On December 2, British Columbia’s Lower Mainland welcomed the latest SkyTrain extension, the long-awaited Evergreen Extension to the Tri-Cities area (Port Moody, Coquitlam, and Port Coquitlam) east of Simon Fraser University. The extension covers 6.8 miles in 15 minutes, roughly the same average speed as Link (27 mph). Most of the line is at-grade or elevated, but there is a short (and controversial) 1.3 mile tunnel in between Burquitlam and Port Moody Stations. The extension makes SkyTrain both the longest grade-separated transit system in Canada (edging Toronto), and the longest fully automated system in the world (edging Dubai).

From a network perspective, the extension operates as the Millennium Line, allowing TransLink to end the loop-de-loop that the line made from 2006-2016, doubling back on itself and serving Commercial/Broadway Station twice. Though the inefficiency of stopping just short of the Canada Line will continue, the Millennium Line will now make more sense as a linear line, operating from Lafarge Lake/Douglas to VCC/Clark. The extension will also offer the first two intermodal transfers to West Coast Express commuter rail outside of Downtown Vancouver, at Moody Centre and Coquitlam Central.

To cover the 2 stations between New Westminster (Braid and Sapperton) that would have otherwise lost service, TransLink has created a new branch of the Expo line. The traditional line from Waterfront to King George in Surrey is now complemented by trains from Waterfront to Production Way/University.

Selection from the new TransLink System Map

Much like the Canada Line, the new Millennium Line will use tiny 2-car trains but at very high frequencies, every 3-4 minutes during peak, 6 minutes off-peak, and 10 minutes at night. Meanwhile, the Expo Line will run every 2-3 minutes during peak on the shared segments between Waterfront and Columbia, with service roughly every 3-4 minutes to King George and every 6 minutes to Production Way/University. The new branch of the Expo line is every third train during peak, and every other train off-peak.

The new network also offers a strange but elegant platform layout where the lines converge at Lougheed Town Centre. Between Lougheed and Burquitlam, Millennium Line trains will crossover and run on the opposite tracks (see 12:35 in the video above). Outbound Millennium Line trains will stop at a new island platform (Platform 3), while the pre-existing centre platform will be shared by the outbound Expo Line (Platform 2), the inbound Millennium Line (Platform 2), and the inbound Expo Line (Platform 1). This is done to facilitate same-platform transfers between Coquitlam and New Westminster.

Diagram by the Author

Though SkyTrain is highly regarded for its frequency and reliability, it shares the suburban biases of most North American systems, underserving the City of Vancouver. But if you want to see a best-case scenario for TOD for Federal Way or Lynnwood, TransLink’s extension offer a hopeful glimpse of Towers in the Park(ing lot).

59 Replies to “Vancouver Welcomes the Evergreen Extension to the Millennium Line”

  1. I still wish they had built the Broadway line out to UBC instead, but oh well. Looks like a nice new line! I’ll have to go ride it next time I’m up there.

    1. Eventually they will. Someone else can chime in, but my understanding is that the UBC extension should be done in five years or so. I have no idea why this extension was done first. I realize both were cut short, but it seems like the UBC is a lot more important.

      1. After the failure of the vote the Broadway extension is still unfunded, but they’re still hoping to start construction in the early 2020s. Even so it’ll stop a few kilometers short of UBC, with SkyTrain as far west as Arbutus, with the 99-B-Line the rest of the way through Point Grey to UBC.

      2. I believe it was regional politics, city vs suburb, the last line, the Canada Line was mostly an urban line benefiting Vancouver. There’s been similar grumblings about the order of construction with the Broadway subway and Surrey LRT lines, though luckily Broadway is next up.

  2. It’s interesting at how Vancouver has branched two of its three trunk lines. With its high frequency, I understand how they can do this. Still, it does illustrate a contrast from ST proposed operations.

    1. Yep, huge contrast. Not only the branching, but in the entire approach. For example the Evergreen Line doesn’t end in downtown, and never will. Eventually it will go out to UBC, but either way, it means lots of people will have to transfer.

      The branch is another contrast. Branching in general only makes sense when you have roughly equal demand on both sides, with the core section being the high demand area. This is a good example of that, as it splits in the suburbs. The other thing about this branch is that they both serve the same number of stops, which means that service to either end takes about the same amount of time. This means that the trains don’t sit at a station waiting for a long time for the other train to go through.

