Photo: Jordan Stead/

Sound Transit is currently developing a consulting contract to oversee the process for selection of West Seattle and Ballard route alignments as part of Sound Transit 3 (ST3) light rail expansion.  They’ve concluded that by selecting a preferred alternative prior to the technical work of the Environmental Impact Statement (EIS), they are able to simplify the study work and thus reduce the total planning time by as much as a year and a half. This means that a preferred alternative will be selected for these lines by the end of 2018.

After discussion and a vote of our board, in order to further our founding goal of building a subway system that is rapid, reliable, frequent, convenient, and useful to all, Seattle Subway is officially taking the following key positions regarding a preferred alternative for the Ballard to West Seattle corridor:

  1. We are concerned that drawbridges, regardless of frequency of openings, pose significant operational challenges. Not only would drawbridges open and delay trains (trains which will run very frequently in the future), but drawbridges may not close. That failure would cause catastrophic delays throughout the system. Therefore, Seattle Subway will only support a high static bridge or a tunnel across Salmon Bay and the Duwamish Waterway.
  2. Expansions from Ballard — northward to Crown Hill and eastward to the University District — are included in Sound Transit’s Long Range Plan. Seattle Subway’s position is that any proposed design solutions must include the potential to expand north and east, such as a wye junction. We will only support designs that provide for in-station transfers at the Market Street station and seamless system expansion beyond Ballard that doesn’t compromise future transit service.
  3. Likewise, an expansion from West Seattle to Burien is included in Sound Transit’s Long Range Plan. Sound Transit must design light rail to avoid a dead end in West Seattle and allow for future expansion that doesn’t interrupt transit service.
  4. With optimum efficiency in mind, any new additions to Sound Transit’s ST3 network must be designed to accommodate 90-second headways. Stations and track alignments must be 100% grade-separated from traffic with no rail-level crossings for passengers.

If you agree with these overarching principles, you can weigh in now by emailing the Sound Transit Board, your elected representatives at the City of Seattle, and the Seattle Department of Transportation (SDOT). (Director, SDOT)

Andrew.GlassHastings@Seattle.Gov (Transit Division, SDOT)

Seattle Subway is organizing a community of grassroots transit supporters to channel public enthusiasm for fast, reliable high capacity transit into actionable goals. We champion a vision of a connected city and region to accelerate our region’s transit investments.

146 Replies to “For Link a Drawbridge is a Bridge to Nowhere”

  1. I thought a draw bridge was only one possible solution for the line to Ballard. Is ST really considering a draw bridge alternative to West Seattle? Seems like a joke if we can build a bridge like a freeway for cars, limit it to 40 mph, and then create a light rail line with a draw bridge decades later that makes cars more reliable.

    1. That does raise a good issue, though – why should West Seattle get a high bridge and Ballard a drawbridge, when Ballard has higher ridership?

  2. I love this so much. Now what will it take to elevate the stretch through SODO that has street crossings? Or reroute or close the crossings for street traffic? I know Lander is getting an overpass, but do you know if there are plans for Royal Brougham and Holgate? I’m not even going to try to list all of the crossings for the South Seattle stretch.

    1. The West Seattle line is projected to have new elevated tracks where the current busway runs, not to use the existing Central Link tracks.

    2. Another privilege for West Seattle.

      I guess we’d all better move to West Seattle where the trains are fastest.

      1. And where growth, while not quite on the level of Ballard, is happening rapidly; with a West Seattle Bridge that will not accommodate the long term increase of people and traffic on the peninsula. A right of way transportation option to get commuters over or under the traffic on the bridge and enable them to connect into the broader link system to commute to other areas of the Puget sound without driving.

        And yes draw bridges are bad period for link service.

    3. They’re overpassing the heavy rail tracks on Lander (the place where BNSF or Sounder trains will just stop in the road for several minutes), not the light rail line. I’m pretty sure.

      1. That’s right. There are no plans to overpass the busway. Instead , there will be a new parallel set or elevated tracks. I hope they make provision for a flying junction at the south end, so that Central Link can use them too.

    4. So are there any plans to grade separate that region? Street vacations?

      What about burying/elevating the South Seattle segment? Is that even possible?

      1. See immediately above. It is entirely possible to divert Central Link to the new elevated tracks if the Lander Station is “stacked” so that the junction immediately to the south can avoid level crossings. Neither Central Link nor West Seattle is going to be running any more often than every five minutes even at the rush hour. That would give a two and a half minute average headway on the shared segment.

        Now, there’s a problem a the north end: all that overhead concrete for highways makes stacking Stadium Station much harder to do. So in the end it may not be possible to share the right of way, because the Green Line is headed into the new tunnel while the Red Line from West Seattle is going into the DSTT.

      2. What might work is to have the separation immediately to the north of SoDo Station and have Central Link be elevated at least past Holgate. North of there running at-grade is less of a problem.

        Sharing platforms at SoDo would make transfers between West Seattle and Central Link easier. To go from West Seattle to SLU/LQA/Expedia or Central Link to UW and points north would be same-platform following train. But to reverse direction (e.g. West Seattle to Rainier Valley or the Airport before Burien gets built or vice-versa) would require climbing or descending a level.

        But that would be much better than the current plan to keep the cheap SoDo Station for Central Link and make everyone transferring between trains change level up to or down from the new elevated station.

        There is supposed to be a huge spaghetti bowl of trackage between Stadium and IDS to allow trains from either south end line to enter either tunnel. But making SoDo a shared platform stacked station handles all the crossing over without adverse movements provides a much better transfer experience.

  3. My recall is that a Ballard drawbridge would open 2-4 times a day, always off-peak when headways are lower anyway so the bridge openings could be coordinated.

    What do we know of the frequency of potential bridge openings on the Duwamish, and the costs of taking that path vs alternatives?

    1. It seems whenever I cross the Ballard Bridge around 6:30 PM weekdays it always opens for a sailboat.

  4. I preferred Ballard to UW because I had a feeling that ST would try to go cheap on the ship canal crossing which became inevitable when West Seattle was getting light rail.

    Anyone who has been on the 15X, 17X, and 18X at 6pm when it approaches the Ballard bridge (particularly when the weather is nice aka summertime) knows that the trip takes an extra 10 minutes for the bridge to open and close (usually for a sailboat) and for traffic to clear.

    A tunnel is unlikely (no matter what the Ballard Chamber wants/thinks) due to cost, but we can’t go cheap on the bridge.

    1. Fortunately, the train won’t have to wait for traffic to clear.

      How long does a normal bridge opening take? Last week at the Montlake Bridge, I wasn’t watching my watch but it felt like only three minutes at most.

      1. It’s been previously reported in the Times that the closure on a Ballard drawbridge would be four minutes. Potentially, that could put it within the off-peak headways.

      2. It may be possible for an opening to occur between headways, but we can’t assume the best case scenario, as that’s a recipe for disappointment. Also, if the Coast Guard today can dictate openings whenever they want, what leverage would ST have that SDOT doesn’t have?

        We already regret the cheap routing through RV that has permanently limited the system. This would be another penny-wise, pound-foolish move.

      3. A 70′ high bridge would not open as often as the existing bridges. The main problem is if the bridge is lower, like 35′. Some want a lower bridge in Ballard, so that it could double as a bike/pedestrian bridge and potentially a car bridge. Not a good idea.

        But the point about the bridge being stuck open is a more serious concern I hadn’t thought of. The West Seattle bridge got stuck open for years, and that’s why they built the high-level bridge. West Seattle has only two or three ways out of it so people had to go miles out of the way when the bridge was stuck open, and they threatened to secede from the city if they didn’t get a high-level bridge.

      4. Dan,
        While a tunnel would cost more, now that Link is elevated through Interbay there is no cost difference between a high bridge and a drawbridge, at least according to ST.

        The issue with the high bridge is that the city is scared of landowners concerned about their views.

      5. The problem with a draw span, is when a road bridge locks down, there are no rail or overhead connections to lock down as well. That could be 18 or more points of failure to add depending on what kind of bridge is built. And any one of those not locking properly could hold up service until someone can get out there and fix it.

      6. >> By “penny-wise”, you mean several hundred million dollars, right?

        and worse for the people who actually board in Rainier Valley. Folks don’t often talk about that. Yes, it would be great if the train was able to run more often, but when you board at a Rainier Valley station, it is nice that you don’t have to go deep down, under the ground (like at so many of our stations).

        There are instances where I think we were penny wise, but pound foolish (Mount Baker station jumps to mind) but serving Rainier Valley with a surface line doesn’t seem that bad.

      7. While a tunnel would cost more, now that Link is elevated through Interbay there is no cost difference between a high bridge and a drawbridge, at least according to ST.

        The issue with the high bridge is that the city is scared of landowners concerned about their views.

        Agreed. A high bridge (as high as Aurora) would be ideal, but was met with a lot of opposition when proposed by ST.

