Ballard Light Rail At-Grade

101 Replies to “Ballard Light Rail At-Grade”

1. I mean, would it be crazy if we just built both of the at-grade alignments? We’d be getting some serious geographic coverage of central Seattle for the first time.

   1. I wonder why my suggestions over the last decade or so have been ignored.

      Community Transit?!

2. With the at-grade experience with our street cars and Link along Rainier Valley, I don’t think there is appetite for at-grade alignments within the urban core, but what about using short automated elevated trains along those routes and avoid tunneling? I really like the Leary Way / Westlake Ave route as it has the potential to branch in Fremont towards the UW. I could also imagine a Seattle Center / Lower Queen Ann branch. If you automate the line, then branches get much easier as a train can just go in and out, you may even single-track such a short branch.

   1. I also wonder if bored tunneling through/under Queen Anne Hill might actually be inexpensive. Its quite different than subway tunneling and it avoids the huge expense of utility relocation, rebuilding streets, construction impacts. Portals at both ends where you stage construction. A mid-tunnel station for Queen Anne Hill would be nice but expensive… maybe you just skip that with a simple tunnel.

      Also it seems to me the further east the line crosses the Ship Canal the lower and cheaper a bridge would be roughly speaking, no? The cut by Fremont is also quite narrow benefiting a bridge here too.

   2. There are definitely bits and pieces of an automated line that could be elevated and even some that could run on the surface. Along Elliot there is a big stretch that could be completely isolated from cars and pedestrians despite being on the surface. You could probably do that with Westlake as well although a much smaller section (e. g. https://maps.app.goo.gl/YUu1u4Psmvjv1SXU9).

      But elevated rail becomes problematic as you enter the heart of the city. It can be done, obviously (we do it with the monorail) but it gets trickier. If we are at the point where ideas like this are being seriously considered than running elevated should definitely be on the table. The main benefit of this proposal is cost. The main drawback is that it would be slower and couldn’t easily be automated. Elevated might be able to split the difference. A Leary Way/Westlake seems well suited for elevated rail given it is basically a straight shot. From South Lake Union to Westlake Center is not a major automotive corridor (in my opinion) so taking whatever needs to be taken to put up the pylons for an elevated line seems quite reasonable.

      It seems like the trickiest part would be connecting to Westlake. The monorail is basically in the way. You could shorten the monorail (maybe) but then the connection from Link to the monorail isn’t as good. You could somehow try and stack them (but that seems messy). I’m not saying it can’t be done, just that it would be tricky.

      Of course it wouldn’t have to 100% elevated. I could see an automated line being elevated (or surface running) from Ballard to Denny (via Leary/Westlake) until it goes underground. Vancouver does that with the Canada Line. With this picture you can see the train heading into the ground: https://maps.app.goo.gl/oC8VRk9tT1bQuf398. To the right you can see an elevated station: https://maps.app.goo.gl/786M6SjcwNYBRnJG7. They build this cut and cover and we could do the same thing (for a much shorter distance). It mean a small part of the line would be cut and cover (basically between Westlake Station and Denny) while the rest of it would be on the surface or elevated. You would spend more on that Westlake Station but that station is crucial anyway. You would need to get deep enough to go under the existing tracks but my understanding is that they are relatively shallow there. I also think that any extension south (to First Hill) would have to be underground and you would be ready to eventually do that. You spend quite a bit on the interchange area (where you need to) and a lot less north of there.

      Of course you almost immediately run into politics. Folks generally don’t like elevated in their neighborhood although it should be pointed out that people voted for the monorail numerous times. This wouldn’t be a monorail but something more like SkyTrain.

      1. Yes, many people don’t like elevated lines, part of that is the size of it. SkyTrain has done ok, but not great. I still think that’s why people prefer monorail or other modern guideways as I explained https://seattletransitblog.com/2023/12/03/evolution-of-urban-guideways/

      2. This is exactly why I said I like the Route 40 alternative: it sets up for an extension along Fourth and/or Fifth if on the surface and under Fifth if elevated transitioning to tunnel. Skytrain makes the transition from elevated to tunnel quite attractive in the illustration you shared.

         If you go elevated out along Westlake the Queen Anne bluff between Wheeler and Halladay is a perfect place to curve left across Westlake into the hillside and dive down under the Ship Canal. You could have a fairly deep but still usable station at 35th and Fremont and then come back up onto structure somewhere west of 1st NW. There are several run-down lots between the BGT and Leary in which to make the transition.

         Or, you could swing across Westlake at the same place and stay on the hillside, heading upward to just west of the Aurora Bridge and cross with a seventy or eighty foot clearance directly adjacent to the AB. There’s enough room between the Aurora supports and the closest buildings south of 34th to land on the north side, but between the support by the lakeside and the support just north of 34th the guideway would have to wiggle to center under Aurora. To pass through that second support a short section of gauntlet track would probably be required.

         Gauntlet track, as the name implies, is a double track structure in which the train envelopes merge to a greater or lesser degree. Instead of the full nine feet of distance between the inner tracks of each direction the tracks get closer together, even directly adjacent. If the gauntlet to be run is narrow enough (typically, a bridge), there may be frogs that allow the inner rails to cross each other at each end of the gauntlet. I think frogs would not be necessary here, though; the Aurora supports are almost wide enough for a double track railroad to pass through them, but I think not quite.

         Gauntlet is functionally like a section of single-track, but without the need for the mechanism of a pair of turnouts. They are completely “passive” though of course they need strong signal protection.

         North of that last major support the other Aurora supports are plenty wide enough to pass a double track railroad. Anyway, the trackway would clear 35th North, which might need to be lowered a couple of feet, and then the guideway would dive into the hillside in the Troll Avenue roadway between 35th and 36th Avenues North. Thirty Sixth might need to be closed except to bikes and pedestrians because it would be “humped” there.

         The station would be between 34th and 35th; I can’t imagine a better location for serving the entire business cluster between Stone Way and Evanston.

         The tine would then follow the route through the south toe of Phinney Ridge we discussed earlier for a Ballard-UW extension, and I would hope that a junction just inside the hill would be stubbed in for a Ballard-UW line; the two lines couldn’t share a Fremont Station, but the station for the east-west line could be right under 40th for a great connection to the E.

         You’d like the tunnel to portal to elevated near Market; I think farther south is better because of the development possibilities west of Third NW, but that can be decided by the statistical experts and politicians. There might be just as much development along Market by the time this thing would be built.

      3. Martin, read my comment about making the Monorail a “dogbone” between the Amazon campus by the freeway and the neighborhood just west of Seattle Center. Yes, the stations in the Seattle Center dogbone would be “one-way”; a rider couldn’t go from Second West and Harrison directly to Mercer and McCaw Hall, but since the trains would be running in a dogbone they could get there by riding around in twelve or thirteen minutes.

         The Monorail doesn’t have the speed or capacity to be a “trunk” carrier, but it could surely enhance downtown Seattle by making a couple of extensions.

      4. I think a lot of the attitudes about various modes are based on very limited experience. It is like people complaining about a bus being “stinky” even though buses are much cleaner now (and many are electric). It is one big reason that “light rail” became popular. It really wasn’t based on the specific benefits of the mode. It was all about perception. Light meant cheaper, less intrusive, cleaner, newer. Heavy rail was big, noisy trains in big cities full of grit and grime. I’m surprised they didn’t call it “Lite Rail”.

         Same goes for elevated options. This city has a monorail and we like it. We associate it with good times. It is fun and effective. In contrast a lot of Americans associate an elevated metro with Chicago and for a lot of people that means graffiti, crime, dirt and grime. Of course that is nonsense. Europeans would laugh at such descriptions. There are plenty of elevated lines that are quite attractive. Stephen Fesler has an essay suggesting we move Link to Rainier Avenue and elevate it: https://www.theurbanist.org/2022/01/19/the-case-for-improved-light-rail-in-rainier-valley/. He makes the case from aesthetic as well as transportation standpoint. Elevated lines can be attractive. They have them in Paris. Of course it looks nice. It’s Paris (https://maps.app.goo.gl/YQsYkEFUreMy1SQx7).

         But that is in part why we ended up with rail on the surface in Rainier Valley. The folks there wanted the line to be buried but they sure as hell didn’t want it to be elevated. So a compromise was made — run on the surface. Hard to say whether their would be opposition to more elevated rail in the city. Most of it — whether built or planned — is in industrial areas or close to freeways or major (ugly) arterials (like Elliot/15th). The original plans were to have the line in West Seattle be elevated — very close to one of the more attractive, urban parts of the peninsula (The Junction). It does seem possible (although not there apparently).

      5. “But that is in part why we ended up with rail on the surface in Rainier Valley. The folks there wanted the line to be buried but they sure as hell didn’t want it to be elevated. So a compromise was made — run on the surface.”

         Only part of that is accurate. The first American light rails in the 1980s followed a pattern: 95+% surface-running to keep capital costs low. The precedents were Portland, San Diego, and San Jose. (I don’t remember if Sacramento and Dallas were built then or later.) Link was modeled after those to keep capital costs around 20 million per mile. That was seen as crucial to getting a yes vote because voters are stingy about transit. The downtown tunnel was a given because it was already built, and the tunnel would have to be extended to the U-District because of the hills and Ship Canal. But otherwise it would be surface where topography allowed it.

