Map Of The Week: Seattle Subway Vision Map


Seattle Subway, a local grassroots transit advocacy group, has a new regional vision map out that charts where they think future light rail lines and extensions should go after Sound Transit 3. The map emphasizes new light rail lines and extensions between Downtown Redmond and Woodinville, Ballard and Kirkland (via Sand Point), Ballard and Woodinville (via Crown Hill and Northgate), and a loop line in central Seattle. The map also draws extensions in Tacoma and Everett, and between West Seattle and Renton (via Burien).

The new Seattle Subway 2017 regional vision map. (Seattle Subway)
The new Seattle Subway 2017 regional vision map. (Seattle Subway)

I recently caught up with Keith Kyle, Executive Director of Seattle Subway, to find out what went into designing the new vision map. Kyle said that the map reflects many aspirations of Seattle Subway’s membership, yet is grounded in political, technical, and funding realities. The regional vision map is one tool that Seattle Subway uses to convey its message to communities and supporters, Kyle said. Seattle Subway has kicked off a new campaign this summer to engage and activate supporters in advance of detailed project planning for lines and stations to be built in the next major phase of Sound Transit’s light rail expansion progam.

The 2015 regional vision for "ST Complete." (Seattle Subway)
The 2015 regional vision for “ST Complete.” (Seattle Subway)

Despite being less geographically accurate, the refreshed map should be fairly familiar to eagle eye readers since it is largely based on the “ST Complete” map that Seattle Subway put out in 2015. Kyle said that some inspiration for the vision map also comes right from Sound Transit’s long-range plan (e.g., Northgate to Woodinville, Ballard to University District, and West Seattle to Renton) and planned high capacity transit corridor studies (e.g., Everett to North Everett, Northern Lake Washington, and Tacoma Dome to Tacoma Mall) funded in Sound Transit 3. Sound Transit considered many of these alternatives during development of the Sound Transit 3 expansion package. Big ideas like Ballard to West Seattle and South Kirkland to Issaquah won out, but not as big ideas as many had hoped.

Sound Transit's adopted 2014 Long-Range Plan map. (Sound Transit)
Sound Transit’s adopted 2014 Long-Range Plan map. (Sound Transit)

Significant Changes to the Vision Map

There is a variety of significant differences between the ST Complete and new vision maps, which includes the operation of lines, cutting out and adding some destinations and stops, and creating new corridor concepts. But there a few other basic changes as well, such as renewed approaches to identifying lines in operation, planned for construction, and envisioned in the future, Tacoma Link and Sounder commuter rail drawn in with their extensions, and existing Amtrak lines getting minor honorable mentions.

New Operational Structure

The new map presents a much more complex operational makeup of light rail lines using joint corridors. For instance, the new map envisions five lines converging at International District/Chinatown Station, though such an interchange would necessarily require multiple platform and track sets to facilitate the interchange. The 2015 map only pegged four lines converging for the station there.

In the new vision map, three lines would also emerge north of Downtown Seattle–two heading toward Ballard and another looping around North Capitol Hill and the Central Area. The Ballard-Downtown corridor would split into north and east of Central Ballard with the Green line heading toward Woodinville and Purple line toward Issaquah via Sand Point. The 2015 map, however, only hand one Ballard-Downtown line with a secondary Ballard-Woodinville line terminating in Central Ballard. The new Green Line threads together portions of the 2015 Red and Orange Lines creating a single line running from Woodinville to Sea-Tac Airport serving Ballard and Downtown Seattle in between. Likewise, the Purple Line blends together portions of the the 2015 Red and Green Lines by picking up a cross-lake connection so that service is provided seamlessly from Issaquah to Ballard via Kirkland and continuing on to International District/Chinatown Station.

Seattle Subway’s envisioned network mirrors the complex type of service pattern found in a system like the DC Metro, where lines share common corridors in core areas to create frequency, allow for practical transfers, and provide for deeper coverage.