      8. I doubt the ballard neighborhood wants a towering bridge over the bay. The “ideal” solution is a bored tunnel, and the ridership projections justify it.

      9. Why would ridership be higher with a tunnel versus an elevated bridge? If anything it would be the opposite (better views, more riders).

      10. Ridership would be higher with a tunnel because it would be more reliable and trains could run at full speed. Trains may have to slow down on or at the edges of a bridge depending on how it’s designed.

      11. Seattleite

        The landowners can go pound sand. To be quite frank I would love to see a sleek train zipping over the river from my window.

      12. Are views really going to be an issue? NE side of Magnolia, NW side of Queen Anne, SW side of Phinney Ridge.. all of them have a view of Salmon Bay and the industrial area, which I don’t see being marred too terribly by a bridge. SE, SW, W, and NW Magnolia are the parts you’d have to worry about… that’s where the (greater amounts of) money is. NE Magnolia has a lot of multifamily and is where the greatest density is… I don’t see as many complaints coming from there as one might think.

      13. “…. The “ideal” solution is a bored tunnel, and the ridership projections justify it.”

        actually, given the relatively shallow controlling depth of the ship canal channel (about 25′), the short distance it has to go “under”, and the silty soils there, the ideal/easiest/simplest (etc) solution of going BENEATH the channel would be an “Immersed Tube” tunnel.


  5. A system that goes both north and east from Ballard changes the ideal location of the existing terminus, which is penciled in at 15th & Market. With no future extensions 15th & Market might be a reasonable spot as it can serve as a bus intercept for Rapid Ride D and 44. But with extensions north and south both those buses become marginalized and have other intercept locations besides, so spreading stations out for maximum pedestrian coverage is best.

    With an Eastern extension it must be realized that that the ST3 site will be the farthest west the system ever goes. But Ballard is already well to the west of 15th, with the heart of it both in terms of residences and businesses in the Ballard Ave /Leary area (23rd and 22nd, block-wise). You already have mid-rise apartment complexes that would be 10 blocks from the 15th/Market location, and 20 years from now that density is just going get further and further out.

    With a Northern extension, the likely next stop is at 65th & 15th, only 10 blocks north of 15th & Market. That’s crowding things, with a mere 5 block walk at the shortest point between stations.

    It’d be expensive, but I’d like to see a subway station at Leary (22nd) & Market, with the BHS extension north and 8th & Market east. That puts an enormous swath of the region within 10 blocks of one station or another, and the longest walk from within that triangle would only be about 7 blocks. That location is right at the entrance to the Ballard Sunday Market, only a block further from the hospital, and halves the walking distance to the Locks and the new Nordic Museum. The 44 and D buses would have to intercept a stop earlier or later, and the 40 would gain an intercept, though all 3 are largely replaced by these lines. One advantage of that location is that there’s already a small park there (with the tall mushroom sculptures) that could house a subway entrance, and it has excellent pedestrian access.

    1. It would be expensive, and would it bring the tunnel below the ship canal option into play? (remember, being cheap with a bridge is clearly in play here)

      I just can’t see the Ballard “Let’s fight the Burke Gilman trail” Chamber ok’ing construction on Leary Way and 22nd.

      1. I don’t really dig a tunnel (heh) under the bay. Look at how that turned out for UW, capped at under 55 MPH, ear discomfort for riders, super big ascent in UW station that takes 3 minutes to even get into the damn thing.

      2. I think the time is takes to get in/out of the station is a great point, and I hadn’t thought about it until Ross raised this point on another post. If an elevated station is only quick flight of stairs from the sidewalk, but an underground station is multiple elevators to access, that’s a compelling argument for elevated over tunnel.

        Sure, off-peak some people might be delayed due to a bridge opening, but on-peak many more people will have a quicker ride due to easier station access (and a quick transfer, which is important because the Ballard station will be the primary bus-Link hub for NW Seattle for potentially decades).

      3. U-Link isn’t capped at 55mph because of the tunnel, it’s the light rail technology and US regulations. Ear discomfort? Yeah, there’s a loud hum…but that’s the noise of being able to get from the UW to CHS in 3 minutes. I’m sorry, but no surface option could ever do that. And, it’s a severe exaggeration to say it takes 3 minutes to get from the surface to the platform at UWS. I do that descent everyday, and it’s 90 seconds from tap to train if you stand…and just shy of a minute if you walk down briskly.

        It’s a huge stretch to say a subway is less accessible than an above ground station, and in most cases it’s exactly the opposite if you actually care about the “rapid” part of rapid transit. Anyway, accessibility has a lot more to do with station placement than whether it’s above or below ground.

        So yeah, “look how that turned out for UW” – “it’s the most awesome thing ever”, said everyone alive.

      4. It’s capped because of the technical spec ST chose. There are faster light rails, although there may have been fewer choices in the 1990s. You just need trains spec’d for a faster speed, and wider curves and shallower inclines. ST is looking into faster speeds for the Angle Lake-Tacoma segment. I don’t know how feasible that is given the existing track.

      5. AJ absolutely. If your trip is 15 minutes, spending at least 1.5 minutes underground is already 10% of the time. Imagine if we could improve transit operations by 10%, we’d be spending millions on that when we could accomplish it simply via better station access.

        Felsen no what I mean is we aren’t even hitting 55 MPH which is the upper limit, and when I asked ST reps they said it’s because of passenger discomfort, and I don’t mean the sound, I mean literally the pressure changes from high speed ascension/descent. I do the trip daily too and what I mean it takes 3 minutes to transfer from bus to rail, making the time from out the bus door at UW station to stepping out of Capitol Hill Station roughly 7-9 minutes, depending on which exit you take). Don’t get me wrong, UW station is still amazing, but a lower depth and better bus transfers would have been even amazing-er. IMO if we had made UW station have elevator bays similar to Beacon Hill Station, that would have been wayyy better for the average commuter (though game days would still necessitate escalators)

        Mike Orr really? That’s amazing! I’d love to see faster trains. Like it’s super frustrating watching your train being overtaken by cars in traffic.

      1. I like 17th also. There won’t be much development west of 24th; the houses are too expensive to replace with multi-family. But the strip between 15th and Mary is a great place for big buildings. And 17th is close enough. Unless a 15th Avenue station is elevated with a mezzanine — essentially a subway station in the air — folks are going have to wait for lights like they have to do on MLK but with even more traffic.

    2. One thing to consider is think about what’s best for the network, bus routes can and should move so don’t consider them being any kind of current constraint. Instead of thinking about connects with routes 15 and 18, locate th station so it allows connections on market, allows connections on Leary, has a bus loop that access both major streets, etc.

    3. The 15th & Market Station should planned as a future crossing/transfer station, not a wye junction. Extend directly north to Crown Hill/Lake City, but an east-west line should be independent so it can travel further west to Old Ballard and the Ballard Locks.

      The current line could be elevated across Market, while in the future an east-west subway could be built under Market with escalator/elevator connections.

      1. Exactly. What Eric and Chad said is completely correct. There is no need for a Wye Junction. It makes way more sense to build a crossing/transfer junction, as that would enable a train to be extended to 24th. I think it is pretty easy to make the case that a station at 24th is actually *better* than any station east of Brooklyn, given the density there (as well as crossing bus lines).

      2. It depends on the long-term vision you agree with. Seattle Subway advocates for a wye because they see not just Ballard to UW but Ballard to Kirkland as a desirable goal. On the other hand, if east of Brooklyn is completely undesirable to you, then it makes sense to advocate for a crossing junction with UW to Ballard as more of a spur off the spine.

        To me, east of Brooklyn doesn’t make much sense right now, it might in the future. It would require much increased density around University Village, Children’s, and Magnuson to work, grown with upzones and bus transit, but those are solid institutional anchors to work around. ST4 might be a bit aggressive, but it seems possible by ST5. As such, it makes sense to me to push the Ballard line as far west as is reasonable, then split a Ballard-UW spur off of it in ST4, with the hopes of pushing eastward in ST5.

        Chad’s plan sounds great, the problem is that the underground route is expensive. If that doesn’t pencil out, could the trains run along Shilshole from whatever crossing to end up at 24th? Does it make sense to rebuild the Ballard Bridge to be multimodal, putting trains on the top deck?

      3. Ballard could start as a wye junction, and later be an east-west crossing? If it’s an east-west crossing with no way for the east-west trains to access the north-south line, ST needs to build an entirely new OMS facility somewhere in north Seattle.

        A wye junction in the U District is a non-starter – ST is very clear they need 100% of the current downtown-UW tunnel capacity to serve trains coming from Lynnwood. If you argue for a wye junction at the U District (the hallowed Ballard-UW spur), you are rejecting Lynnwood Link ridership forecasts.