         So the original vision of Link was surface from Intl Dist around the northeast side of Beacon Hill to Rainier Valley and on to SeaTac and presumably Tacoma (in the middle of Pacific Highway).

         Rainier Valley residents were divided: some wanted tunnel, others surface, others elevated, others nothing, for different reasons. Tunnel advocates said it would be best for neighborhood walkability/retail. Surface advocates said it would be the shortest walk to the platform. Elevated advocates said it would be less expensive than underground but still allow it to run full speed. “Nothing” advocates like Save Our Valley didn’t want light rail at all, and if they couldn’t get that, they wanted ST to do U-District first and then come back to Rainier Valley.

         ST’s position was that only surface was justified in a flat area, and that anything more expensive would be contrary to voters’ expectations. That view prevailed, and is why Rainier Valley and SODO are surface.

         The next segment to go through design was TIB. ST wanted a surface alignment on 154th and Tukwila Intl Blvd. The City of Tukwila objected, saying it had just beautified Tukwila Intl Blvd and didn’t want it cut up again, and objected to ST taking a corner of Southcenter’s property. ST acquiesced to Tukwila’s demands and made it elevated at TIB station and on the the Airport Freeway. Then, because it had to clear several highways between there and Rainier Beach and it couldn’t have steep inclines, the entire thing was elevated from TIB to almost Rainier Beach.

         The next piece was Roosevelt. The original plan was for it to emerge from the tunnel at 63rd (Ravenna Blvd & I-5), and continue along I-5 to Northgate. The Roosevelt neighborhood asked for an underground station in the center of the neighborhood. ST acquiesced this time, unlike in Rainier Valley. Further engineering studies also showed it would be less expensive to extend the tunnel to 95th instead of weaving up, down, and around I-5’s infrastructure. The freeway was fifty years old and at its end-of-life, and if ST damaged it, it would have to pay for restoration, which would be very expensive. So extending a tunnel that was going to be built anyway turned out to be cheaper.

         Then grade-separation became the norm, and at one point in ST2, everything was grade-separated. But then Bellevue wanted a short tunnel in front of City Hall. It said it would pay for half of it, and begged ST to economize in other areas for the other half. That’s what caused the Bel-Red segment to be lowered to the surface. Also, East King was originally going to pay for everything east of Intl Dist station, including Judkins Park station. East King begged North King to take on the cost of Judkins Park station and the intermediate segment, to free up money for Bellevue’s tunnel. North King and the then-Seattle mayor agreed.

         Later transit fans realized Judkins Park station would give route 7 riders faster access to downtown than the 1 Line, and direct access to the Eastside, and more development appeared in Judkins Park. Thus it became clear that Judkins Park station would be very beneficial to North King after all.

         After that, everything was grade-separated again. And when ST got worried about the number of collisions in Rainier Valley and SODO stacking up, it vowed to have no more surface with level crossings ever again. (Surface in a freeway right of way leveraging the freeway’s under/overpasses to avoid level crossings doesn’t have this problem, and is part of Lynnwood Link and Federal Way Link.)

         “Hard to say whether their would be opposition to more elevated rail in the city.”

         Of course there’s opposition. See West Seattle. The representative alignment in the ballot measure was elevated. After the vote ST activists revealed their opposition to elevated and asked for a tunnel. It wasn’t just for the houses/businesses that elevated would displace; it was also the aesthetics and noise of elevated they objected to.

         The Monorail project had similar pushback: 2nd Avenue businesses objected to it running past second-floor windows and people looking in, and to losing a car lane for the stanchions. That convinced the monorail team to move it to 5th Avenue. Apparently 5th Avenue businesses have less clout. This opposition of elevated downtown was part of the general opposition to the Monorail.

      6. Tom, the monorail dogbone is certainly interesting.

   3. ALRT.
      Box on 5th Avenue, and maybe CAF can adapt the Baltimore Link Purple rolling stock for Linear Induction traction like Bombardier.

3. When the at-grade options were studied in the 2010s, was the assumption that the 3 Link lines would go through the original downtown tunnel? This suggests that ST thought it would be feasible to run 3 lines in 1 tunnel…

   1. It’s certainly “feasible” in the sense that there’s no technical issue. I don’t think anyone disputes that. There are three main operational objections to putting three services on a single two track line:

      1. You’re forcing each branch to accept 1/3 of maximum capacity, or if you want one branch above that on one branch, another has to go below. This is how Washington DC ended up with 12 minute headways for a heavy rail subway (Blue Line).

      2. The operational challenge of merging three lines at one choke point means you’ll never achieve your theoretical maximum headways. Trains and train operators just need a little time and space to safely merge, so even if your system can have a train come “every 2 minutes”, you’ll never have trains zipper merging at that rate from the three branches. This might be less severe for Link since the the West Seattle branch point is one station south of the East Link branch point.

      3. It creates a large set of single points of failure for the entire system. In the Seattle case, a hypothetical southbound issue at U District has now affected service to Rainier Valley, West Seattle, and the Eastside. Again this is well illustrated by issues in DC over the past decade.

   2. No, the at-grade option was never routed through the tunnel in the 2010’s design. Either stubbed at westlake or continued down 1st avenue

   3. The three-line assumption was made for the early West Seattle corridor, but not Ballard/ SLU.

      The Ballard/SLU study assumed every alternative would begin at Westlake specifically.

      WSDOT prepared a macro summary presentation of all the studies, linked below:

      https://wsdot.wa.gov/partners/erp/background/ERP%20150505%20HCT%20Corridor%20Studies.pdf

4. On the comparison chart for 15th Ave. options, it’s not clear how much of the travel time advantage of the light rail option over the D line comes from the actual construction work vs. the train simply having a routing that is more direct which serves fewer stops. If we wanted to, the D-line too could be made faster overnight, simply by making it an express route that skips Lower Queen Anne and serves only the stop locations shown in the first diagram.

   The problem, of course, is that by converting the D-line to an express route, you lose coverage, so you either have to pay the service costs to run the existing D-line alongside it, or force people to transfer for some trips that are one-seat rides today. In other words, a large chunk of the light rail time advantage over the bus is not really anything inherent to light rail, but a service decision, and having the Ballard->downtown route be rail, rather than bus, doesn’t really make these tradeoffs go away.

   1. Yes, that is a very good point. I could easily see the D skipping Lower Queen Anne. The 8 could be extended to Smith Cove. That would mean a transfer for some riders; roughly 1/6 by my calculation.

      To be fair, that isn’t the only thing this would do. While it would run on the surface there would be a new bridge. You would also eliminate the delay caused by serving Dravus. But those improvements could both be done for buses for a lot less money than running rail the whole way.

      Not only that but you could make incremental improvements. For example you would start with right-of-way. Make it a BAT lane all-day long (not just peak). Fix the situation for Dravus — add a bus stop underneath the overpass (both directions). The bridge is tricky. One option is to spend money and improve the approaches (from both sides) so that buses can basically skip to the front (or close to it) when the bridge is up (similar to what exists on both sides of the Montlake Bridge). Another option is to just take a lane. Seattle is slowly transition to one lane (each direction) for each arterial. At that point it might make sense to “re-imagine 15th” and run the buses in the middle and open the street so that you can just walk across, like this idea for Aurora: https://i0.wp.com/seattletransitblog.com/wp-content/uploads/2024/03/segment1_busway-1.png?w=986&ssl=1. Maybe you don’t worry about a bus stop at Dravus but have one at Bertona instead (there is pedestrian access from both directions). The crossing there would only be used by pedestrians. If the 31/32 were both sent to Magnolia via Dravus (as I think they should) this would actually be an easier transfer (either direction). Just walk from the center stop to one of the curb-side stops.

      Again, this wouldn’t have to happen all at once. If the biggest selling point of the surface-rail proposal is that it can be done sooner then this would be much better. You could get improvement right off the bat by restructuring the routes while spending more on service. Various improvements to right-of-way could occur very quickly as well.

      I’m not saying that is the best approach. I think an automated metro from Westlake to Ballard has promise. But it is the major speed improvements (and the high frequency) of such a system that adds value. But if you are going to make the sort of small changes described in this essay you might as well just do that for the buses.

      1. What about the safety concerns and community segmentation that come with at-grade?

      2. For using streetcars, they are much lighter and can stop in a much shorter stopping distance. (for comparison articulated bus around 50~70 thousand pounds) the seattle streetcar is around 66 thousand pounds. While accidents are of course not ideal, I don’t really see how it is better/worse that the center lanes has trucks (up to 80 thousand pounds)/cars instead of a streetcar or bus. Especially for 15th ave/ elliot. For downtown seattle, it’d be traveling slower (for better or worse) and it isn’t that different from trams in european cities.

         > What about the safety concerns
         For light rail vehicles, it’s a bit of a different story. Each car is around 100k and then two of them would be at around 200 thousand pounds. While it is better than four of them for braking distance purposes it’ll still be hard to stop soon enough. (Idk how to calculate this exactly). It’d be similar to san diego trolley.

         > … and community segmentation that come with at-grade?
         If you’re talking about the 15th/elliot unable to cross that’s the existing street design, I don’t think light rail could make it worse. For downtown, there weren’t plans for close intersections, that’s why the light rail/streetcar is limited to 2-cars due to the block length.