Changing Stops and Destinations

A few destinations and stops were cut from the new map, including a stop at Tacoma Community College (which will be served by the Tacoma Link streetcar), Tukwila Town Center, and select stops on the West Seattle-Renton and South Kirkland-Issaquah line. Some of these changes reflect new realities from planned investments in Sound Transit 3. Meanwhile, Seattle Subway’s vision would include a number of added stops, such an S 38th St in Tacoma and a terminus at Renton Landing. The latter is a burgeoning urban district for housing, services, and jobs. As a terminus, the station could be a good transit center location that connects with local pulse bus service to districts to the north and east, adding to its overall utility.

New Service Concepts

The new vision map brings back the concept of a second cross-lake light rail line (I-90 being the other). Seattle Subway’s concept provides the cross-lake connection via Sand Point and Kirkland, similar to ones proposed in the past. The map also highlights a new way to thread together cities on the Eastside with an extension of the Seattle-Redmond line to Woodinville. Previous concepts have generally focused on such a connection via Kirkland, but this new one is interesting since Kirkland trips could require a transfer via Bellevue or two transfers in Seattle to head toward Woodinville. Admittedly, that is probably not a high ridership corridor today, but with the growth of areas in North King County, Southwest Snohomish County, and other parts north and east of the King County line, it could prove a promising and useful service.

Where Is First Hill?

First Hill’s absence from the map is notable. Kyle explained that crossing I-5 for University Link was an immense technical challenge and expensive alignment to construct as part of the initial Sound Move program. Doing it again as part of Sound Transit 3 would be, perhaps, even more complex since light rail would have to dive under three levels of roadways–twice–within a short distance and that a station in First Hill would have to be deep-level. The costs to do that would undoubtedly be high even if technically feasible.

Instead, Seattle Subway is hedging their bets for a station near Fifth Avenue in the office core of Downtown Seattle that would also offer a connection to the Madison RapidRide line. This would provide better coverage and access to First Hill while creating a larger light rail market for Downtown Seattle, but Seattle Subway is still hopeful that technical analysis from engineering and budgeting experts may cast a more favorable light on a First Hill station since serving the district is a regional priority. It is also possible that Seattle Subway’s loop line–which links Mount Baker with the Central Area, North Capitol Hill, South Lake Union, and Downtown Seattle–could make a deviation to partially serve First Hill, but the utility of that is not entirely clear right now.

If You Map It They Will Fund

I asked Kyle why there were few extensions envisioned in Snohomish and Pierce counties. Kyle said that light rail, as a regional project, will probably complete its destiny after Sound Transit 3 and that specific subareas, such as those in King County, are more inclined to do big things on their own. Kyle also admitted that his organization was less familiar with local constituencies in Snohomish and Pierce counties. So while there may be some future for more light rail expansions in those areas, Kyle was not certain what those might be.

Readers will undoubtedly have their own quibbles (feel free to share them in the comments) with Seattle Subway’s new vision map. But it is a bold statement on what a more extensive rapid transit system could look like in the region. For that reason, it will be a powerful tool to frame future planning discussions in the minds of the public, planners, and politicians in the years ahead.

Seattle Subway will hold a fundraiser and straw poll tonight from 7pm to 9pm at Hattie’s Hat in Ballard. The straw poll will center on primary candidates for mayor and city council in Seattle.

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.

Stephen is an urban planner with a passion for sustainable, livable, and diverse cities. He is especially interested in how policies, regulations, and programs can promote positive outcomes for communities. Stephen lives in Kenmore and primarily covers land use and transportation issues for The Urbanist.

Inline Feedbacks
View all comments
frank discussion

Obviously missing is the Lynnwood – Bothell and Renton – Bellevue segments, as anybody who’s driven I-405 during rush hour recently and which BRT is a half-measure, since there’s close to that level of bus service there today. I-405 should have had light rail with ST-3, but the board made the excuse that they had agreed not to do that, which they then conveniently forgot about in proposing a line from South Kirkland Park and Ride that partly duplicates East Link while traffic between Issaquah and Renton has jumped 95% in recent years.