        OTOH, the Downtown-Ballard line doesn’t have the entire Snohomish county funneling through, so it is much more conducive to having a wye and splitting frequency north and east (or west) of Ballard

      4. Yes and No. It’s a BIG elevation change and how do you do a service interchange? They should both be subways.

        The north-south line will be deep if it crosses the Ship Canal in a tunnel, so make the mezzanine to serve the east-west line as well.

    4. Bus routes going up 15th can’t deviate without getting into the turn-turn-turn overhead that we deride at TIB and on the 150 in Southcenter. So the question becomes what is it transferring to and how close to the end of the route is it? If a north-south route starts at Market Street or a few blocks south and transfers to an east-west train, then it doesn’t matter much if it does it at 15th or 17th or 22nd; we can make it work. But if a route is going a long north-south distance on 15th, then a deviation would be annoying overhead. Metro;s LRP envisions three north-south routes on 15th:

      * Rapid: Ballard Fred Meyer to 15th – 85th -Wallingford – Northgate stn – Northgate Way – Lake City.
      * Frequent: 85th & 32nd – 32nd – Mkt – 15th – 65th – Linden – 80th – Northgate stn.
      * Frequent: Interbay – 32nd W (Magnolia) – Emerson – 15th – Mkt – 8th NW – 145th stn.

      The first two end in Ballard and can meet the station wherever it is. The third is a minor route that already zigzags a lot, so one more zig won’t matter. It could come up Leary Way and turn right on Market if the turn isn’t too sharp, getting closer to “real Ballard”.

      As for the D, the 2020 plan straightens it out on Elliott Ave W, but in the 2040 plan it’s gone, so irrelevant re station location.

    5. A key technical question is how deep would a subway station in Ballard have to be. A few others have already noted this!

      Consider how deep the UW station is, a Ballard underground station could take lots of time to get to and from platforms. That makes a tunnel an inferior option for transit riders.

      Consider too how inflexible a tunnel is. If the other Ship Canal crossing at Montlake was a high bridge, we would have had money and potential to have added a station at 520, as an example.

      1. It’d twice as far from the waterfront to Market as it is to HSS. Track level could be st the second story down I am pretty sure.

    6. I like the ‘triangle’ idea. It would provide very good walkable coverage and it puts a stop in the heart of Ballard. I don’t think you can get much better.

  6. Just a thought about 100% grade separated –
    Charlotte’s under-construction light rail to UNC is in the median of a highway and trains will have the right-of-way 100% of the time as the intersections use crossing gates instead of a traffic signal system. It seems like the interbay area would be a sensible place for this type of solution, tracks at grade with crossing gates.

    Given the choice, I’d rather see extra money spent on a high fixed bridge that a viaduct in the interbay area.

    1. 15th Ave W has no at-grade intersections either. That’s why this alignment was chosen: it allows protected tracks either in the 15th footprint or in the railroad footprint just west of it, which is much cheaper than a tunnel and also happens to serve Expedia and the emerging village at Dravis.

    2. Crossing gates are not 100% grade separation as our own SODO transit corridor proves, there are frequent accidents when cars and people disregard those gates. The new LA Metro Expo line suffered a car incursion on its tracks on its first day of operation.

      Each time these “accidents” happen, it inconveniences thousands of people and delays operations in a cascading fashion.

      It is vital that we do not “settle” for anything less than true, 100% grade separation so that there is no possibility of a single point of failure.

  7. This seems uncompromising, to say the least.
    I’d like to see these recommendations as preferred result, but not at the cost of other stations in ST3, or at more schedule delay.

  8. It’s complicated design issues like this that make a single, unified environmental statement a rather bad idea. This issue alone will require extra environmental study and Coast Guard approval before it can be allowed. I can only imagine how many hidden issues could arise that could change how this moves forward. It has the potential to delay every extension with single documentation.

    Having said that, I would think that a high bridge is the least impactful to the environment. Even a new drawbridge would seem to require more disruption to the ecosystem as the counterbalance mechanics would need to be supported by bigger piers in the water. A tunnel would seem to be lots more expensive and even more environmentally disruptive.

    While I’m saying this about the Ship Canal, the circumstances are similar for West Seattle. In fact, the West Seattle crossing comes earlier in the program so resolving this now is more urgent. Of course, areas west of that crossing are so much steeper in elevation, a low bridge and especially a tunnel would constrain the track design pretty severely.

    1. ST doesn’t have to build the Preferred Alternative; it can built any of the alternatives disclosed in the EIS. The Preferred Alternative is just the zero point in the EIS to compare the other alternatives against, and presumed frontrunner.If it comes out significantly more expensive than assumed, that may give ST motivation to reconsider.

      1. That’s not the issue, Mike. The issue I’m talking about is that when ST finally is doing the permitting, dealing with the bridge crossings alone can add to the delay. Sure, ST can adopt a preferred option quickly, but when they get into permitting and design, the task becomes much bigger — not just because of the geographic scope, but also because of the number of additional agencies, permits and issues that have to be addressed. The other issue is that as details and debates emerge, delays and design changes will inevitably happen.

        Consider how it could be an issue with West Seattle, as the same environmental document is proposed to clear both West Seattle and Tacoma segments (and they are set to open in the same year). If there are issues with either one, the other one could end up getting delayed waiting for permits and that would push back construction and opening dates if the two segments are tied together waiting for permitting.

        That’s not even getting into the issues of having assignments with “surges” in them — from public meetings to meeting for permits to getting the right preliminary design staff working on them. Having different documents at different times is ultimately the best way to create the system. Is it really wise to have all the public meetings for design details of a preferred alternative for Everett, Tacoma, West Seattle, Issaquah and Kirkland in the same few weeks — and perhaps the 522 and 405 and possibly even infill station discussions scheduled at the same time?

        I am amazed when people naively think that a single-environmental document with a preferred alternative now is going to save lots of time. It’s like saying that the track meet can start sooner if we design a system where everyone shows up at once. If every seat has to be filled in a stadium before anything in the track meet starts (a comparison to the permitting process), it’s going to start later than if parts of the track meet can start once a section of stadium related to that event has everyone seated. The bigger the scope, the more that permitting can add delay, and it doesn’t take much review of history to point out that permitting delays things more than documentation and research do (unless a specific mitigation issue arises in the preparation).

        I view this time-saving promise as mostly political cover for such a long program, and administratively the various segments will end up with technical delays or political debates which will delay the program — easily more than if the documentation was spaced out into smaller approvals and honed preferred alternatives. At the end of the day, I think that ST will probably end up with so many supplemental environmental studies that by 2028 or 2030 we’ll wish that we hadn’t tried this blanket approach.

        Seattle Subway, the STB and others need to be advocating for that really critical piece of analysis — the ridership loads in various operational plan alternatives. We just don’t know how many people are going to transfer between train lines at Link stations. Heck, we don’t know how many people are going to transfer between Link train lines and the feeder buses! All these things inform a preferred alternative — the size and position of stations and their platforms, the track configurations and even the size of storage and maintenance facilities. It’s bad enough that we were not discussing these details before ST3 was proposed, but we really need it now if we’re supposed to get to a preferred alternative soon! And to bring it back full circle, we don’t even know how many people riders or how much train operations will be potentially affected by drawbridges! We need the operations plan and detailed rider data made public to have any substantive discussion about drawbridges!

  9. LINK drawbridge? Perfect example of what happens when a place changes it’s town-owning company from a first rate machine-manufacturer to a cut-rate grocer and bookseller. Even worse for transit than an aircraft manufacturer trying to design streetcars. Though regarding same reasons.

    Granted with many variables, passengers aboard a southbound bullet train regenerating into Seattle through the bedrock under the Ship Canal will have no idea why they won’t be able to get upstairs and retrieve their luggage five minutes later.

    Compare paragraphs on BART, MUNI, and the Oakland Bay Bridge. And whatever year anymore digging starts, tunneling machines will be farther advanced than they are now. Many thanks for the list of links, Seattle Subway. Also for the wonderful tone of understatement.

    But Mike, I since my average transit trip takes me to or through half a dozen neighborhoods and subareas, I’d consider the best possible transit through West Seattle to be the exact same privilege as for everyplace else in the Central Puget Sound Region.

    Of which, since the first big white shield broke ground northbound under Jackson Street, I’ve considered Olympia, Bellevue, West Seattle, Bremerton and Everett part of. Hate what’s happened to Ballard lately, but still not as bad as if it went back to living between its own city lines.

    But, hopefully on my way to make today’s condition of Ballard temporary, I won’t appreciate being aboard the tenth of a line of trains waiting for tracks on a drawbridge or realign. Would also rather be moving underground than drowning underwater. Between Olympia and Everett, and the Sound and Snoqualmie Pass… NIMBY in spades.