   2. We already have an express version of the D called route 15. It’s currently suspended and was unidirectional peak-only, but it could be frequent all day. People have been asking for that for years or for the D to bypass Uptown to get reasonable Ballard-Downtiwn travel time but Metro has always refused. ST could provide it. That would be better than the current 30-45 minute travel time

5. I think it’s worth mentioning that the alignment in ST3 was created without being an alternative earlier. The 2014 alignments (shown in the article) were either paired as Interbay and Belltown, or South Lake Union and Fremont. And of course DSTT2 appeared after these studies too.

   The cost increases are due partly to the inaccurate estimates created in 2016 because they were not studied. ST also implied in the process that the stations would not be very deep. ST likes to imply that inflation is to blame — but it’s obvious to me that the source of the initial low cost estimates put into ST3 were always ridiculously off. Both the SF Central Subway and the LA Regional Connector projects were well into construction by 2016 and their per mile costs were orders of magnitude greater than the estimated cost into ST3.

   1. Yup, and I begged the previous editors of this blog in a bunch of open threads to reach out to Scott Kubly, the genius behind the Seattle ST3 proposal, to discuss what he was thinking. (Martin Duke and the old STB regime were tight with Kubly and wrote a few flattering pieces about him).

6. There is a strong argument to be made for running *part* of the proposed Ballard Link on the surface. From Mercer to the ship canal there are only a handful of cross streets that don’t have overpasses. Gilman is the biggest but you could easily force cars to go north to Dravus and cross there. From a pedestrian standpoint there is very little reason to cross over Elliot/15th. If and when the city develops the Armory area (south of the Interbay Athletic Complex) this could change. But it still wouldn’t be like Rainier Valley where the train cuts through an urban area with major (surface) cross streets. The case for surface running along this fairly long section (about two miles) is pretty strong.

   But once you go south from there the idea collapses. It makes more sense to run buses. There are a couple reasons for this (and to a certain extent they go together):

   1) The fundamental advantage of rail is capacity. Right now we don’t need rail for any of our buses. (In contrast Vancouver needs rail for buses on Broadway). But if some of our buses were really fast then they would be very crowded (which means you would want rail). In this case though, the trains would not be that fast. At least not fast enough to justify the extra expense of rail.

   2) Buses are a lot more flexible. Imagine we make the same right-of-way improvements for the bus that is proposed here. For sake of argument consider the “15th Ave / Elliot Ave” option. So this would mean a new bridge just for buses (and maybe bikes) along with other improvements like the buses going under the Dravus overpass (with a stop there). Ideally you would then want the buses to go into a tunnel but since you aren’t proposing that for the train the buses would run on the surface as well. But the buses would not have to follow one route. If the increase in speed leads to a lot of demand then run the 15, 17 and 18 again (they are currently suspended for lack of demand). But not just during peak. I could easily see the 18 being an all-day bus to complement the D. The same thing is true for the other option. A completely new Fremont bridge (used by transit and buses only) could be used by a bunch of buses going various directions. You’ve got the 31/32 which ties together Magnolia, Fremont, SPU and the UW. The 62 connecting Dexter, Fremont, Wallingford, Roosevelt and Sand Point. All these buses — not just the tram replacement for the 40 — would be much faster. Other options exist as well. The 5 could drop down into the heart of Fremont — thus providing good bus service from Lower Fremont to Upper Fremont, Phinney Ridge and Greenwood — and then follow this new, enhanced transit corridor. With an easy crossing of the ship canal the 28 could go over to SPU and then merge with 3, 4 or 13. A new transit bridge — used by several buses — would add a lot more value than just one new tram.

   Reece Martin once called trams a niche mode. They make sense when the buses are overwhelmed but you don’t want to invest in a metro. In this case neither really applies. It should be one or the other — either build a full fledged metro (ideally automated) or make improvements for the buses.

   I should note that our light rail system is effectively a metro (it just happens to be light rail). While people whine about the surface running in Rainier Valley (and SoDo) that is mostly a safety issue. From a transportation standpoint it is not much different than a regular subway line. The average speed from Tukwila to downtown is still very fast for a metro (over 25 MPH by my measurement). The main difference is that there aren’t that many stops. If the train stopped every half mile or so from Tukwila to downtown (like a normal metro) it would take just as long (if not longer) even if it was underground the whole way.

   1. I should add that the idea of running on the surface *only* on the parts north of downtown was explored in this essay here: https://seattletransitblog.com/2013/12/06/some-thoughts-on-ballard-option-c/. To quote one of the key paragraphs:

      Lots of people worry about the top speed of transit service, but it’s not very important for in-city services (say, typical trips of less than ten miles, stops about every half-mile), because even a fully grade-separated train spends much of its time accelerating or decelerating for stations; frequency and reliability matter most. Assuming any of these lines will be both very frequent and reasonably reliable, the most important factor to minimize trip time is to avoid extended periods of very low speeds, e.g. slogging at-grade through the city center.

7. The tough thing about Ballard and West Seattle is their distances from Downtown. They seem just too far to work well with an entirely surface/ streetcar approach.

   A hybrid approach could be better — like how Boston’s green line operates or SF Muni Metro operates. In a driver-centered concept, this seems more optimal. And the design question becomes where and how long do cheaper and easier-to-reach but slower surface segments go and where do the grade separated segments go. Where do shorter surface segments make the most sense from an access and cost perspective?

   Of course, ST has already chosen the option that has all the disadvantages of both at-grade vehicle technology (slower max speeds and costly drivers) as well as a fully grade-separated line (deep stations and high capital cost). That’s the macro difficulty here.

   So maybe the first question that should be asked is how much surface running should a driven train have? Consider that if MLK had just 1.5 miles of surface running the discussion around it would be quite different than it is, for example. I would begin by setting that distance target — and then going segment by segment to see where that treatment makes the most sense.

   At its core, the big elephant in the room is to me deciding on the right technology that fits in each circumstance. Automation has really changed the choices but it generally means full grade separation. This is the question that ST will not put on the table because they are limiting the alternatives to be trivial in their variations.

   1. So maybe the first question that should be asked is how much surface running should a driven train have?

      I think it depends on the corridor. 15th/Elliot is very different than Rainier Valley. Notice how rarely people mention the surface running of the train in SoDo. Some of it is the distance but the bigger issue is the crossings. How many pedestrian and automobile crossings are there and how big are they. I think you’ve only got three for SoDo and that’s it. Pedestrians can’t cross anywhere else. In contrast for Rainier Valley I count six full automotive crossings between Columbia City and Othello alone (and several pedestrian-only crossings). It is just a much busier place (and there are no overpasses).

      In contrast 15th/Elliot is not. From Mercer to the bridge you already have three overpasses. You only have one significant crossing other than those — at Gilman. But cars from Gilman can easily be forced to Dravus if they want to turn around. That leaves pedestrian crossings of which there are few needed. It is a greenbelt much of the way. Unlike downtown it is well suited for surface running.

      That is the whole idea of this proposal: https://seattletransitblog.com/2013/12/06/some-thoughts-on-ballard-option-c/. Surface on areas where speed would be quite good and underground in areas where they wouldn’t.

      1. Rather than suggest track distance, perhaps the definition should be how many at-grade crossings are acceptable — or maybe how much ADT crossing tracks is acceptable — given a 5 miles stretch.

         Like you mention, few complain about SODO. But that may be due to the fact that there is no car traffic next to it and there are by extension no curs turning at the link crossings. That also means that the signal goes for Link and the busway and crossing pedestrians or it goes for the multimodal cross traffic/ activity (two phases). People get less anxious with just two phases. .

         The more I think about the topic, the more I think eliminating turns across tracks on MLK has to be part of a solution. The crosswalk-only signals on MLK had few to no accidents in a recent Blog post for example.

         Anyway, I’ve never liked how the Interbay segment of Ballard Link summarily discarded 15th Ave as an option. It would be very messy to construct and the Ship Canal grade would be a factor, but the west side of Queen Anne will yield more riders than the rail yards and short buildings on Intervay’s unstable soil will ever produce.

      2. Al, Link could climb the approach to the existing Ballard Bridge. To get up to a 70 foot clearance would, yes, require a longer approach, but the existing roadway grades are pretty mild. I think that the ST plan was the 70 footer would have been to the west of the Ballard Bridge, so it would have its own approaches on both sides of the Ship Canal.

         That plan ran into the buzz saw of the Port’s push-back, which is why any Link-only bridge is now planned to be in the Fourteenth Avenue envelope.

         I’ll be honest, I think the alternative that many people suggested — a new mid-level bridge for northbound rubber tired vehicles and (eventually) bikes would be built in the 14th Avenue ROW with the cars switching back to 15th one or two streets north of Market. Initially the bikeway would also be used for one lane of traffic so there were still two lanes in each direction.

         When that was open and serving traffic a 70 foot replacement for the existing bridge for Link and southbound vehicles in the 15th envelope would be built. I think that this is almost inevitable. The Ballard Bridge does not have that many opening cycles left in it; it’s going to have to be replaced, and the only possible Federal help will be for a highway bridge. The Red States have a pretty much unbreakable hold on 52 seats in the Senate for a very long time since Ohio and Pennsylvania flipped. Georgia might even go starting with John Ossoff next year, though it won’t be such a “lock” for the Republicans. At any rate, there won’t be any more big transit aid packages for at least a decade.