Steve Corley

Late to the discussion. The Kirkland/Sand Point crossing is ridiculous. A new bridge? Loop from S. Kirkland, through 68th (Google), Kirkland Center (brand new office/res development) then east on 85th to Redmond to connect to the Redmond line. If you want another bridge crossing then it has to be 520, way out in the future. Sand Point/Magnusson is not close to dense enough for a station. Buses are the solution for that area. Running anywhere near a shoreline will not let you get enough density because a large part of your radius is water.

Stephen Fesler

Ross –

You tripped the automatic filter and used a blacklisted phrase (an admittedly silly thing to filtered out). Please take the presumption that moderators are not arbitrarily tossing comments into the ether. This happens from time to time since Disqus’ filter system can be somewhat sensitive. Just shoot us an email if you have concerns with comments not appearing. Anyway, yours is restored.


Anyone know the reasons for the 8/gold line skipping the Cap Hill station? The way this map is designed Roy/15th seems too far north of John – but the distance is only a quarter mile. Seems totally possible for the line to hit both stations – and if for some reason it can’t (maybe track geometry?) why prioritize less density/connections for a station that’s already built? This means a ride from UW to the CD would require getting off at John and walking/bussing uphill to the station on 15th, or riding all the way to westlake and transferring.

Also if Sound Transit does decide to use Harrison/99 as the South Lake Union stop, it would be tragic to bring the 8 line that far west and not add a more central SLU stop, either on Fairview or further north on Westlake.


I had the same question last night, when I wrote my other comment. I have no idea where that comment went, though. I was quite critical about this proposal (I find it ridiculous — and skipping the connection is one example) so maybe the editors of this blog don’t want dissent.


One more route idea, though not necessarily one for ST4:

Run a line down 99 from lynnwood, hit greenlake, lower fremont, westlake then connect it to belltown, then run a little ways downtown, then up madison to replace the madison BRT.

That should be ST 5.

It would connect the missing segments on seattle and allow more TOD in snohomosh county. The big problem with the current lynnwood link extension is that it runs along the side of the freeway, so you can’t build that much next to it.


This is something I’ve been thinking about too! It would round out all the big (downtown) misses from ST3 and the current Seattle Subway map. My alignment is a little different, but a 99 -> Belltown -> First Hill line just makes too much sense.
comment image


Belltown makes sense, but you would have a very tough time replacing the Madison BRT. There are just too many stops on Madison. It would be an extremely expensive proposition to add all those stops if you built a tunnel, and a train can’t make it up on the street. Nor would a surface train be any better than a surface bus, unless it turns out the Madison BRT is so popular that buses are full even if you run them every couple minutes. Also keep in mind that a surface bus means that getting from the stop to your actual
destination is much slower than if you took the bus.

First thing you need is the Metro 8 subway. That will cross Madison (but it isn’t clear where). That gives you two stops on Madison — one downtown (with the ST3 line) and one well up the hill. At that point you have a good argument for simply staying put. Even if a connecting bus isn’t super fast, it won’t matter for the relative distance — what would matter more is headways. The Madison BRT would likely have better headways, thus making the benefit of a line heading up Madison fairly marginal.

I suppose you could have a line that works as a fast connector. It would go from one Madison line to another. This wouldn’t save much time coming from the north or the south, but would save time from other parts of downtown, depending on how the Metro 8 is connected to the rest of the line.


Well, I’m more thinking this is for ST-5 than ST-4. So sometime in 2060-2070. At that point I think the population density will be a lot higher and I think we will be struggling to have any major route handled by buses alone. If Seattle stops growing today, the line I’m describing isn’t necessary; however, I don’t think that’s likely.

As far as stop spacing, it depends on the location:
1. For central Seattle, we just need light rail with more stops. The current stop spacing, where capitol hill only got a single stop, is just ridiculous.
2. For long distance lines, the existing plan is to keep buses running in parallel. This is the plan for the Ballard line. Some version of the D will still be running parallel to it. It kind of simulates having both a local and express line on a subway.

My idea to go up 99 is to allow more development in snohomish county. The ST3 line up I-5 was cheaper to build, and hits a few key locations; but it’s pretty limited in terms potentially for TOD due to the freeway and various parks being in the way. 99 to lynnwood is 10 miles of developable land.