    Mark Dublin

  10. Seattle needs to just e-mail Translink and ask for the Skybridge blueprints. Then build.

    Oh and rant time: You people in transportation need to CEA-SE AND DE-SIST RE-IN-VEN-TING THE SIM-PLE STUFF AND GO BUILD. I do not want umpteen years of process, I want you to find what works around the world and get us our championship light rail NOW, NOW, NOW

    You wonder why I equate MULTIPLE Sound Transit employees to Seattle Seahawks? Its because I want passion and commitment to excellence out of Sound Transit, a desire to win NOW for the 12s! /rant

    1. Joe, when you finally get elected to the Sound Transit Board, I think you’ll find that like any other tool, “process” itself is neither a menace or a salvation.

      Government or private, or even within the same mind, nothing ever survives, let alone gets done, without some level of process.

      Seattle’s problems with our transit-related processes stem from fact that main thing process demands in order to work is that everybody has finally agreed it’s time to make a decision.

      When that finally happens, you’ll find out that not only does every process start to show effective and orderly results, but the project will have more or less painlessly collected a leader.


    2. Which part of Sound Transit represents the Seahawks Offensive line? The guys who designed the UW line without allowing a Ballard to UW line to interline? The ones who are trying to sell Ballard on a low drawbridge.

      1. North Sounder doesn’t go where the people are! The population is centered around Lynnwood, not around the Mukilteo and Edmonds shore. Just cancel Sounder North and put the money into replacement buses and accelerating Everett Link.

      2. “Grade-separated BRT”, Joe? While it might be possible at Mulkilteo to take a pair of lanes and put barriers between them and traffic, there’s no way that’s going to happen along 196th.

        But it would be nice to have real priority for the buses.

  11. This should be a tunnel if for no other reason than that the east west line to the U District will certainly be tunneled. A transfer between an elevated Ballard Link and a Ballard-UW line would require a fifty foot elevation change. Cities need subways, not elevated structures.

    1. At the same time, underground stations are very expensive, and especially adding a transfer to a perpendicular line to an existing underground station. We’re already facing this at U-District Station, where ST has refused to say how the Red/Blue lines will transfer to a 45th line or what stubs or interfaces it installed to facilitate this (apparently none).

    2. >> Cities need subways, not elevated structures.

      New York, London, Vancouver, D. C. and Chicago all have elevated “subways”. My guess is most subway systems are a mix, typically underground in the heart of the city, and above ground outside of it. New York City even has above ground in Manhattan. It’s not that big of a deal.

      As far as the transfer goes, A Ballard to UW subway (if it is ever built) might not be underground once it reaches 15th. It will certainly be underground through Phinney Ridge, but it could easily pop up around 3rd NW (there is even a little park there — It might also be cut and cover through Market, which means that the difference in distance would not be that big.

      The essential thing is that ST should consider how this will eventually interface with an east-west subway. They need to address the very issue you mentioned, and we should be able to figure out whether it is worth it. What concerns me about underground, is that it is very expensive not only just to get to Market, but to go farther. I could deal with an awkward transfer if it means we can have stations at 65th and 85th, and still have money left over for a down payment on things like the Ballard to UW line, or the Metro 8 subway. But the point is, the issue should be addressed. We should figure out what the various options are (elevators, escalators, etc.), what they would cost, and what they would mean for typical users. Worse case scenario, we do something similar to what Mike fears — we build it underground (at great expense) only to ignore a crossing, and then end up with something just as awkward as an elevated to underground transfer.

      1. There’s no reason that a subway can’t surface in the middle of 15th NW somewhere around 60th. What’s hard about that? North of there 15th is a less critical roadway so unless the eventual destination is Lake City, going on the surface from there north isn’t horrible. Maybe there is a stub track between the tubes north of the station for turnback Green Line trains while some other line (Renton?) continues on.

        I like the idea of lines weaving back and forth across each other. It produces great transfer opportunities, especially if one is more an express (a la North Link) and one a more local. For instance, folks from Kenmore could ride to Northgate on the “Orange” line and if headed downtown could change there for a four stop ride to Westlake rather than a nine stop version (Greenwood and Holman, 85th, 70th, Market, Dravus, Smith Cove, LQA, Gates, Amazon). But folks coming from points north headed say to Gates could change to ride around through Ballard. They’d get two more stops and the same number of transfers (one), but the trains are likely to be less crowded.

        It that’s the plan, then the extension north on 15th should be grade separated.

        But there’s another option. Turn left at Fremont North and go up the Interurban ROW to the middle of Shoreline. There are significant parts which could be at-grade with no or only slight auto conflicts because of the cemetery and the big box stores north of 130th. In that scenario running at-grade in the middle of 15th and somehow at-grade for a way alongside Holman (north of 85th the traffic picks up noticeably) would be fine.

    3. I don’t think Ballard-UW needs to be a tunnel. Given the steep slopes of Phinney Ridge (the ridge, not the neighborhood), a tunnel portal east of 8th NW (basically what Ross suggests) is very reasonable. And east of U District, the steep slope just west of 25th Ave means you could do the same thing – tunnel under the U-District, and then pop out & run elevated along 45th Ave to Childrens.

      And Ross is spot on – many cities have elevated lines. Chicago’s system is literally called the “L” for that reason. And isn’t most of Vancouver’s system elevated?

      1. Vancouver is not elevated downtown, or most of the way to the airport. And, I sure hope Chicago isn’t being used as some sort of aspirational example. The L is hideous, in terms of its structural footprint. There are places where being elevated is appropriate, but if money was no object, a subway is better. Surface and elevated can mimic the pitfalls of BRT, where compromises slowly chip away at any preconceived advantages (High, high bridge to higher drawbridge to not-so-high drawbridge).

      2. You expect to put an elevated railway down the middle of Market Street? You must not like Ballard.

      3. P.S. I do exactly like the idea of erupting from the hill just below the dorms about “44th”. In fact, you could put an elevated station for East Campus and U-Village just northwest of the Pend O’Reille intersection with Montlake and it would be a winner!.

  12. ST can save some money between Whole Foods and the Ship Canal by running at-grade next to the BNSF yard. That puts the Dravus Station midway between West Queen Anne and East Magnolia, which has more development opportunities.

    Also, instead of elevated over Elliott, the track between the tunnel portal and Smith Cove Station should run in two notches cut into the hillside. Put the station at the north end of the ped bridge and there is a good route between the Magnolia Bridge on-ramp and the hillside. No need fir s sixty-foot high overpass of the bridge approach.

    The savings from doing these two things should pay for a tunneled crossing of the Ship Canal.

    1. I agree, they really are missing the boat when it comes to running this elevated. It is unnecessary for 15th West. This isn’t MLK Way, but I’m afraid folks think it is, and fear the worst. Unlike MLK Way, there are only a handful of level crossings, and all of them are minor.

      That being said, I’m not sure if you could save enough money to pay for a tunnel to Market, let alone 65th or 85th. Plus you would upset this very group. Seattle Subway was insistent that this be built elevated, even though it is pretty easy to argue that it isn’t worth the money.

      1. Well maybe not pay for the Ballard Station; you’re probably right about that. But I submit that the tunneling will be no more expensive than the bridge and its big, unsightly approaches.

        Importantly, there’s no need to dig a subway all the way to 85th. Between Market and 85th 15th NW is less important as an arterial than it is to the south of Market or the north(east) as Holman north of 85th. ST is talking about running in the middle of 15th which means at least one lane has to go and in most places two. So….surface at the jog at 65th and have the station at 70th rather than 65th. There’s a “wide spot in the road” in the block or so north of 65th which could be taken for the surfacing portal. Think of the Duboce Portal in San Francisco. It is very little wider than the two tracks which use it.

        Whether the tracks continue on up to become elevated or simply transition to at-grade (see my post above), two lanes would have to be taken to 70th or so. But again, traffic reaches a local minimum between 65th and 70th as folks coming from or heading to the south turn on or off and the same thing happens for northbound cars. I don’t think elevated along 15th NW north of there is a show stopper; there aren’t important views to protect. But please remember that elevated will certainly take one lane and make even a “turn refuge” problematic. In any case in “downtown Ballard” both lines must be subways, if for no other reason than the needed service connection.

    2. Has ST committed to elevated at Interbay? Is this just a terminology issue? We use elevated to mean “no grade crossings” but it can be technically surface, like Lynnwood Link in Shoreline and Federal Way Link along I-5 with underpasses at the cross streets. I thought that’s what ST intended, not putting it in the air unnecessarily.

      1. Exactly, Mike. This is the point of what I wrote. If you notch the hillside behind the smattering of small semi-derelict businesses along the north side of Elliott an run the trains there you can have a station at Smith Cove with shore-side platforms and no pedestrian crossing. The platform of the northbound track is above the track and part of the platform for the southbound track. Then you’re set up at the proper elevation to transition into the one stretch of elevated structure that would be necessary, a quarter of a mile from about the Brown Bear car wash, across 15th NW to the Whole Foods parking lot and around behind the store to Armory Way.

        Extend the Helix Bridge across Elliott and you have a VERY nice stop for Expedia and no impact to the traffic capacity of Elliott for significantly less money that putting Ballard Link on stilts for a mile and a half from the tunnel portal to a high bridge.