         The rail portion of such a bridge could host either existing Link technology or an automated Light Metro that forwent a fishook extension to the U-District.

         Well, maybe an elevated “SkyTrain” could turn up Shilshole and hook around onto Market, but it would be very controversial to try that. It would for sure require taking the oyster restaurant in the southeast quadrant of the intersection. And of course it would be a huge visual interruption all along Shilshole. An at-grade Link could certainly go up Shilshole and turn on up 24th as far as 64th where the streetcar tracks used to turn. But that wouldn’t serve “downtown” Ballard very well.

         At-grade Link could also go up Leary Way with a station a bit southeast of 20th NW then drop into a short tunnel under Market to a terminal station between 20th and 22nd in the B of A parking lot on 56th. The platform could be one level down for comfortable waiting and the “mezzanine” fare-paid cordon at grade level, also with cover but probably not sides. If the platform selection scissors were between the terminal and 20th Avenue stations, a walkway across the east end of the station could continue under 20th Avenue to the south side of Market, allowing boarding and alighting passengers using eastbound buses on Market to have a stoplight-free access to the platform.

      3. But that may be due to the fact that there is no car traffic next to it and there are by extension no curs turning at the link crossings.

         Yes, I think it is all of it. There are only a few crossings in SoDo. There are far fewer crossing per mile. The crossings that do exist have relatively short cycles. All of that matters.

         I don’t think it makes sense to set a bar though. It is just a trade-off. At some point it isn’t worth going on the surface because there are too many crossings (or the crossings are too bad). But you need to consider all aspects of the crossings (as well as how many).

         The more I think about the topic, the more I think eliminating turns across tracks on MLK has to be part of a solution. The crosswalk-only signals on MLK had few to no accidents in a recent Blog post for example.

         I agree. It is tricky though because the “three right instead of a left” approach is not always easy. For example imagine I am driving down MLK and want to turn left on Orcas. Right now I just get in the left lane and wait for the green arrow. But three rights is this: https://maps.app.goo.gl/fb9M7PcU5Y4tx7cm7. I suppose you could do this: https://maps.app.goo.gl/ATYxR4Agx8p7krWP9 but my guess is the neighbors wouldn’t like that. You want an alternative that involves arterials but there aren’t good options for that.

      4. “I agree. It is tricky though because the “three right instead of a left” approach is not always easy.”

         True that. Maybe they should be banned initially at places where pedestrians cross to get to stations. Then ST could see if it helps or not. During that time, a driver could just proceed to the next left turn picket and make a U-turn.

         If it reduces accidents significantly, it’s a good value for the cost. Then maybe buying some parcels to eliminate them could be done next. It’s a much lower cost solution than full grade separation!

   2. “Automation has really changed the choices but it generally means full grade separation.” I don’t get why this is the case. Driverless cars in mixed traffic already are statistically safer than human driven cars, and the train runs on a fixed corridor that can be blocked off with barriers and train gates. And it’s not like the region doesn’t have expertise to be first at this. We’re opening the first rail line across a floating bridge within a year. We designed the long land bridge to overcome with ground stability issues with Fed. Way Link. And we’ve got decades to figure this out. We got this!

      1. Nervous Nellies of all parties will object to at-grade automation. Nobody else is doing it, because it’s a HUGE liability honeypot.

      2. I do think it is only a matter of time before trains that are in mixed traffic are automated. A lot of the trains with drivers that could be automated simply aren’t because retrofitting is expensive. New York is slowly moving that direction but they are nowhere near that. Toronto decided (at somewhat the last minute) to build their new line (the Ontario Line) using automated technology but they are nowhere near doing that for the existing lines. In general it takes a lot of effort to convert a line after the fact. It makes sense to design a line for automation from the beginning. That doesn’t mean you can’t cross your fingers and hope the technology is widespread when it comes time to build it (worse case scenario you just run the trains with drivers) but my guess is ST would not do that.

      3. In Europe there are several metro lines being automated: London, Hamburg, Berlin – but it’s not trivial.

      4. Cars run on rubber tires. That provides more surface contact for braking.

         Link trains run on steel wheels. The pounds of pressure applied to the point of contact is much greater when braking, and it takes a much longer distance to stop a much heavier train as opposed to the distance to stop a car.

         Unless ST was to convert the lines to rubber tired trains, this will be an automation challenge. A driverless car may only have to think ahead for a block to avoid most collisions. That distance for a driverless train is much further.

         I’m not advocating for rubber tired trains. However I do think that it’s a technology that could be considered to resolve some of Seattle’s unique constraints .

      5. Al is correct. Don’t propose driverless at-grade except in freeway envelopes or with emphatic fencing.

      6. If Ballard Link will be a standalone automated line, can be rubber tired as well.

      7. Light rail trains can be brought to a stop very, very quickly. It’s not as if there are thousands of tons of freight behind slack action couplers controlled by an air brake system with a mile of mechanical action delays.

         The standard braking limit is set based on the danger to standing passengers. If you stop as fast as the braking system really allows, it will throw standing passengers off their feet and probably hospitalize a few that hit handrails or are hit by dislodged luggage or what ever.

         So, maximum braking is only used in emergency situations.

      8. @Glenn in Portland,

         Thanks for setting the record straight.

         And you are correct, the risk to standing passengers from abrupt braking is very real. It’s usually not as much of a problem with urban rail where the brake applications are more moderated and smooth, but with other systems it can be very real.

         And it isn’t just buses that have this problem. I was flying back from Singapore once and our 747 was taxiing at SFO. The flight attendants had already unbuckled and were in the aisle when the pilot recognized a runway excursion and slammed on the brakes. Those poor flight attendants were thrown at least 4 rows forward, and this was in business class where the rows are long.

         Nobody got hurt, but I still think about that to this day. Keep your d*mn lap belt fastened until the airplane is at the gate! For your own good.

      9. Glenn is talking about “track brakes” which are a couple of feet long between the axles on each truck or “bogie” and use electromagnets to “stick” the brake shoes very firmly to the track below.

         Normal wheel beakes can lock and cause the wheels to slide on the smooth track surface. Track brakes are like rail grinders;they bring the train to a very quick stop, as Glenn notes, usually toppling a large fraction or standees.

         But the problem is not that LR (or HR, for that matter) trains can’t stop rapidly enough. It’s that automation is pretty poor at identifying pedestrians who might intrude on the trackway.

         If some kid gets sliced up by an automated Link train along ML King Jr Way, it’s going to be a HUGE public relations fiasco. And a gigantic tort payment.

         It doesn’t matter if the radars actually triggered the track brakes, the tort ghouls know ST has enormous pockets. It would be a circus.

         And Ross, New York does not have any Light Rail, at least not yet. Places where the Subway runs on the surface are stoutly fenced.

      10. And Ross, New York does not have any Light Rail, at least not yet. Places where the Subway runs on the surface are stoutly fenced.

          I never said they did. That is my point. New York has trains very well suited for automation — in every respect. They carry huge numbers of riders. They run frequently in complicated arrangements. They are essentially 100% grade-separated. And yet they aren’t automated. Even New York City has not (yet) transitioned to an automated rail system. The same is true with the Toronto Subway. Cities are transitioning to automation — but very slowly. It is mainly new lines that are automated. Toronto is again a great example of this. Late in the planning stages they decided to build the *new* lines as an automated line. This enabled them to gain some of the *design* advantages of automation — smaller platforms.

          The point being that building a surface rail and then just hoping that someday it will be automated is really not a great strategy. If you want to commit to automation at the design stage (which means smaller station and smaller platforms) you really have to have 100% grade separation.

      11. I’ve been on a few MAX trains that have had to stop quickly. Even without using the full force of the track break, it was fast enough to elicit swearing from a few passengers

          Among other things, modem light rail wheel slip detection systems puts sand on the rails if heavy braking is required.

          It’s pretty remarkable how well these systems, as well as magnetic track brakes, can work when needed.

      12. Oh, OK. That’s clearer now. I agree that “allowing for automation” rather than just doing it as a part of construction is dumb. It’s not like anyone needs to wonder if it will work. It will.

          But I continue to believe that it will work safely only in a fully controlled environment. Perhaps I’m wrong, but I wouldn’t want to be responsible for a $10 million tort award if it isn’t.

      13. Lazarus, that’s a vivid example of “good braking in mass transportation vehicles”. Thanjs.

      14. Ross, apologies, you said the same thing in the last paragraph of your reply.

      15. Yeah, I was thinking more about automating Ballard Link as a standalone line, as opposed to automating the entire Link system. It’s clear that even as a standalone line, it would have issues, and big challenges! Still, I don’t see any issues above that cannot, in principle, be addressed with engineering. Legal liability is its own issue, which applies to every at grade rail line in the country! Let the lawyers figure it out. You would definitely want to fence off the at-grade sections, except at intersections, where I would envision train gates. The cost of going slower is less in the Interbay compared with Rainier Valley, because the stretch with intersections where pedestrians cross is much shorter. And you wouldn’t be operating at as high a speed as a grade separated system (also the case for human operated trains!). Unfortunately, it can be hard for LRT to stop in time for pedestrians whether it is automated or not, hence the need to design for avoiding these conflicts.