      A split at the UW would have made a lot of sense. Getting to Ballard would be faster, but if Link ended at 145th it would be the same number of stops and roughly the same amount of time. That would have been a logical terminus, although it would have taken a lot more work to get that station in shape (building bus only ramps, for example). Ending at Lynnwood would mean that the Ballard train (even if it ended at 24th) would not be as frequent as folks might like.

      Splitting to serve Lake City or Aurora makes sense from an abstract standpoint, but I’m not sure how it could have been achieved. It starts getting really expensive (lots of tunneling) or controversial (elevated over quiet areas) pretty quick. A split down south to Renton might have made sense if it was grade separated down Rainier Valley. It is hard to see any splitting in the new lines unless the city grows like crazy.

      Of course you could easily view our system as simply a series of branches. While Vancouver, like most subways , has crossing lines in the city (one downtown, one just outside it) we don’t. The only crossing line we have is in Bellevue. The new downtown subway tunnel will be built right next to the old one, with the only additional stop a couple blocks from an old one (built 30 years ago). If they add a stop on First Hill (unlikely) then it would change the dynamic, but for the foreseeable future, we have something that resembles the bus system forty years ago — every trip requires a transfer downtown.

      1. Every rail transfer would take place downtown. Many rail-bus and bus-bus transfer will not require a downtown transfer.

      2. One nice thing about branching to go to Lake City is that LCW-to-Broadway would be a one-seat-ride in both Seattle and Vancouver!

        (confession: I spent almost an hour last night trying to make jokes about place names we have in common with Vancouver and this was the best I came up with.)

      3. @AJ — Yes, definitely. I wrote that and immediately realized it could be misinterpreted. People wonder sometimes why my writing is so verbose and ponderous. It is because when I state something that is fairly clear (how rail transfers will take place downtown, and how that is quite similar to bus transfers of old) the statement can be taken literally if I don’t add a bunch of extra phrases. So, to be more precise:

        For the foreseeable future, rail in Seattle will resemble most of the bus system forty years ago — every rail transfer occurs downtown.

      4. @Al — Yeah, funny that they not only have a Lake City, but a Lake City Way.

        I still think the most amusing thing are the last two stops on that line. What if you live in between? Which stop do you take? Ah, the old Lincoln — Douglas (station) debate.

      5. Seattle is more downtown-centric than most American cities. Part of it is geography, with the water on both sides puahing people in a more north-south orientation and downtown being in the middle of the isthmus, and part of it is economic because jobs didn’t flee as much as in other places. In Dayton OH I’m told downtown is no denser than the suburbs and has no more jobs, so suburbanites have no reason to ever go there. But downtown Seattle has more than 10% of the region’s jobs and that was before the growth in SLU. The swap in the 80s that movrd Nordstrom into the Frederick & Nelson building, created Westlake Park, and built a city-owned parking garage at Pacific Place was notorious because it spent funds intended for low-income housing, but it paid off hugely in terms of revitaliszing downtown. It continually stuns me how thick the pedestrians are on Pine Street even at 9:30pm.

        The PSRC has a report that downtown is heading toward a transit capacity crisis if it doesn’t significally increase capacity, with tens of thousands of people who won’t be able to fit on the 2014 buses or trains to get in/out of the area or circulate within downtown. That was part of the reason for the second tunnel, the increase in RapidRide lines, and moving non-RapidRide routes out of downtown.

        Not having Link on 45th misses the opportunity for crosswise synergy, where people make perpendicular transfers as well as going straight. But we mustn’t overstate the impact. The people who would benefit are mostly along 45th, not all of north Seattle. Somebody coming southbound on the 5 or 26 would not be advised to transfer to the 45th line and again to the Red/Blue lines: that’s two transfers within a mile. Especially with the 26 and 28 being express. Those living further north would be advised to take the 45 or 40 to Link instead. That makes it critical to get good transit lanes and signal priority on those RapidRide lines. The remaining ones who would lose out then are those going south on a bus to a destination on the 45th line. It’s hard to believe that those people would be more numerous than the tens of thousands of people who could more easily transfer downtown. If north Seattle continued further east-west (like the Bloor-Danforth subway in Toronto going several miles out) then it would be a bigger issue.

      6. Thank you, Mike. This fantasy of people getting on a bus at 65th and Phinney, transferring to a relatively infrequent subway at 46th and Fremont and then again to Link’s main stem at 45th and Brooklyn is absurd. San Francisco doesn’t make people living along Quintara change to the N Judah at UC 9th and Judah. The #6 goes the rest of the way downtown. Some people probably do transfer, but most do not. And that’s just one transfer.