        And you have a better location for the Dravus Station, too. East Magnolia doesn’t have views to protect as does West Queen Anne, so you’re going to find it easier to put density there.

  13. Operationally, a tunnel would work best and advocating for that is a fine thing to do. But I’m guessing that once the analysis is done, building a bridge will be sufficiently cheaper and lower-risk versus a tunnel at this location to muddy the waters. I suspect/fear subsurface soil conditions (which we’ll need to evaluate) will prove unfavorable to a tunnel, especially in Interbay. A drawbridge would occasionally open, which is a problem, but a high-level drawbridge would open rarely. Views from the trains on such a high bridge would be glorious every day, to the extent that’s any consolation.

    ST could potentially leave in place an empty southbound train at a tail track just north of an Dravus/Interbay stop in order to maintain southbound headways if the bridge does have to go up. Are such things done anywhere else?

    The thorniest issues I see are not actually the ship canal crossing, but station siting and the potential for expansion. For station siting, is consternation over an obvious location around 15th/Market where RapidRide lines will intersect for the many years we are building this line. That location is not in the center of historic Ballard, but it’s only a short walk away, and it may be too tight in the historic core to fit in anything but an expensive, mined station whose construction impacts would still be painful. Bus transfers must also be considered. Station siting will get harder the more big, expensive projects are built on large parcels where the station is desired, so it’s good we’re planning now (given that years ago is not currently an option.)

    As for planning for expansion, that is something ST has IMHO done a rotten job of in years past, and that needs to change in this round. Expansion to the east (to UW) has long been part of the long-range plan. The only way it wouldn’t make sense to connect those two lines is if the east-west Ballard-UW line were to continue west — but there’s only so far west it can go (one or two more stops?) and I’m not sure how much that would buy in terms of ridership. A wye that splits to the north and east would create a lot of very useful future trip pairs; figuring out how to design and construct that will require ingenuity. You could argue that expansion to the north, or east, seems more important today, but building for an uncertain future and we must not design ourselves out of the flexibility we’ll need to make the best decisions later.

    1. The only way it wouldn’t make sense to connect those two lines is if the east-west Ballard-UW line were to continue west — but there’s only so far west it can go (one or two more stops?) and I’m not sure how much that would buy in terms of ridership.

      I think it would be very good. Even adding one station would be great, really. A station at 24th would probably serve more people than all of the stations east of there (with the exception of Brooklyn, of course). There is not only a lot more density close to the station, but it connects well with the north-south streets (there is lots of density along 24th).

      A wye that splits to the north and east would create a lot of very useful future trip pairs;

      Would it? I mean I’m sure there would be people that would go that way, but I don’t see it as being huge. You essentially have a circle, with three connection points. The trips that are close to Ballard (but don’t start or end at Ballard) are probably the lowest ridership trips on that circle. You basically have any combination of Expedia, Interbay, West Woodland, and Phinney. Once you get close to Westlake (Lower Queen Anne, South Lake Union or Capitol Hill) you are better off going through Westlake. With longer trips, it is the same story. If you at Northgate and headed to Lower Queen Anne, it makes sense to transfer at Westlake, instead of the UW, even if they build the Wye.

      I’m not saying there isn’t value in a Wye, but my guess is that the time saved per rider by enabling a Wye is far less than if you added a station at 24th.

    2. “….. I suspect/fear subsurface soil conditions (which we’ll need to evaluate) will prove unfavorable to a tunnel….”

      while that is very true about a bored-tunnel, that isn’t the smart way to go beneath the ship channel anyway.
      The better (cheaper, less change in elevation) way is the “immersed tube” method of tunneling. soft soil conditions actually are a positive factor for dredging the necessary “trench” across the channel for this type of construction. Think of it as “underwater cut & cover”….

    3. Jonathan,

      Technically, what you’re advocating is a “junction”, not a “wye”. Traditional railroad wyes have all three sides of a triangle filled. You want a simple routes divide junction, unless you’re considering having some service from UW-Ballard which turns north along 15th NW.

    4. Oh, and the way you make such a “routes divide” junction is by stacking the station on the north-south heading and placing it about 52nd. That’s a little far from the center of activity in Ballard, but maybe the center of activity would just move around it.

    1. Tillikum Crossing in Portland, opened in 2015, is such a bridge — for bikes as well — — and it’s beautiful. It actually has a 14 foot bike path on not just one but both sides of the bridge. Bus compatibility does slow things down operationally, and it’s not clear to me what buses would need to use this crossing once there’s a Ballard Link line, so long-term bus compatibility might not pencil out.

      Meanwhile, the failure in the city’s bicycle network represented by the current Ballard Bridge can’t really wait for any new Link bridge to be built here. I don’t know how to solve that one without a new bridge. Building two new bridges here seems excessive. To the extent it makes sense to build a bridge here for Link (versus a tunnel), maybe it would make sense to build a bridge as the first element of the project such that it could be used for RapidRide in the interim and bikes/peds too (with an elevator and approach ramp on each end.) Though it would have to touch down at grade to be useful for RapidRide in the medium term, which sounds like a big constraint…

      1. The multi-modal thinking that you discuss is badly needed. It often seems that SDOT, Metro and ST are living in singular, tangential worlds. That’s one more reason why I don’t like the fast-tracked ST-only process.

        I could see value in fast-tracking a crossing before Link opens in the 2030’s. If SDOT pays for it early, ST could repay for it later in the ST3 program.

        I think a two-bridge strategy would be best. If a higher new bridge for buses and peds/bikes first that opens in about 2026, the current drawbridge can be closed and replaced within the existing alignment while Metro could retain a RapidRide line and high-frequency bus shuttles could also run during construction. When the rail finally opens, the buses can go away except in emergencies and ped/bikes can have two ways to cross the Ship Canal.

      2. It’s easy enough to rough in the Link alignment and have the buses use ramps to the side. There will still be buses going that way after Link comes. The successor to the south end of the 40 comes to mind. It might turn in SLU and continue on Boren to the hospitals. And if whatever facility is built does accommodate buses, then a UW to East Magnolia and Smith Cove route via Fremont and Leary Way would make sense.

        The truth is that if it’s a tunnel, the buses can use it too. It’s not that far that ventilation would be a problem. I guess the buses would then have to have half-turn corkscrews up to the surface and would have to wait for trains. With Market so close to Dravus, the trains wouldn’t be meaningfully slowed for bus stopping distances.

        A big “X” centered on the Ship Canal crossing.

      3. When Johnathan said slow he meant slow. I was in Portland last spring and saw the trains on the bridge, and I rode a bus from the west side station to downtown. We want trains running full speed so that there’s not so much overhead getting to Ballard — that’s the point of Link in the first place.

        There was one thing on that bus route I liked: between the bridge and downtown there are exclusive-ROW segments where the bus just zips along like a MUNI Church Street train going around the hill above Dolores Park.

      4. That section where the buses zip along is also mixed MAX and bus elevated right of way.

        Buses go faster over the bridge than MAX does. I think they do that because of the track level pedestrian crossings at each end of the bridge, combined with the blind curve at the east end.

        Those particular issues will not be built into any sensible Link structure.

    2. For future streetcar lines caliber of South Lake Union, Connector, and First Hill, no harm. But with the best bus designs, lightning-fast wheelchair securements and trained assistants, doubt buses would not always be in the way of a six minute rail headway. Or less.

      But for any dual-mode line, if joint use years haven’t taught us anything else, we’ve long been looking at minute to minute evidence is that these systems demand communication and control far beyond anything has ever had. With painfully visible results every rush hour minute.


      1. No, Mark. The buses wouldn’t serve the same stations. They’d stop on the street at both ends, before and after they pass through the tunnel. The trains would have absolute priority, so you don’t need to worry about wheelchair tie-downs delaying the trains.

  14. Politically speaking can you tunnel to Ballard under the ship canal, and not tunnel into West Seattle?

    if the answer is no, then ST will need to find money for both projects, and if that can’t be done then both routes end up without a tunnel.

    My money is on ST accepting compromises to the route, which means no tunnels.

    1. Or, Seattle passes a levy to pay for tunnels. There is a precedent for this with Bellevue – if the city wants a tunnel, but the value engineers at ST says it isn’t worth it, the city can always chip in for the incremental cost.

      1. Good God, if we pass a levy, I want to pass a levy for more important things. Like a new rail line, or just fully funding the RapidRide+ lines.

    2. Great point. I forgot to raise this (in my long rambling rebuttal below).

      There is also the flip side of this.The more urban, elevated rail we accept, the more we will get. Since elevated rail is cheaper, the more overall grade separated rail we will get.

    3. Why do you need to tunnel into West Seattle? Just build a new Link bridge next to the West Seattle Bridge. Yes, it has to be longer because the grades must be gentler, but you have to overpass both SR99 and the West Seattle Freeway, so the thing is going to have to be 60 feet high soon after leaving Lander Street.