I’m also not sure if an Aurora rail line would make sense either. If you run it on the surface, it has all the same issues as a bus. If you bury it, then it becomes very expensive. Even just adding another bridge for it becomes very expensive.

Meanwhile, you really don’t have huge density on Aurora. The reason the E is so popular is because it is so fast. A grade separated train (or bus) would be faster, but not that much faster. If you are going to spend the money on grade separation (and I think you should) then it makes sense to build lines that are much faster than the fairly popular surface buses they replace. Good examples of this are the 8 and 44. A Ballard to UW bus would be faster than driving, even at noon! This gets people out of their cars, as not only is a trip involving two stops very fast, but with transfers it is fast. For example, Ballard High School to UW would be faster than driving, even though it involved a bus, a transfer, then a train ride. You really can’t say that about a train on SR 99, since 99 moves fairly quickly most of the day (and the bus lanes move even faster).

One way to measure these things is to look at time saved per potential rider. In the case of a 99 bus, the number of riders is fairly large, but the time each one would save is relatively small. If you take that measure (time saved per rider) then divide it by the cost, you come up with precisely the same measurement that the government used to use when deciding to allocate grants. I’m afraid that a SR 99 train wouldn’t perform well on that measure.

On the other hand, the E is our most popular bus, and if it continues to keep adding riders, we may want a train just for the added capacity.


I agree with Ross – the E has room for more bus improvements, but works really well along a long, linear corridor. The stop diet that a rail replacement would require would, IMO, result in inferior service compared to the existing E. The only real reason to upgrade the Aurora corridor into a rail corridor is if there are capacity issues, but I think simply extending the Ballard line north until it curves and has a station at Aurora (the most likely extension) should resolve that as riders north of that station transfer, freeing up capacity farther south.
(For example, we should see the same thing with the 40, where the Ballard station should relieve capacity issues in the Fremont-Downtown section because Ballard-Downtown trips will likely prefer a bus-Link in Ballard over the longer, one-seat ride through Fremont.)


The biggest weakness of this map strikes me as the eastside. Most of this would be built in the 2050’s. I think by then the eastside will be heavily urbanized on the scale seattle is today. I think the eastside will need considerably more transit.

Also, if you are going to do the sandpoint crossing rather than use 520, why go to bellevue? It makes no sense. Bellevue is already accessible via the i-90 crossing. It really only makes sense to cross at sandpoint if you are going to continue on to redmond. In that case you are providing fast access to microsoft from north seattle and lynnwood.

If you don’t think a faster connection to microsoft is worth it, then just cross at 520, which already is designed to be upgradeable to carry light rail ( though that won’t be cheap either, it will be much technically simpler than a tube at sandpoint).

warren trout

The Chinese will build a city of 10 million in less time than we will spend planning this rail. We suck.


Not “one of”, but *the* most densely populated area in all of the state. At least according to the last census (the really dark block in this map — the only one with over 100,000 people per square mile — is in Belltown.

Chloe P. H. Lewis

“Seattle Subway’s envisioned network mirrors the complex type of service pattern found in a system like the DC Metro”

I like the following optimistic sentence about how that _should_ work, but this decade the DC Metro plan doesn’t sound like a good one.


I’m not sure if I follow you. If you are suggesting that this map looks nothing like the DC Metro, then I agree. The DC Metro covers all major parts of the city* with criss-crossing lines that enable people to get around quite well within the city. It has just enough tentacles to some of the more popular suburbs to serve them as well, thus providing a very effective system for everyone. This has produced one of the highest per capita subway usage in North America, and this looks nothing like it, since it spends huge amounts of time dithering around in the suburbs, and manages to miss important connections in the city.

If you are talking about the maintenance problems that exist within the DC Metro, then that is irrelevant, except to say that this would probably encounter the worst of both worlds. The failure to properly connect up the central city will lead to low ridership, and the seemingly random lines throughout the suburbs won’t make up for it. This will lead to low fare recovery. Meanwhile, the maintenance cost is generally related to the miles of subway you build. This means that spending billions sending multiple lines out to remote, underpopulated areas will likely lead to the same sort of operational difficulties that the DC Metro has.