      16. P.S. I realize if you are talking about going all the way to Westlake at grade, that is much more complicated than just doing the Interbay segment at grade. I would not recommend going through Denny, SLU, and Uptown with an at-grade automated train!

          (Signing as Brandon K. henceforth to distinguish from the other Brandon on here).

8. One of the main factors that drive up costs is the ongoing delays with trying to start a project like this. The time it takes to get these projects started allows labor and material to increase.
   At some point in time, the decision must be made to get started, the sooner you start the sooner you get done and the people can ride the train and have far better access to West Seattle and Ballard

   1. > At some point in time, the decision must be made to get started, the sooner you start the sooner you get done

      That rationale works as long as it is under budget. It isn’t so sound transit needs to wait decades to fund the elevated/tunnel alternative anyways.

      It’s like saying rather than building a 1 million dollar house one must rush forward with building a 10 million dollar house on the rationale that any delays means interest rate/labor costs might increase.

   2. Seriously, ST3 was approved by voters in 2016… 8/9 years ago. Even West Seattle line, the first line, is nowhere near starting construction, let alone advanced design phase. The clock is ticking, the money is burning up every day.

      1. But again, that isn’t the problem. The cost of waiting is minimal. The money isn’t burning up — if anything the longer we wait the better chance we have of actually building the thing. This is an important and misunderstood idea. We don’t just have a huge chunk of money to spend (that is sitting in a bank). We have money coming in every day. We are also spending money every day. We can borrow money as well to build the things we want to build. But we can’t spend too much money too fast or we exceed our debt limit. This is why a lot of these projects get delayed. It isn’t the planning — it is the cost.

         Imagine the following: We finally figured out exactly what it we want to build. Great! It costs 20 billion dollars. Workers say they can build it in six months. Great! Except we can’t, given our debt limit. We basically tell them we can give you a little bit of money every month. That’s fine, but that means that they only build a little bit every month. Or we tell them we can give then ten billion in ten years. That is the boat we are in (more or less).

         This explains why West Seattle Link — a fundamentally weaker project — will be done before Ballard Link. Ballard Link is too big! We knew it was too big from the very beginning. Except now West Seattle Link is really big as well. This is why things keep getting pushed back. They are way more expensive than originally estimated.

      2. What I’m getting at is the money for ST3 (~$56B in 2016) is buying us less product due to inflation and much higher construction costs over where things were 9 years ago.

      3. > What I’m getting at is the money for ST3 (~$56B in 2016) is buying us less product due to inflation and much higher construction costs over where things were 9 years ago.

         While partially true and an easy scapegoat, that is not completely true. The largest culprit are design changes that now overly focus on minimizing community impact. An easy example is say the bellevue downtown tunnel that was argued over for years because of an increase in 300 million dollars.

         Now compare that to west seattle link where we are adding a medium tunnel at an increase of a billion dollars. The long span bridge over the duwamish also doubled the cost from 1 billion to 2 billion.

      4. What Wesley said. The main problem is the cost of the projects. Let me put it this way: Imagine it is the day after the election and planners magically are able to calculate how much money it actually costs to build this thing (the way they currently want to build it). It would still take a very long time because it is too expensive and they don’t have the authority to exceed the debt limit. So you really have two things working in opposition. On the one hand things are getting more expensive. But at the same time ST is constantly raising more money, enabling them to have enough money to pay for the projects (without exceeding the debt limit).

         The whole idea of a debt limit (and its impact on projects) is not intuitive (at least to me). Here is a good explanation: https://seattletransitblog.com/2018/02/28/sound-transits-debt/

      5. The original ST3 2016 idea in ST was to get West Seattle Link open by 2030. The “realignment” pushed it to 2032. The same year, (Tacoma Dome Link was also supposed to open in 2030.)

         However, the original West Seattle Link idea was for an alignment with no tunnels. Adding a tunnel not only takes longer to build but it takes longer to design and evaluate. The more and deeper the digging, the most costly and longer to build things get. Just look at U-Link and how long that took. It was about 7-8 years of construction.

         The delays have also been partly driven by the increasing number of design changes in the corridor. Because ST had assumed such an unrealistically low contingency budget, there was no money to pay for these changes. FTA generally recommends 30% at the EIS stage, while ST assumed 10% years before the EIS stage.

         And ST has always culturally focused on mitigating the construction and final design. They knew there would need to be lots of property acquired. The problem with this is that if public discussion is focused on mitigation in its many forms, the vibe is that the project is more of a nuisance to be mitigated than it is a good investment. The result is a more costly project — and also a project not designed to improve the journey of a transit rider.

         With West Seattle Link now more than triple its ST3 estimate, half of the added cost appears to be driven by three identifiable major design choices/ features added after 2016 (ST refuses to admit the impacts of these so we can’t say for sure):

         1. A deep bored tunnel with a deep vault end station. (ST could choose to forgo the last station and save $1-2B.)

         2. A signature bridge over the Duwamish. That’s a new feature just added this year that adds $0.5-1B.

         3. Station footprints. In every case, ST underestimated the amount of property needed in its 2016 plan. That mistake combined with proposing palatial stations appears to add another $1B.

         So to me, the delays and cost increases were inevitable.

         What I didn’t expect was to see the costs become more expensive per mile than a heavy rail D (Purple) Line subway under Wilshire and through Beverly Hills and Westwood in LA — with only a fraction of the predicted ridership.

         The cost-benefit of West Seattle Link (high costs with limited travel time improvements and ridership) is so bad that it could be the poster child to undermine the entire New Starts program nationally.

   3. For a lot of the projects that is the case. But for West Seattle and Ballard Link it isn’t. Most of high costs are due to a more detailed analysis of what actually needs to be done. In other words if they started construction five years ago it would still be massively over budget (if they built what they are currently planning on building).

   4. Only a third of the tax revenue is going to ST3. The rest is going to ST2 and debt service; i.e., finishing the 2 Line and building south Link to KDM. That’s why WS/BLE is on a long timeline. The full tax stream will be available for ST3 projects only when ST2 is finished and the ST 1 & 2 bonds are substantially paid down. That was expected to be in 2023, but ST2 isn’t finished yet.

9. I’m concerned also with the capacity as mentioned here… two car more streetcar-like trains versus the 4 car LRVs like now.

   1. I think that would be enough capacity but it shows the whole niche nature of this project. Why exactly are we building a brand new rail line instead of building something for the buses that would offer *the same* level of speed improvements? The only reasonable answer is because the trains couldn’t handle the load. If you made these sort of improvements then you end up running buses every couple minutes — that gets expensive.

      Fair enough but with two-car trains you can’t replace that many trains. Keep in mind that running one train (with a driver) is more expensive than running one bus. So the savings for running a train only occur when you can run the train less often. Thus this approach has to find a happy middle-ground; enough crowding to justify a train but not so much crowding that you aren’t better off with a bigger train. It is a niche solution.

      This is different then running a small automated train. Each run of a train is cheaper than running a bus because there is no driver. You really don’t have that “niche” problem. The only real question is whether the operational savings (from running automated trains) and benefits (from running trains quickly and frequently) are worth the capital cost.

10. The Westlake-Leary route is kind of intriguing if built well to maximize speed and conflicts. I’d bypass the Fremont Bridge and go with a 3rd Ave transit/bike/ped bridge. A Fremont station doesn’t need to be right in the heart of the Fremont commercial district given the traffic and poor locations for a station.

    Westlake could be all center-running transitway from Olive downtown to near SPU & a 3rd Ave bridge. Could be worth considering whether there could be benefit to sharing this transitway with buses or if it would slow down the trains. Europe has a lot of these shared transitways between trams and buses.

    Maybe some underpasses/overpasses make sense around congested areas by Mercer and at the south end of the Fremont Bridge.

    Would also be nice if somehow an at-grade route could be designed for future elevated but it seems impossible without a long closure (as would be the case with MLK) as the columns would go where the at grade tracks are plus needed work area for construction.

    1. A Fremont station should be in the heart of Fremont. That maximizes walk-up usage and transfers. A bike/ped connection on a higher-elevation transit bridge makes no sense — bikes and pedestrians, largely using paths right along the canal, benefit most from the low height of the existing Fremont Bridge. A bridge that’s just slightly west of the Fremont Bridge could serve a few very specific trips better than the Fremont Bridge (like Ballard to SPU) but that’s about it. It’s not far enough west to be a good Ballard-Magnolia connection.

       If anything is moved off the Fremont Bridge to bypass the heart of Fremont it should be car traffic, which suffers most from the low height of the Fremont Bridge. Or, rather, buses suffer most when cars clog up the local streets due to the low height of the Fremont Bridge. Without so many cars in the way bridge openings wouldn’t be so much of a problem. To be sure, the Fremont Bridge does open too often, and a better policy would distinguish between working boats and leisure craft, which shouldn’t get to disrupt surface traffic nearly as often as they do. But that’s one of many instances in this country of a stupid policy that would be nearly impossible to change…

       1. If anything is moved off the Fremont Bridge to bypass the heart of Fremont it should be car traffic

          Exactly. We were discussing this the other day on the Bike Blog. If we build another bridge in the area we should build a car bridge at Third Avenue and make the existing Fremont Bridge transit and biking only. Of course it would be much cheaper — and almost as good — if we just take a lane of the bridge (and surrounding streets) and make them transit only. The problem is we aren’t there yet. This city is fundamentally very conservative (in the classic sense of the word). We are taking very tiny steps towards building a city that we should have built decades ago.