      The west end can use Pigeon Point to descend to station elevation at Andover. Then slant up across the valley and poke into the hill just below Avalon. Easy-peasy.

  15. Does a high bridge vs. a drawbridge make any difference in terms of bike & ped lanes than can be included in the bridge?

    I remember the high bridge requiring a prohibitively long approach ramp for bike lanes, so if the drawbridge also comes with an improved bike/ped crossing of the ship canal, that might be something in favor of a drawbridge, vs. a high bridge or tunnel.

    1. That’s the concern about a high bridge, that it would be infeasible for bikes and peds. Compare the Fremont and Aurora bridges for trips from Fremont or SPU.

      There is a separate movement for another bridge around 3rd Ave NW. In one concept it would be an automobile bridge and the Fremont bridge would become bus/bike/ped only. And for bicyclists it seems close enough to Ballard. Peds can use the existing Ballard Bridge.

    2. This assumes that the light rail line and the pedestrian / bike trail are parallel to eachother.

      If you build the pedestrian / bike path as a lower level of a high bridge, then that doesn’t have to be the case.

      Something like this, only with the pedestrian section not as an afterthought and with a draw section:

      1. If you sling a pedestrian crossing underneath you lose the benefit of having a taller drawbridge, unless you pick a design that allows the pedestrian portion to open independently of the main span. Portland’s Steel Bridge can do something similar to this, but it seems like it may be unnecessarily complicating things.

  16. Just throwing out a ridiculous idea, that might be more cost effective when the length/ heights of the approaches are considered.

    What about two double track bridges just above water level and separated by enough distance that one bridge would be open at any given time for continuous train ops, on a set schedule, while boats would be forced to platoon between them. Like the pedestrian paths at the locks

    1. That’s an intriguing idea! The big issue that I see with it is that building a one higher bridge is much cheaper than two low drawbridges. It would also take more land. It would orotect views though.

  17. I’m afraid this is the type of thinking that has gotten us into our current mess. There is no cost benefit analysis. Improvements that are very minor but expensive are considered essential, while far more valuable additions lack funding. I’m still waiting for the First Hill station, and it isn’t even mentioned in your piece. Instead we have expensive, but relatively frivolous suggestions.

    I’ll address the ideas by number:

    1. Movable bridges can fail, but so too can other aspects of the system. The more miles of track you have, the more likely you will have a failure. If you are worried about the reliability of our subway system, maybe you should support shorter lines.

    I find it hard to imagine a train ever waiting for a boat, let alone wait a significant amount of time on a routine basis. There are a few things to consider:

    This bridge will be taller than the Ballard bridge. A taller bridge means fewer openings.

    The bridge will never open during rush hour. That is the case now, so it is silly to think that it would change for the worse. It is far more likely that the closure period will be extended (for both bridges).

    The operator decides when a boat goes through. The boat may have to wait minutes — it is up to the operator.

    If this train is paired with the Rainier Valley line (which is the plan) then at most, you have have 6 minute headways. Outside of rush hour, Link runs every 10 minutes. This will likely improve, but it is reasonable to assume that at best, Ballard Link will have 6 minute headways outside of rush hour. More likely it will be 8, but I’ll assume 6.

    A bridge operator will consider both the northbound and southbound trains. If they are synchronized to cross right at the bridge, then their will be 6 minute gaps. If they are completely out of sync, then there will be a 3 minute gap. Realistically, it will be something in between. Again, I will err towards the worse, and assume 4 minutes gaps.

    That means that an operator simply has to wait a couple minutes until there is a 4 minute gap, and open the bridge. It is very rare, if not unheard of, for an opening to last 4 minutes. This is not like the current bridges, where the timing is measured when the traffic lights turn yellow. This is the time where the bridge actually starts going up, to when it is completely down (and ready for the train). As soon as the train passes over the bridge, the operator opens the bridge.

    If it turns out that the operator does keep the bridge open too long, all it means is that the train simply waits — for a very brief period — at the nearby station. Most riders wouldn’t even notice, and those that do will assume the slightly longer dwell time is due to a passenger who failed to clear the entrance (which will be a much more common occurrence). We are talking about a very rare (or quite likely nonexistent) problem.

    While a high bridge would be ideal, a tunnel would not. To begin with, an underground line would be less popular. While I know that there are more important things, we shouldn’t dismiss the value of providing more entertaining transit, especially if it is actually cheaper. A high, beautiful crossing of the ship canal would likely lead to more ridership, which would increase fare recovery, and make further investments more popular.

    But it is the extra cost that is the big issue. A tunnel would be very expensive, not only to serve Market, but to serve areas north of there. If a train gets to Market above ground, then it is highly likely that it will continue above ground, which makes adding stations at 65th and 85th relatively cheap. Put it this way:

    Would you rather have one underground station or three above ground stations in Ballard?

    To me the answer is obvious. This should be above ground, and the crossing should be via either a high or movable bridge.

    2. Transfers are essential, a Wye junction is not. If a Ballard to UW line is extended beyond 15th and Brooklyn, then the line will be extended west, to 24th NW. There is more density there than anywhere east of 15th (other than Brooklyn, of course). Even if it just goes between 15th and Brooklyn, it is hard to see how a Wye would be essential. It would be nice, sure, but what is essential is a good transfer. On that we agree — ST needs to have a plan for making the transfer from a north-south line to an east-west line. But such a plan should include the likely (and ideal) extension of an east-west line to 24th.

    3. Burien lacks density. Heck, it just lacks people. As it turns out, the highest concentration of people sit extremely close to the intersection of two freeways. It is remarkably easy to serve those people with express bus service — you don’t need another long distance, money losing, hard to maintain train. Not only am I afraid that the good people of Burien will be suckered into paying for another suburban project they don’t need (“Welcome to the club”, says Kirkland) but I don’t want Seattle to pay extra for that. Seattle has way more important needs. Seattle should be spending their money on a Metro 8 subway, or a Ballard to UW light rail line.

    4. Improving headways sounds great, but this is by no means the most important line. I would rather put money into the section between the UW and the south end of the existing downtown tunnel. It is simply a more important line. There are more stations downtown, and it serves Capitol Hill and the UW. If there is any section that is likely to have capacity problems (due to weak headways) or want better headways to improve transfers, that is it.

    1. Ross, I’ll point out that building the line to Alaska Jct for a future extension basically costs nothing. Just some tail tracks with the right kind of switches. Like what we have an Angle Lake. There’s no reason not to do it, even if there never is a line to Burien.

      1. Fair enough. If so, I’m all for it (why not). What I am not interested in building an extensive amount of infrastructure so that we can build something that will likely never need to be built. For example, my understanding is that the tracks will come in from Alaska and end at California. This would be where the tracks would turn, and head south (on California). It may be trivial to accommodate the turn — it might not. I would hate to spend a lot of extra money making a turn, only to see it never be used, or worse yet, be justified for an extension that isn’t worth it.

      2. No. The tracks won’t be headed west on Alaska, they’ll be heading south under California. There is not going to be surface rail in West Seattle, bank on it. So there will be a bored tunnel from the hill underneath Avalon curving curving gently until it achieves a north-south heading maybe a block north of Alaska.

        West Seattle will not allow anything else and they have political power.

    2. I’m curious, Ross. When you think about transit planning and building, what time-frame are you working in? Considering what’s happened to every civic system starting with transit in the last three years, what makes you think Burien, or anyplace else, will always be too small for a subway?


      1. Fifty years is a reasonable time frame, but I would go out a hundred. I don’t know of any place in North America that is spending the kind of money per person we are spending. At some point, I figure it has to slow down. I can think of several projects that are a better value than an extension of this line, and will likely always be a better value. I expect Burien to grow, but I don’t expect Burien to grow at a faster rate than Seattle (it isn’t now, and I see no reason why it would in the future). Since success in transit is largely based on proximity and density, I don’t see why a Burien extension would leapfrog those other projects. In short, either we build one of the largest, most expensive subway lines in North America (for a city that isn’t that big) or we build things like a Burien line before more important projects.

    3. If the line to Ballard is elevated, does that mean an extension of Ballard-UW to 24th would have to be elevated or on the surface in order for some sort of junction/connection to be built at 15th so that the Ballard-UW trains have access to a maintenance facility?

      Would it be cleaner and cheaper and gain some functionality to do as SS proposes and build a wye at 15th thereby forgoing a stop west of there

      Don’t get me wrong. I think 24th (or 22nd) makes a lot of sense due to the density of what’s there, and would love for ST to plan for it if possible.

      1. I think you could have a wye junction to avoiding building a new maintenance facility, but still have the actual line have stations west of 15th&Market. So revenue service would be an east-west line, requiring a transfer to downtown, but non-revenue trains would use the wye junction to access the OMF.

        I proposed elsewhere in the comment that Ballard-UW could start as a wye junction that interlines with Ballard-Downtown, and later be extended west.