* With the exception of Georgetown, which for whatever reason didn’t want the subway.

Andre Tsang

What connections major connections is this map missing?


I would imagine the deep-bore SR99 tunnel is going to make this difficult, if not impossible. Best bet might be rehabilitating the Monorail (!, yes I said it) for local transit use.


The Sand Point crossing seems like a moon shot, funding-wise. As awesome as it is, it will no doubt cost an arm and a leg to build. Where would the money come from? A full federal grant isn’t happening with the current administration, but might be a possibility with the next. Self-funding would require tons of bonds, which require a 60% majority to pass. Definitely possible, but it’ll be a tough sell for Snohomish and Pierce county voters. Is there a way King county could go at this alone? Fund the crossing purely with King County taxes and bonds, then “contract” Sound Transit to operate?


Check out Sound Transit’s concept of Subarea Equity. It basically means the Eastside has plenty of tax base to (help) pay for the Sand Point Crossing.


I’m well aware of Subarea Equity, and I’m sure between the City of Seattle and the Eastside there’s plenty of tax base to pay for this thing. I’m saying the hard part would be getting a measure passed to fund construction. Just because voters in the other two counties aren’t paying for this doesn’t mean they can’t be goaded into voting no on an ST4 measure that has this included.


The tougher part is selling this to both cities. Why on earth would Seattle chip in for this? Seriously, I have friends and family that live in Kirkland, but this is way too expensive for what it delivers. Seattle gets three fairly weak stations in one of the lowest density areas of the city, and a connection to Kirkland. That is just a terrible value for what this would cost. Meanwhile, they don’t even extend light rail where it will be needed — to the west. West Ballard (24th) has way more people per square mile than anywhere in Kirkland, and serving them would be way cheaper. Oh, and this doesn’t even hit the most densely populated part of Kirkland, which is Juanita. Which means that this would be a very tough sell to Kirkland.

No, what Kirkland (and Redmond) could really use is a much better connection to existing Link. The 520 bus lanes are fast, but getting from 405 to 520 is tough. Either the state of Sound Transit needs to build ramps connecting the two HOV lanes. Then of course you need a better connection from 520 to the UW station. Maybe another bridge, or even a tunnel. That would be extremely expensive, but way cheaper than this, and result in much faster times from a typical neighborhood in Kirkland to the UW (and in turn, downtown Seattle).


WSDOT’s 405 Master Plan includes HOV ramps for both the 90 and 520 interchange. Rebuilding each interchange is roughly $500M. It’ll happen someday, if only because the existing ramps can’t handle the traffic, but I don’t think it needs to be a part of any transit plan – a one or two station extension north from the existing South Kirkland station would be nice, but as Ross says, anything more would be overkill.

Andy Barr

> A full federal grant isn’t happening with the current administration

Why not? The President and his Admin are looking around for infrastructure projects to fund.


Cool to see how the map has evolved. obviously a lot to unpack here, so I’ll just pick a few interesting ones outside of Seattle.
I’d love to see a Redmond-Issaquah line, very similar to what Zach proposed in his “grand bargain.” No changes to stations & alignment, just a different service pattern. I’m sure Kirkland would rather have the 1-seat ride to downtown.
For Burien-Renton, I’d strongly recommend checking out the ST alignment study from a few years back, which included a Link station at the Tukwilla Sounder station, which opens up all sorts of transfer opportunities, especially once Sounder gets closer to all-day service.
Very interesting they chose to connect Bothell, i.e. all of Snohomish county, via Redmond, rather than via Kirkland, which is would be more direct access to Bellevue & Seattle. IMO, Bellevue is the bigger destination than Redmond/Overlake, plus I think the 405 corridor is preferable from a TOD perspective than a Bothell-Woodinville-Redmond line that basically follows the edge of the UGB. Especially if there is a UW-Kirkland lake crossing, as that turns the Kirkland station into a strong transfer node.


Kirkland’s very vociferous opposition to utilizing the Eastside RAIL Corridor for its envisioned purposed, sealed their fate as a transit desert.