11. And once the Feds deny any federal funds for West Seattle and Ballard where will that leave the project?

    1. I wonder if the monorail authority could be then used to build Ballard to UW, with the understanding that West Seattle light rail gets even more priority when the federal funds turn on.

       1. That’s an idea. It could build anywhere up to around $1 billion. There’s a “not light rail technology” provision though. There’s disagreement on whether that can be set aside, and there has never been a definitive legal opinion on it. The provision was added at Seattle’s request, to prevent the monorail money from being subsumed into Link. It doesn’t define what “light rail” is, which leaves open the question of what all it can be used for. But Seattle could push to use it for some kind of Ballard-UW train, and ask the state to repeal the provision if necessary.

       2. If you can’t use it for light rail, you could build the Ballard line with TSB maglev technology and use their superior cost structure. Maybe as an elevated line so that you can build it in phases.

       3. mdnative,

          Who said anything about the Federal funds being “turned on” again? It’s all up in the air right know politically, but National politics are likely turning against big transit projects no matter who’s president.

          Seattle also has huge problems with housing and education, I wouldn’t even count on overwhelming local political support more light rail.

          Here’s something to think about….. for what Sound Transit is spending on 2 light rail lines, it would be possible to build out roads, utilities and new light rail for a new city of over 200,000 people (we’re talking over 20 billion dollars here)

          Seattle is a city where developers buy a million dollar house to redevelop 4 $750,000 condos. There’s no way to build anything cheap in the City, transit or housing. If the Feds decide they’re not helping Seattle with billions for a subway…. it just won’t happen.

          And maybe that’s what needs to happen?

       4. Tacomee, Ballard to UW would need political support from West Seattle. No way Dow, Bruce, or whichever Chamber of Commerce person is running the city is going to tell West Seattle rail to take a hike.

12. This could work, but, please, not on First Avenue. Yes, First needs some sort of trunk transit, preferably with bus lanes north of University Street and some genuine police intervention to the chaos at Pike.

    But West Seattle and Ballard deserve service to the center of downtown Seattle, by whatever technologies their primary connections are made. While, sure, at Pike / Pine it’s a flat walk over to the retail core. But it’s emphatically not “a flat walk” on Marion between First and the office core.

    “They can take the G!” Um, isn’t forcing that just about the most egregious one-stop bus trip imaginable?

    I expect that the tracks can’t go on Third, because of the concentrated weight on the top of DSTT. And the Second / Fourth couplet suffers from the same problem as does two-way on First; granted, it’s somewhat higher on Second at Marion than it is at First, it’s still a nasty climb to Fifth Avenue.

    So the couplet should be on Fourth / Fifth.

    The pooh-bahs are having the vapors now!

    If the City isn’t willing to give a lane on each of Fourth and Fifth for a surface version of Link, then it should shut up when ST tells it “Without Federal funds, there simply won’t be any part of WSBLE built. We can’t afford it, even with an extra ten years.”

    And I personally would vote for the Fremont / “40 Line” version because it serves downtown Ballard better and SLU much better. Extend the Monorail a couple of stations in a loop to handle Uptown and “West Seattle Center”. Yes, something would need to be done to create a matching loop downtown, and you can’t turn into Pine to reach it because DSTT extends to the curbs on both sides of the street; there’s no place for the supports. Maybe turn into Olive and then Howell with a station at Sixth, another and Ninth and have the loop out at the Amazon complex by the freeway. Loop at Court Place over the freeway.

    Instead of turning slight right into the old Westlake ROW (and squeezing together), the guideway could swoop around the Bank building into a station here: https://www.google.com/maps/@47.61275,-122.3374092,3a,75y,318.69h,77.49t/data=!3m7!1e1!3m5!1sUngYyO5zkTuS740xofGNww!2e0!6shttps:%2F%2Fstreetviewpixels-pa.googleapis.com%2Fv1%2Fthumbnail%3Fcb_client%3Dmaps_sv.tactile%26w%3D900%26h%3D600%26pitch%3D12.514820193383684%26panoid%3DUngYyO5zkTuS740xofGNww%26yaw%3D318.689409321085!7i16384!8i8192?entry=ttu&g_ep=EgoyMDI1MDIyNi4wIKXMDSoASAFQAw%3D%3D

    Yes, this is a lot of hand-waving, but I think that the Monorail, could become a useful transportation artery if it weren’t asked to be the major carrier but rather a neighborhood enhancement.

    1. Caveat. The above is only as a fall-back to the consensus BLE-as-an-automated-Light-Metro we’ve come to.

    2. What about turning onto 6th Ave and continuing on it, turning at Spring St and then onto Boren as an elevated line?

13. Too bad Seattle can’t just extend and build out a nimble monorail system… which would be cheaper and have less visual impact than elevated light-rail… oh wait my bad! That was happening, and we killed it.

    1. I don’t think it would have been cheaper than just regular elevated rail (or a mix of elevated, surface and underground). Part of the problem with the monorail is that the folks insisted on monorail. If they had simply said “metro” or even “SkyTrain” it would have been very different. SkyTrain uses two different technologies but they are both very wide spread compared to monorails. Copying our nearest neighbor — one of the best cities in North America when it comes to transit — would have been very smart. Alas, we had other ideas.

14. It would be a major disappointment to build the Smith Cove station at Prospect instead of closer to the Magnolia Bridge, connected to a ped bridge to the Port. Yes, the cruise passengers have luggage and most of them would likely just take a flat rate taxi or an Uber. But the Port itself has over 2000 employees, in addition, the cruise ships and contractors during cruise season. Let’s be a world class city and connect the Port to our regional transit system. You’d also be serving the Expedia campus, just on the north side instead of the south side. Seems like it would also be easier to connect to the Elliot Bay trail.

    1. Except the bus stop that is closest to the bridge (at 15th & Galer) gets hardly any riders, despite being served by four buses. Here is the ridership inbound (towards) downtown on average each day:

       Route 24 — 12
       Route 33 — 13
       Route 32 — 8
       RapidRide D — 69

       In all four cases the stop at 15th & Prospect gets more riders. But ridership there isn’t very good either. In general it just isn’t a very productive stop. 15th & Armory (Newton) for example gets 181 riders on the D — way more than all the buses combined for Galer. Dravus gets 265, Leary gets 456 — you get the idea. Not that many people take transit to Galer and I don’t see how a mode change would make any difference. It is one of those stations they add mainly because there is a huge gap otherwise. If anything there is a case for simply moving it to Armory (but that would upset Expedia).

15. One very good thing about such a surface alternative is that it keeps Line 1 in the tunnel, preserving the RV / South King County (and soon, Tacoma) to UW trips.

    But Ross is basically right, one could make the same sort of right of way improvements for buses that one would provide for Link. Would digging a bus tunnel under the Ship Canal as in the “Route 40” alternative improve bus service as much as would doing it for a train? Well, if the tunnel portaled a bit west of Fremont, it could, though that would deep six running the 5 through “downtown” Fremont and wouldn’t help the 31 / 32.

    1. Would digging a bus tunnel under the Ship Canal as in the “Route 40” alternative improve bus service as much as would doing it for a train?

       Digging a tunnel (for buses or light rail) would be massively expensive. It would blow a hole in these (optimistic) estimates. What would make more sense is to do as folks have suggested in other comments: build a new bridge (for cars and trucks) over on Third and convert the Fremont Bridge to buses and bikes only. It is the same idea in Ballard. Keep the existing Ballard Bridge but make it for bikes and buses only. Then build a new bridge at 14th for cars.

       There would be slightly different goals in these cases. With 15th being the transit route it means transit is the fast option. If you are trying to get from Interbay to 85th & 15th you would take a bus that goes straight up 15th. But if you are driving you have to cut over to 14th and cut back. In the case of Fremont it is having transit serve the main neighborhood (Fremont). Oh, and just because the new bridge on 3rd is “for cars” doesn’t mean that a bus couldn’t use it. I could easily see the 28 going across the ship canal there and continuing straight SPU up the hill. There would just be a lot more buses going over the Fremont Bridge. (In that sense it is like the buses that run on Fourth Avenue South instead of the SoDo Busway.)

       With a dedicated bridge instead of a tunnel it means that the buses would occasionally be delayed by bridge openings, but they wouldn’t be delayed by the backup caused by openings. It is the same basic idea with good surface rail (e. g. Link — not our streetcars): the vehicles are delayed by traffic lights but not by congestion. It is a compromise — just like this rail proposal.

       1. I think some people will be pretty upset about having to go all the way to Third Northwest to cross the Ship Canal, especially folks from anywhere east of Fremont.

          If they’re going toward Magnolia or Queen Anne, it’s probably not an issue. But if they were headed to Westlake if would be pretty irritating. If they’re headed to Dexter they can use Aurora and the Dexter ramps, even if it means a bit of a double back. But folks headed specifically to Westlake are going to be torqued if you force them way west to Third.