      2. How trains in Ballard get reversed is also an important design consideration for track configurations. Tail tracks? Sidings? A wye or two sets of tail tracks in Ballard could be required if train frequencies are high enough no matter what happens with future extensions.

      3. THIS is exactly who elevated will not work in central Ballard. Such a service connection needs to be hidden away underground. To do it in the air or on the surface would require taking the north half of the block at the southeast corner of the 15th and Market intersection. And please remember that there is a twenty-two foot elevation change, unless you want to have a four diamond crossing! Though there is one at the northwest corner of the Loop on the El, it’s an operational nightmare.

        The right design for this station, whether at 15th or 17th is to have the lowest level be the north-south tracks with a center platform. The next level up would be the east-west tracks, again with a center platform with escalator connections directly between the platforms. Above that would be a mezzanine with a fairly low ceiling (think Powell Street) oriented east-west in order to provide a pedestrian passage across 15th NW (Yes, if the north-south tracks are at 17th the platform for the east-west would lie to the east with the mezzanine extending under 15th.

        The service track would be deep enough that it could under-run whatever is built in the southeast corner

        THIS is the right station design for Central Ballard. Build for the future, not some scrimping “budget”.

      4. P.S. Since Market Street is seven lanes wide it might be possible to run the tracks at the edges of the street envelope and put the Mezzanine between the east-west tracks above the center platform. That means that the ceiling of the east-west platform would be fairly low as well. The upper portion of the track box with the catenary could project above the floor of the mezzanine.

        I know that it is really nice to have soaring ceilings like in the DSTT, but they’re not essential.

      5. P.P.S The reason I mentioned taking the northern half of the block is that Link is not a streetcar. Yes, it can take some fairly tight curves, but not an ordinary intersection.

        I suppose that if the elevated tracks on 15th were at the west edge of the street envelope and those on Market at the northern edge of its envelope, such a track could make the curve and stay within the roadway ROW. But WHAT an ugly blight on the intersection.

      6. The connection between the two lines at the UW will of course be done underground. You could connect the tracks there so that trains could access the maintenance facility. It is not essential that they connect in Ballard.

        As far as a passenger connection between the two lines, you raised the same point above. It is a good point, and as I said above, the issue should be studied. How long would it take to transfer if you end up with an elevated east-west line and an underground north-south line (the worst possible scenario)? My guess is if you designed it right, it wouldn’t be that bad. More to the point, I don’t see it as a major transfer point, unlike the other ones.

        Either way, money matters. It is quite possible that they simply don’t have the budget to put in an underground stations. If so, then what? Would you delay the project until we can pass another levy? Would you get rid of some other Seattle project, and if so, what?

        If we do manage to scrape up extra money, would you rather have an underground station, or two extra stops in Ballard (at 65th and 85th)? These are the trade-offs, and to me it just isn’t worth it.

      7. An underground station, for sure, as long as it was designed to accommodate an east-west line as mentioned above. A shallow mezzanine along Market ties the neighborhood together across the 15th Avenue car sewer and even across Market itself. This works even if the north-south line as along 17th. The tubes for the east-west line are next below and below them are the north-south tubes. Yes, this has to be cut-and-cover because of the shallow depth, and that means utilities problems.

        I was in San Francisco when the Market Street subway was being built; it was amazing; they hung the utilities from a latticework of girders on which the temporary roadway was supported. They actually moved almost none of them, though if pipes were corroded they replaced them of course. They just excavated around them and accessed the area beneath from a few places where there were few utilities or they did move a stretch of them.

        Yes, it was disruptive, but there were plenty of people still walking along the sidewalks. Of course that Market Street is a lot wider than Ballard’s, but the block between 17th and 15th doesn’t have that much retail presence.

      8. Ross, as I said in a different reply, since ST made no allowance for such a junction in the North Link tunnels, it is very difficult to build and will certainly require making a mighty hole in the UW campus somewhere around the Law School. You can’t just poke through the shell of a bored tunnel to create a junction; you have to create a box around bores where the cross-overs and the junction itself will be, peel the bore cladding off the trackway and then install the trackwork.

        Fortunately the cross-over itself would probably be trailing point, a plus for safety, though the interchange itself would likely be facing point on both lines.

    4. The big silent elephant in the room is the forecasted overcrowding between UW and Downtown once Link opens to Lynnwood. Imagine being unable to board a train at Northgate or Capitol Hill to go south in the mornings.

      At some point, there will probably be a need to offload riders onto this line north of UW. Whether it’s at Northgate or U-District, an extension to connect the two lines is probably going to be needed.

      1. The DSTT can support 90-second trains with some capital improvements. ST chose not to make them in ST3 because of the second tunnel, but they’re still possible.

      2. Good point Al. The section between the UW and Capitol Hill is likely to be the heaviest one in our system. I am pretty sure bottlenecks like this are worse in the evening. For example, an event at the UW combined with commuters heading home. It is likely either way that things get better as you get past the U-District.

        So that does sound like an argument for allowing the train to turn and head towards the UW. Except that during heavy use (rush hour) making a transfer is no big deal. It is when trains are running infrequently (e. g. every ten minutes) that it is a pain. The key, really is that all the trains be able to run as frequently as possible.

  18. If the Ballard Link’s crossing of Salmon Bay ends up being a drawbridge or fixed elevated crossing, Sound Transit and Seattle would be missing an opportunity to improve the city’s and region’s disaster recovery resiliency by building that structure in a way that allows emergency vehicles, etc the ability to cross the Ship Canal when existing bridges (Ballard, Fremont, Aurora, University + Montlake) are knocked offline due to earthquake damage. If Portland built its transit-priority Tillikum Crossing with seismic resiliency in mind, Seattle should be building it’s next Ship Canal crossing with megaquake recovery in mind.

    Transit advocates have made a strong case for ST3 from a mobility standpoint but haven’t promoted Link expansion as an investment in regional earthquake resiliency. Resiliency and mobility should be advocated for simultaneously with Ballard Link.

    1. Michael, I agree with you that LINK should be incorporated into the whole emergency response plan. But I’m skeptical that a transit bridge will be any better able to withstand a ‘quake than a regular one. It might be better to structure and equip an underground tube, for evacuation and rescue and rebuilding response.

      Emergency equipment can be stored at stations and other places in the tubes. And the cars themselves could be build with removable seats so as to carry machinery and supplies.

      Also, watching Beacon Hill Station constructed with vertical cables in a hole the size of a skyscraper, I wonder if serious beam lengths an supply weights could be likewise carried if stations were constructed with large surface entries built into the structure.

      In the Nordic countries and Finland, every subway station has been a bomb shelter ever since it was built. Doubt we’ll ever be hit by manned bombers- B-52 kind, not Uni-kind. Has never happened to us before. But I think dealing with the damage of a national disaster is a lot more legitimate a national defense measure than most new weapons systems.


  19. Another thought on a wye: As I understand it one of the benefits of a crosstown Ballard-UW line is to serve as bus catchment from a large part of NW/N Seattle.

    The problem I can foresee is that if bus’s are dropping people off at this line’s stations and they are forced to transfer again to get downtown at either U-District, or 15th and Market, then what is likely a one seat ride today becomes a three seat ride.

    I am coming to the conclusion then that making one good transfer to a train that goes downtown (through the wye) is going to get more voter support than a line that forces a transfer to another train. If, as AJ suggests, an extension to 24th could remain possible, then this probably ought to be part of the calculus here.

    1. A 45th line would replace bus 44, maybe. It wouldn’t replace north-south buses. Any north-south buses that are going away or are being truncated at the Ship Canal will be in the North Link and Ballard Link reorgs. Metro’s long-range plan shows its latest thinking.

    2. @John — If people are going to go downtown via the Ballard to UW line, it is quite likely they will go through the UW, not Ballard. So if we do build a Wye, that is where to build it. The reason is that it is faster for most riders to go through the UW. Here is a breakdown of the stops you skip, from West to East:

      8th NW — Via UW: 5 stops. Via Ballard: 6 stops.
      Zoo — Via UW: 4 stops. Via Ballard: 7 stops.
      Wallingford: Via UW, 3, stops. Via Ballard, 8 stops.

      Furthermore, if you are close to the zoo (e. g. Aurora and 45th) then you are just going to take the E to downtown. In the case of Wallingford and 8th, taking the train as a way to get downtown is reasonable. The thing is, Wallingford is likely to have way more riders.

      Of course, there are valuable stops on the way if you go through Ballard. However, the most important ones are close to Westlake. To reach them, it makes a lot more sense to head the other direction on the Ballard line. So (to break it down again):

      8th NW — Ballard is a good way to go to Lower Queen Anne and South Lake Union.
      Zoo — Again, you just stay on the E and either walk to your destination (South Lake Union) or ride the train (or perhaps a bus) the other direction (to Lower Queen Anne).
      Wallingford — It makes more sense to go through the UW, and then transfer at Westlake.

