. g. Fremont to downtown) greatly exceeds that. The two articles contain links to other points as well. West Seattle BRT (including the WSTT) would be a bigger improvement for West Seattle residents than a single light rail line (because it would avoid the transfer penalty and serve more people).

I would say that there is no consensus with how a Metro 8 subway should look. It isn’t obvious. Serving Belltown would appear to be essential, but then what? Curving around and going downtown would appear to make sense. Should that connect to the WSTT then? Or maybe just share the existing tunnel.

Unless you are thinking this could be done as cut and cover, I see no reason to follow Denny. Quite the opposite. I would go a bit north of Denny, to actually serve the heart of South Lake Union. This puts the station a fair distance from Westlake, which would increase coverage. After Capitol Hill, you might want to go closer to First Hill. Then add a stop on 23rd, so that it connects well with buses. Something like this:

As I said, though, picking an ideal route for a “Metro 8” subway is not obvious. What is apparent, though, is that no matter how you pick a route, it should be a higher priority than West Seattle light rail. When you compare the costs, the number of people served, and the alternatives, it a much better value.


It probably is more expensive to build the WSTT for dual use, but cheaper in the long run. It is important to remember that the transit tunnel has saved a lot more time for bus riders than it has for train riders. That will probably be the case for the next twenty years, if not fifty (even as buses are kicked out).

Mixed operations are not very good right now, but I can’t help but think that the main reason is that neither agency is that concerned with making it work. They know the situation is temporary.

In the case of the WSTT, I don’t think we can assume that. If anything, it is the opposite. The demand from the south is simply too dispersed. So buses could be expected to be there a while. It is also clear that at least through the tunnel, the buses would operate as BRT. You would have off board payment and level boarding. This means that a bus would be able to have shorter dwell times than a train, and thus be able to get out of the way much faster.

There are really three possibilities. One is that it is simply converted over to rail. The downtown tunnel is really a strange project. There is just enough of it built to force joint operations, yet not enough demand to justify kicking out the buses. If this was built with high priority projects first (the way most cities build their light rail) then you would start with U-District to Rainier Valley, and kick out the buses the day it opened. But instead we have to wait until the next phase, which also includes the airport (built first some reason) and Northgate (a nice to have).

The second possibility was mentioned (joint operations). The third is to just keep it as a bus tunnel. But then you have the problem of where to end the Metro 8 subway. One possibility would be in Belltown. This matches the map I mentioned, but it means that a train ends there (and doesn’t share the tunnel). It would mean that someone headed to the South Lake Union stop from a different part of downtown (e. g. Madison) would have to transfer. But transfers here would be easy. Trains don’t always run that often, but buses running through the WSTT would run quite often. Headways would be measured in seconds, not minutes. Meanwhile, you would still have a good chunk of the city with direct access to South Lake Union, and most of the city with a different transfer. If you are coming from the south or east, you transfer at Judkins Park. From the north, at Capitol HIll. It is only the folks from other parts of downtown and West Seattle that would have to transfer. As mentioned before, for most people in West Seattle, a direct bus downtown (and into the WSTT) is better.

This is why I think these set of plans makes a lot of sense. The WSTT, a Ballard to UW subway and a Metro 8 subway pretty much ought to do it. Nothing else in terms of big tunneling projects comes close to adding that much value for the money.


rossb’s map bypasses the heart of the Central District, which is very problematic. A line running along 23rd south from Madison seems really important to serve that dense and transit-needy community.