          You certainly won’t be able to “get a lane” on the existing bridge without an alternative pathway, though. There are two lanes of traffic which turn left off 34th onto Fremont, and they’re backed up many hours of most days. Trying to squeeze that many trips into one land just wouldn’t work. A new bridge is necessary to make the crossing better

          The ideal would be to keep one lane for general traffic in each direction on the Fremont Bridge, but only to connect to 34th headed east or coming from the east. Ditto at the south end; the only allowable direction would be a left turn; no straight to Dexter or right to Nickerson.

          That allows buses in the westernmost “bus only” lane to turn into the BAT lane on Westlake while the cars turn left next to it.

          Currently there are two right-turn lanes from Westlake to the bridge. The left-most one is a little shorty which would be perfect for bus-only, providing a stopping pocket to allow buses turning north from Nickerson or crossing from Dexter to hold the right-of-way when they are running and otherwise giving the 40 / Rapid Ride free access to the lane.

          Looking north from the landing of the bridge, from left to right the lanes would be:
          1. Southbound bus-only; left or straight at Nickerson / Westlake. Right turns take the existing shortcut on Florentia.
          2. Southbound general traffic; left only at Nickerson / Westlake
          3. Northbound bus-only. Straight through at 34th
          4. Northbound general traffic; right only at 34th using the existing cut-through.

          Fremont between 34th and 35th is just about five lanes wide, though the extra lane is divided between the two directions as extra wide lanes at the curb. The road could be configured such that the right-hand land would become standard width which would be the stop for the 31/32, 61, and 5(?).
          Then there would be an island platform in the north half of the second lane, probably a bit narrower than an actual lane. The 40 / Rapid Ride would stop there, using part of the center southbound lane.

          Southbound would be a single lane for half the block which would be shared by buses and cars turning right onto 34th. From about the alley south the same sort of thing would happen as the northbound island, though at this island all buses would stop and then proceed south across the bridge, moving right crossing the intersection to gain the southbound curb lane on the bridge.

          Cars turning right to 34th could not come from Fremont. They could only approach the intersection from 35th headed east. This would incentivize cars from Fremont headed to the new bridge to turn at 36th. If they wish to go to Westlake, they could turn left at 35th and loop around on Troll Avenue.

          Traffic from 34th headed west of Dexter would just continue using the straight lane up 34th to the new bridge. Traffic from Fremont would turn right at 36th or 35th and left at Third.

       2. I missed that there is already a “straight to Dexter” third lane southbound at Nickerson, so cars should be allowed to cross the bus lane to access that. So traffic from 34th to Dexter would be accommidated.

       3. I think some people will be pretty upset about having to go all the way to Third Northwest to cross the Ship Canal, especially folks from anywhere east of Fremont.

          They would just go over the Aurora Bridge. It is simply what you are used to. If there was a bridge over Third right now and it went away people would be very upset. They would not want to go all the way to Fremont. Think of all the people that drive on Third that now have to detour. Of all the bridges over the ship canal, Fremont is the least essential. Aurora Bridge is right there. You could say the same thing about I-5 and University Bridge but the ramps are not nearly as close to the water. It is quite common for people to just avoid the Fremont Bridge and use Aurora because of the delays (potential and otherwise) involved with the Fremont Bridge (I’ve done it plenty of times).

          But if they were headed to Westlake if would be pretty irritating.

          Oh the horror: https://maps.app.goo.gl/CX3yJkoG6yhdMHca7. Obviously that is pretty far south on Westlake. But there really isn’t much in the area close to the Fremont Bridge on the south side. A relatively small number of people would be hurt by this. For example consider someone living in a house on the north end of Queen Anne trying to get to the PCC. But even now the current route isn’t great: https://maps.app.goo.gl/V9Q5TX3LodRF7qFc9. They have really made it difficult to drive through lower Fremont. Might as well go all the way. Being asked to go around via Third is quite reasonable.

          For a lot of drivers it looks better. The only significant group of drivers that would be hurt by this would be those headed to Lower Fremont — something we want to discourage. Even then the driving alternatives aren’t that bad.

       4. Just to clarify Ross it would be at 3rd avenue nw (also called 3rd avenue w) this is different from 3rd avenue N

       5. Just to clarify Ross it would be at 3rd avenue nw (also called 3rd avenue w) this is different from 3rd avenue N

          Yes, absolutely. I should have specified that. When I think “Third” I think of the arterial (on both sides of the canal).

          It is not a new idea. This article mentions McGinn looking at adding a transit/bike bridge: https://www.seattletimes.com/seattle-news/ship-canal-transit-bridge-back-on-mcginnrsquos-agenda/. The very last paragraph is this one:

          However, the 2012 Transit Master Plan developed by a city-appointed advisory committee shows a possible fixed bridge crossing the canal at Seattle Pacific University.

          (In other words: 3rd Avenue W to 3rd NW). A lot of people looked at that idea and figured it made the most sense to just flip things (I agree). Make the Fremont Bridge the transit/bike bridge and the SPU Bridge a general purpose bridge. That way you serve Fremont with more buses. As I wrote above I wouldn’t rule out running the 28 across the SPU bridge and up Queen Anne, but it would share the crossing with cars. Even then it could have a bit of a head start by adding bus lanes on Leary (for the 40) and bus lanes northbound on 3rd West (for all the Queen Anne buses). Third is actually quite wide — all you would need to do is get rid of a lane of parking.

          While this would make things much better I don’t necessarily think it would be worth it. You can achieve a huge amount with just red paint and a little courage. Just take a lane. It is not trivial but I think you could add enough new right-of-way in the area to allow buses to get right to the front of the bridge from both directions (while still allowing cars to use it). The changes for the 40 are a big step, they just need to go farther.

16. Prediction: When the surface line is competed, the train will be slow and unpredictable because of surface traffic. The battle cry will go out, “Transit doesn’t work. Government work. Down with government. We need a transit system that actually works. We need grade separation. A downtown surface light rail is a trolly car line is a fancy bus service.

    1. It all hinges on whether ST keeps the “rapid” promise. If it has full transit-only lanes, it can run at the same speed as cars without traffic. If it has signal priority, it won’t get red lights. People are skeptical it will really get these, because the SLU and First Hill streetcars and the RapidRide lines were so watered down.

       What even the best surface alignment can’t do, is run at full 55 mph speed like underground and elevated can. So surface from downtown to Tacoma would be at 30-35 mph, while underground/elevated would be at 55 mph like the freeways. (And I-5 is actually at 65 mph.) That gets to be an ever-bigger gap the longer the distance.

       For Ballard-downtown, a reasonable travel time is 10-15 minutes, so the mode and alignment decisions should be based on that. Surface may be tolerable if it stretches to 20 minutes. But we really need to eliminate the 30-45 minute overhead for every trip to/from Ballard to a regional transfer point (Westlake or U-District). That’s just unacceptable for a 3-4 mile distance.

    2. John P Woods,

       You know Portland has light rail without tunnels? And it’s a little slow and janky, but it works fine. Some people in Portland think they need a tunnel, but the reality of how much money that would cost and it’s never going to happen.

       There’s never going to be a battle cry.

       1. @tacomee,

          Portland has a tunnel on their LR system.

          It’s called the Robertson Tunnel, is about 3 miles long and contains the deepest subway station in America, and one of the deepest subway stations in the world. It’s a real engineering feat.

          And it contains two interlined lines, just like Seattle will do someday from IDS to NGS.

       2. Is the MAX tunnel in Portland even in the city limits? That’s a heck of a mountain between Beaverton and Portland…. I don’t know if there was a way to get light rail to Beaverton and Hillsboro without it. Maybe the Washington Park stop is still in Portland City limits?

          But there’s no other tunnels in MAX, right? No tunnel downtown even. There is talk about one every once in awhile, but honestly Portland is handcuffed by the same issues Seattle is (housing and education) so there’s never going to be money for a tunnel in your lifetime.

          Seattle does have a tunnel downtown to Capitol Hill, U District and up north, and that was a smart move because those neighborhoods have density and are in the center of the transit grid.

          There’s absolutely zero reason for a ST tunnel anyplace else. Really, Seattle and Portland are absolutely not New York City. Seattle doesn’t have the density (or the money) to spend on a second downtown tunnel going to West Seattle or Ballard.

          Sound Transit just pushes back the timelines on their unrealistic tunnel plans and now it’s likely the Feds are done helping out. Game over. I think if you’re a real transit supporter the question has to be….. As Sound Transit spent years planning tunnels that aren’t going to be built… should had ST invested in above ground transit that would actually happen in tax payer’s lifetimes?

       3. @tacomee,

          “Is the MAX tunnel in Portland even in the city limits? ”

          Yes. The east portal and the Washington Park Station are both inside Portland City limits. The west portal is just outside the city limits.

          Tunnels are wonderful things, and they really aren’t that expensive to dig. The expensive part is building all the supporting systems inside the tunnel.

       4. Yeah, what Lazarus said. The station (in the tunnel) is not only within the city limits but it is about two miles from there to the center of the city (as the crow flies). From where the tunnel pops out of the ground it is about a mile to the center of the city — even if you were to walk it (https://maps.app.goo.gl/tMJ4314mkjsCvXi49).

          In any event Portland should invest in building a downtown tunnel for Max. Whether they do or not is another matter.

       5. Lazarus and Ross Bleakney,

          I’ve ridden MAX from Hillsboro to PDX several times…. it’s a slow slog, but then so is driving to the airport. One advantage to living in a city with walkable neighborhoods is not having to travel very far for day to day living. A trip to a big city airport (or just across town) is always going to a slog however.