Yes, Kirkland had the audacity to hire independent experts to come up with a plan that would benefit more of its citizens, and the region as a whole. Then it went ahead and promoted this idea — an idea that anyone with an open mind would realize is not only cheaper, but much better. Silly Kirkland. Just play the game, and beg for ridiculously expensive and inappropriate rail, because, well, that’s how you are supposed to play.


comment image

Mike Carr

What does it take to move a thousand people? 15 years of waiting for Light Rail, then you can move from station to station.


Building infrastructure takes time. And every trip whether it’s by car, bus or train is multimodal with the end mode being walking. Building stations in dense walkable locations is the best use of our resources and planning for our cities. A train can move 1000 people at a time. Taking the current segment of the Link Light Rail from Downtown to Angle Lake, with 3 minute head ways that’s 20 trips per hour moves 20,000 people per hour per direction versus 300 buses to serve the same number of people. It is a far better use of those 300 buses to serve people to get to high capacity rail lines.


Perhaps you meant “independent” consultants who create the conclusions their clients want.

I see you are “railing” against density in other posts. The truth is, there are 1.5 MILLION people coming this way in the next decades. They are not all going to fit in Seattle. Nor should they. That means the many communities in the region must plan and prepare for growth, including Burien and Kirkland. Creating more compact, walkable communities connected by high capacity transit is the best way forward.


When it comes to envisioning these sorts of lines, I wonder what sort of city you are envisioning. Right now, just about all of the population density is within the city. The only exception is parts of Bellevue, and they will be served quite well with ST2 projects. So now you think that building very expensive trains — whose primary advantage over buses is capacity — to far flung, and low density areas is a great idea. Why? Do you think that Kirkland or Burien will suddenly become Brooklyn? I just don’t see it. Seattle is growing much faster than the suburbs, so if the suburbs actually become large enough to justify that kind of rail investment, the city of Seattle will be twice the population of San Fransisco, and need a much more extensive set of rail projects than the ones shown. You would basically need a network like the old Seattle Subway map (the one that existed when Seattle Subway meant a subway for Seattle, not the greater Puget Sound area). Even then you would have problems with capacity on key lines (like UW to downtown) and need more than six car trains. You would certainly need to drop the headways under three minutes.

I don’t think any of that will happen though. Kirkland will not need trains to get around — certainly not trains heading to Redmond. What Kirkland (and the entire suburbs need is much better bus service).

Here is an example. The map shows a train stop at Totem Lake. Fair enough. Yet right now the bus service from Totem Lake to Seattle really isn’t very good. There are a handful of commuter buses heading from Totem Lake to Seattle in the morning, and that is it as far as a direct connection. Either the bus slogs its way through Kirkland (on surface streets) or you go through downtown Bellevue. Even in the afternoon rush hour, there isn’t a direct bus. All you really need is more bus service, a flyover ramp connecting the I-405 to 520 HOV lanes, and a better way to connect to the UW station. This could be expensive. Maybe you build a tunnel from 520 to Husky Stadium. But that would still be way cheaper than building an additional rail line. It would also be way more popular.

Subways are only popular in densely populated areas. There is no city, anywhere, where the suburban rail line (whether a
commuter rail line, or a distant light rail line) is as popular as the inner city one. Subways are also very expensive to build. Suburban locations do best when they connect to those areas via express buses or commuter rail. Not only are they cheaper, but they provide what people are looking for when it comes to suburban transit — not a connection to the various neighborhoods along the way, but an express right into town, where they can (hopefully) connect to fast and thorough transit network in the inner city.


The cities I’m envisioning match the PSRC growth regions. For example, Kirkland has committed to growing Totem Lake, so a long-term vision should match that.
Totem Lake will get good service to downtown when the 405 BRT opens, which will have an easy transfer to East Link in downtown Seattle. That’s a 2-seat ride on two high frequency routes. If after that, transit ridership around Totem Lake continues to be weak, and/or Totem Lake fails to develop, then yes there would be no need to upgrade to rail.

Andre Tsang

PSRC growth regions are a joke. Ballard isn’t included but Renton and Bremerton are?