      The only destinations where riders really come out ahead going via Ballard are the least popular ones (Interbay and Expedia), just as the only origin that really comes out ahead by going via Ballard is the least popular one (8th NW).

      In short, you can make a stronger case for building the Wye in the U-District (especially since you pick up Capitol Hill). But the case isn’t nearly as strong as it would be if Ballard to UW was the only line serving Ballard.

      1. On the other hand, ST’s current plans assume no Y in the U-District, and would probably preclude one without substantially interrupting service. Given that unfortunate situation, I’d prefer a Ballard Y over no Y.

      2. Of course you can make a better case for building a wye in the U-District; you’ve been making it for a decade and nobody at ST has listened. And rightly, because THERE WILL BE NO ROOM for the trains. And they have made a wye extremely unlikely by not stacking the platforms. To go from north to west a new tube would have to drop down (probably) or rise up (unlikely) enough to cross the heading of the existing tubes. That would put the north extent of the curve at least north of 55th.

        d.p. had a good plan for a service connection, but I don’t think that the tunnels were planned for it. It would have required an underground box to be provided for the cross over and connection turnout. You can’t just poke into an existing tube and add turnouts. To build such a connection ST would have to mine out a volume around the existing tubes, build a retaining box and then quickly — oh let it be very quickly! — remove the tubes and replace the track structure within the box. It would be a HUGE fustercluck of a service interruption.

        So, let the service connection be built in Ballard as a part of the Market Street Station” construction.

        The point of the second downtown tunnel is to serve in-city Seattle riders. So if there is to be a truncation of North Central Seattle direct-to-downtown service ST will prefer it be through Ballard rather than the U-District.

        So, and this is important, the transfer at 15th/17th and Market has to be much easier than the one at U-District Station. If there ever is an east-west line across Wallingford, I personally believe that it would be better to have it go under 43rd than 45th with the platforms to the east of Brooklyn and entrances actually just on campus to the east. It puts it closer to the center of gravity of the new U-District development and sets the line up for a station on the east edge of campus downhill from the dorms that would be very popular.

        But that makes the walk to transfer to and from North Link longer.

        Whichever street they choose, please let us hope that they leave some unused wall in the mezzanine at UDS for an underground connection. Please ST, at least do that for possible future expansion!

      3. For clarity, there would certainly be an entrance to the south at Brooklyn for the east-west line, because that’s the epicenter of the coming development. I did not mean to imply that the only new entrances to the complex would be east of 15th NE. There could be a pair at The Ave as well.

    3. John,

      Well said. There is exactly zero need for Ballard-UW except as a bus intercept. The traffic on the 44 is one 60 foot bus every ten minutes — occasionally nine. Ooooh! Oooh! We needed Link yesterday!

  20. Agreed, I haven’t heard anyone seriously suggest a drawbridge. Both should be tunnels (full stop)
    Related, Im dissapointed you have no demands regarding grade crossings. The only ones we should be considering are in sodo. 15th should be at grade until before the crossing, but is an area already largely grade separated.
    Thank you for emphasizing dead end project termini. It is a short sighted trend we have let happen on too many of these projects. The cost of study and design to optimize is so inexpensive compared to clawing out of the boxes we put ourselves in without forethought.

  21. A couple of notes after reading the comments:

    1). A draw bridge and high bridge would be the same cost.

    2). A high bridge would be 14 stories tall for a short span in the center over the ship canal. There is a long span and a helpful hill between that point and where a Ballard station would go. The line should be able to get to a normal station height in that span.

    3). This bridge crossing is not just about the opening line. In the future there will be interlining north of the ship canal. Even before that happens its very likely that Ballard/DT will run more frequently than the Rainier Valley section can via turnbacks at Stadium due to demand. RV maxes out at 5 min frequency.

    4). Done properly, a high bridge would be incredible to ride and great to look at. An iconic “Seattle Must” on both fronts.

    I’m focusing on the high bridge because $600M is a lot of money. That said, the budget for Ballard/DT is bonkers on a $/mile basis. It could be possible.

    Side note: They also budgeted more time for the constuction phase of Ballard/DT than its taking to build crossrail. So…

    1. For what its worth, high bridge is my first choice as well. It is as cheap as a drawbridge, and has all the advantages of a draw bridge (great view). But if folks object (and they likely will) and we have the choice between spending a bundle on a tunnel or building a draw bridge (that is higher than the Ballard bridge, and unlikely to cause a noticeable delay) I would prefer a draw bridge.

  22. Both the Duwamish and Ship Canal should be crossed with “Immersed Tube” (aka “Sunken Tube”) tunnels. Their construction cost would be comparable to a “high bridge”; and depending on the alignment/grade of the rail bed prior to the approaches, going down into a (relatively shallow) immersed tube tunnel might very well result in less elevation change (resulting in shorter approaches) than climbing to the height of a “high bridge”. And once you’re already designing/building the tunnel segments, you could absolutely incorporate lanes for bus transit as well.

    Other advantages of an immersed tube include:
    –considerably more cost effective than alternative options – i.e. a bored tunnel or a bridge.
    –Their speed of construction
    –Minimal disruption to the river/channel, esp. since crossing a shipping routes
    –Resistance to seismic activity
    –Safety of construction (built in a dry dock instead of underwater)
    –Flexibility of profile


    1. “Immersed tube” is the method used on the Transbay Tubes on BART. A common language term we all know today is “drag and drop”. You build a tunnel section at a site next to the water, launch it with (big…) “waterwings”, drag it where you want it to be and flood the waterwings carefully.

      On the BART tubes divers went down a welded the joints, but I expect that with modern engineering we could have sections that self-join.

      Yes to this.

      1. yeah, “drag & drop” — but oh man the size of the computer mouse you need…..

        seriously though, this ain’t a new idea (first used under the Detroit River in 1903), and it’s fairly common.

        a partial list of Immersed Tube Tunnel Projects:

        1. Michigan Central Railroad (first tunnel in 1910)
        2. Airport Express (MTR)
        3. Marmaray (world’s deepest immersed tunnel)
        4. Cross-Harbour Tunnel
        5. Eastern Harbour Crossing
        6. Sha Tin to Central Link
        7. Tsuen Wan Line
        8. Western Harbour Crossing
        9. Busan–Geoje Fixed Link
        10. Tokyo Bay Aqua-Line
        11. Detroit–Windsor Tunnel
        12. George Massey Tunnel
        13. Louis-Hippolyte Lafontaine Bridge–Tunnel
        14. 63rd Street Tunnel
        15. Baltimore Harbor Tunnel
        16. Chesapeake Bay Bridge–Tunnel
        17. Detroit–Windsor Tunnel
        18. Downtown Tunnel
        19. East Side Access
        20. Fort McHenry Tunnel
        21. Hampton Roads Bridge–Tunnel
        22. Midtown Tunnel (Virginia)
        23. Monitor–Merrimac Memorial Bridge–Tunnel
        24. Posey and Webster Street tubes
        25. Ted Williams Tunnel
        26. Transbay Tube
        27. Sydney Harbour Tunnel

        Between Bertha and the LINK work I know that everyone in Seattle is very conversant in boring machines, but it ain’t the only (or necessarily the best) way to go — especially with these marine crossings.

    2. How deep are the ship canal and the duwamish? As both are navigable waterways I have to wonder if there is enough depth in both places to accommodate an immersed tube tunnel.

      If they are deep enough then it ought to be part of the discussion. Maybe in conjunction with a cut and cover tunnel/station …..

      1. in any case, by design, it seems that a “trench” is (always ?) dredged across the channel before the segments of an “immersed tube” are set in place, and then it is buried with a layer of rock. This is at least in part to protect the tunnel from external damage — a sinking ship, an anchor dropped in an emergency, etc. But to answer your question….

        re: Channel water depths–

        the Ship Canal has an official “controlled depth” of 30′ of water in the navigable channel. That being said, it has not been dredged in a while and has silted in to somewhat less than that — on average 26′ deep at center, 21′ along the sides of the channel as of surveys done in 2010.
        Basically, The Large Lock has a designed minimum water depth over the sills, so the Army Corp would want to allow anything that can make it through the locks to be able to navigate the channel…..

        What is commonly referred to as the Duwamish Waterway actually begins with the East and West Waterways (on either side of Harbor Island), and then the Duwamish from there on up. The intention is to maintain 34′ water depth (at low tide) in the East and West Waterways; 30′ depth in the Duwamish to the 1st Ave South Bridge; then 20′ depth from there to the 8th Ave Bridge; and finally 15′ depth up to the 14th Ave. bridge, which marks the end of the commercially maintained navigable channel. Again, these are designed controlling depths, and depending on silting-in recent dredging the depths will vary — but any tunnel crossing would have to maintain these official depths.


        nautical chart which includes ship canal:

        nautical chart which includes Duwamish Waterway:

Comments are closed.