          The last thing Portland needs is a light rail tunnel. Don’t get me wrong tunnels are helpful, but they also cost a lot of money. Portland (and Seattle) are currently broke and the Feds are likely taking away money.

          I think you both know there was never any reason to build a tunnel to West Seattle or Ballard. There was never any reason to build light rail to Tacoma when a heavy rail line already connects the two cities. But when did Sound Transit ever care about the cost of anything? Honestly, Sound Transit is so preoccupied about local politics and not saying no to anybody, every single project had a massive cost overrun and was delivered late. The light rail in Tacoma is so awful that Troy posts plans to fix by tearing everything out and starting over…. but then Troy doesn’t live in political reality. We’re stuck with the crap ST has already built.

          The honest and straight forward way for light rail in Seattle is for Sound Transit to announce there’s just no money for tunnels now. That’s the truth isn’t it?

       6. Every city has needs. You can make the case that neither Portland nor Seattle should spend a bunch of money on transit. But the point I’m making is that Portland building *downtown* tunnels for MAX is a very promising, quite reasonable project that would add a lot of value. It is quite likely it the best major project (in terms of cost/benefit) that Portland could build. You can’t say that about *most* of what is in ST3. You can’t say that about burying (or elevating) the line in Rainier Valley. Safety aside, trains running down Rainier Valley is really not that bad. It probably doesn’t make my top-ten list of mistakes made by Sound Transit.

          Yes, Portland gets by running trains on the surface through downtown but that isn’t a good thing. It is one of the major weaknesses within their system. It leads people like Jarrett Walker — a transit expert and Portland resident — to suggest that maybe Portland would have been better off with a BRT system (https://humantransit.org/2009/11/brisbane-bus-rapid-transit-soars.html). I really don’t think that would be the case if Portland had built something smaller and automated. It is certainly not the case in Vancouver BC. No one with any sense would suggest that Vancouver would have been better off with major investments in the bus system instead of SkyTrain.

          I’ve said this before. The three Northwest cities have three different models. Vancouver embraces an automated rail system that is only now extending deep into the suburbs. Portland went with a very extensive light rail system that extends fairly far into surrounding suburbs. Seattle is going with a BART-style system that will extend extremely far into surrounding suburbs and cities. Only one of those systems really makes sense — Vancouver’s. That is the only one that carries a huge number of riders. It is the only one that creates a first-class network of buses and trains appropriate for a sprawling West Coast city.

       7. “Slow” and “works fine” are contradictory. I don’t know what “janky” means. That slowness wastes people’s time and adds up over several trips. The proposed Portland tunnel is not to some suburbanish district like West Seattle, it’s right downtown in the middle of the network where the most transfers are. If there’s any place that needs a tunnel, it’s downtown. As Germany figured out fifty years ago, and installed downtown tram tunnels in cities all across the country, in cities like Portland. Seattle, fortunately, built a downtown tunnel too, although it was before rail.

          You need a downtown tunnel in order to provide robust mobility circulation across the entire network, which should be the goal of public transit. A large percent of the ridership have one end of their trip in downtown, or go through downtown, or transfer downtown, so it affects them directly. Even if they don’t, the presence of a tunnel allows higher speed and more capacity, which translates to higher frequency for the same operating cost.

          The Portland region may or may not be ready to pay for a downtown MAX tunnel: it hasn’t been asked yet. It has other priorities like housing too, but just saying transit is so unimportant its needs can be completely ignored is ridiculous, and probably only a minority view among Portland area voters.

       8. “Portland went with a very extensive light rail system that extends fairly far into surrounding suburbs”

          Is it very expensive? We just said it’s almost all surface. I’ve never heard anyone say it has inordinate operating costs.

17. Any plan for surface running needs to be an actual plan. Eg: ML King didn’t work well.

    However, are there places where a better street arrangement can be made? Eg: a four lane road that can be turned into one lane and light rail, while auto trafffic is shifted to another parallel street?

    See MAX along NE Multnomah as an example.

    Or, close the street nearly entirely to road traffic, as done with MAX on parts of SW 1st in downtown Portland?

    1. Well, Sound Transit spent a ton of money on the northbound tunnel…. could they used some of that money to fix problems in the at grade line in Rainier Valley?

       This the problem with building tunnels… they’re no money for anything else.

18. What options offer/preserve the soon encroaching necessity of widening/replacing the Ballard Bridge on 15th Ave. ? Building LiteRail beside it would be a NoGo !!

    1. This is part of the reason why folks have proposed a new road bridge on 14th and a new train (or transit-only) bridge on 15th. You start by building a new road bridge on 14th. When that is done you replace the bridge over 15th. That way you avoid the big outage.

       A new light rail bridge using 15th is pretty straightforward — it would probably be similar to the existing bridge. A new bus/bike bridge is a little more interesting. First off it would definitely be designed for bikes and pedestrians. That is one of the big weaknesses of the current bridge — it is terrible for bikes and pedestrians (people have been killed trying to cross it).

       But the various ramps are also interesting. Right now the bridge becomes a viaduct, crossing roads like Leary after going over the water. The main roadway doesn’t reach the ground until 50th. It is essentially built like a freeway, with exit ramps to Ballard Way (and Leary). It is worth noting that because we aren’t running express buses (like the 15) all of the buses use those ramps. They don’t go over the rest of it. Thus a new bus/bike bridge would not be built that way. Just get rid of the viaduct (between Ballard Way and 50th). All lanes that went over the water would go down towards the surface like the ramps at Ballard Way. This would save a lot of money in the long run (you no longer have to maintain that viaduct). It also means the roadway is a lot narrower — you don’t have ramps and a viaduct — you just have two lanes reaching the surface at Ballard Avenue. This would improve the local landscape considerably — you don’t have all that space for the ramps. You can add a lot of trees while also widening the bike/pedestrian pathways. It also means you could cross 15th much further south. Basically every intersection starting at Ballard Avenue would allow for pedestrian and bike crossing. No more of this crap: https://maps.app.goo.gl/B2LrjAs1AzDYGTMi9 or this: https://maps.app.goo.gl/GXeX1XYQYXUUP1qK9. Just walk across the street (when you get the signal) like this: https://maps.app.goo.gl/dpVdDSwAbuqMUwby8.

       The new bridge over 14th would be built in a similar manner (no viaduct with ramps like the current bridge over 15th). This allows it to be much narrower in Ballard (it is basically two lanes each direction until it hits the ground and then it becomes five lanes). Where it ends (before or after Leary) is debatable. That part of Ballard is industrial — it is just fine to shift traffic around. For example I could see the new car bridge going over Leary (and the other streets to the south) and touching ground just south of 50th (with no ramps before then). You would not be able to cross 49th but because 49th doesn’t go through (https://maps.app.goo.gl/PX7WhjbNDE4Shzic6) it wouldn’t be a huge loss. There would be no left turns from 14th to 50th. So it would basically be a four lane road bridge feeding into a four lane road. This is basically how 15th & 50th is now (https://maps.app.goo.gl/eY4yxmkxoX3FWxjA7) except I would add a traffic light at 14th & 50th and allow cars (and everybody else) to go across 50th. North of 50th the road would widen as necessary if they want to allow left turns. Riders coming from the south headed to Old Ballad would loop around via 11th and Leary (https://maps.app.goo.gl/sMLa7eGdToJc2rEX7).

       1. “That way you avoid the big outage.” And you save money. With a road bridge open at 14th, you can close 15th Ave essentially from the canal to Market; that’s a wide ROW that will reduce the amount of space ST needs to lease elsewhere for construction staging.

    2. The concept of building bridges is lots more complex than building some road or track at the surface. It’s not merely drawing a straight line on a map. There are everything from navigation considerations (span width, depth of tunnel, height of bridge, use of drawbridges) to soil/ bedrock considerations (bridges are heavy and the shorelines are often unstable soils for quite a depth) in addition to how it connects to a street grid or transit network.

       The conceptual problem with this crossing is that light rail spanning the Ship Canal has always been seen as additive — meaning that the obviously needed 15th Ave bridge replacement has never been an alternative component of Ballard Link. (Very ironic in that other light rail alternatives often respond to other non-transit considerations like demolishing county buildings or neighbors wanting to hide trains underground.)

       The idea that Ross mentions of building a new street bridge at 14th first and using the 15th corridor for Link is a great one on paper! However, it’s never been even modestly studied.

       It’s one of the problems with having the wheels of planning Link in Ballard being so limited yet so time consuming. Studies began 8 years ago — and yet every alternative location for the ship canal crossing and a single Ballard Link station vary by just a few hundred feet except at the very beginning when one light rail alternative proposed crossing further west (and was shot down by the Port).

       It’s particularly irksome to me how the navigation channel crossing adds so much to the cost and complicates the design, yet no leader dare suggests that the money should come from anywhere else but transit capital dollar sources. It’s just another case of how transit capital dollars is seen as the cash cow to get other things done. I think there is a good case to make that the Army Corps of Engineers should be more actively involved in both planning and funding this Ship Canal crossings for both Link and 15th Avenue.

19. 19 minutes is not bad at all. I wonder if it is still the case if analysis was updated to based upon today’s condition.

Comments are closed.