      7. OK, interestingly enough, this article is about the Millennium Line. Specifically, it is about the suburban extension. Eventually they will fix the other end, and replace an overcrowded bus with a subway line. Where does the bus go?


        The 99-B Line in Vancouver carries over 50,000 people and it doesn’t go downtown. When they finally finish that essential part of the subway system, it will obviously carry more. It is quite possible that just that tiny section will carry more than our entire line — yet it doesn’t go downtown. It will go to a major university (imagine that) and serve tens of thousands of people who are either OK with a transfer to downtown, or are simply headed somewhere else.

        You are right about someone at 65th and Phinney not likely taking the train to downtown (nice straw man). However, they would take the train if they were headed towards Ballard, Wallingford, the UW, Capitol Hill, First Hill, or the Central Area. The UW is a major destination (likely second in the state). First Hill is also a major destination. Ballard and Capitol Hill are big nightlife areas; the former is a major medical center, the latter has a college. It all adds up, just as it does in Vancouver.

        But just for kicks, let’s play this out. Let’s say someone felt like taking the train to downtown. OK, first you have the transfer to the train. The train then makes a stop before the U-District and then continues downtown. So 4 minutes to the U-District, and 8 more minutes to Westlake. The transfer will likely take a couple minutes (to get down to the platform) and maybe 2 or 3 minutes average wait.* So the whole thing — to get to Westlake from 45th and Phinney — takes a little over 15 minutes. Is is worth it? It depends. If you are headed to Belltown or South Lake Union it definitely isn’t — you just stay on the bus(es). But let’s say you are headed to the south end of downtown, like Pioneer Square, or the I. D.? Unless the bus manages to travel through downtown quickly, the train is probably faster. Or how about if you happen to be headed to the airport? You have to transfer anyway, so you might as well make the transfer early (sounds like a wash from a timing standpoint). Besides, I’ve been told by some people (cough, cough) that reliability is the key — so if it is close you take the train.

        Oh, wait, I know what you are thinking. Somehow we failed to actually build this — https://www.flickr.com/photos/67869267@N07/9152772373/in/photostream/. OK, fair enough. Somehow the east-west train keeps going east. This means that there is something over there worth going to (which means more riders). It also means that ideally the two lines are timed to work with each other. This is trivial, because as luck would have it, the key sections are grade separated. So a west bound train waits for a north bound train. An east bound train arrives a minute or so before the south bound one. So now, with the transfer, you add a whopping 60 second or so.

        Of course that means that the other connection is not timed. Big deal. At least it exists! At least people could get to Ballard, (upper) Fremont and Wallingford from Roosevelt, Northgate, Lake City and places north.

        That is what I mean when I say it transforms the region. The type of trip you mentioned that might not be much faster — Phinney to downtown — is already fast! That is the point. Ballard to downtown is just about as fast as what is about to built (only two minutes slower) but more importantly, it connects everything else along the way.

        These are the trips that people make. These are the trips that explain why folks who work downtown have the highest transit ridership, yet Vancouver has three times the ridership per capita. Not everyone is going downtown for all of their trips. They go all over — for work, school, doctor visits, entertainment or just to visit friends and family. In Vancouver you can do that — with an excellent combination of rail and bus service. In Seattle you can’t.

        * Oh, and Richard, a train from Ballard to UW would be very frequent. It would be the most cost effective line in our system. It would be about three miles, with five stops. Running trains every 3 to 6 minutes would be cheaper and get better farebox recovery than running trains to Everett or Tacoma every 10 minutes.

      8. @Ross–those are great points. The 65th and Phinney example is kind of a worst-case scenario as the Ballard-UW line wouldn’t be as helpful for that fairly specific area (which you deconstructed nicely). However, as another example that I actually lived – for years, I lived in Lake City, have family in Ballard and also enjoy the restaurants, shopping, and nightlife in Ballard. I’ve driven that crosstown route uncountable times and it sucks…to the point I just wouldn’t do it sometimes. For me it would have been a no-brainer to take a crosstown train–EVEN if it meant transferring at 45th from North Link. I just assume that people who are leery of transfers are such because they aren’t used to extremely high frequencies (unlike the horrible train-to-route 11 walk and transfer). Anyone who has spent any amount of time in cities where your only issue is deboarding, crossing the platform or going up/down a level, and waiting at most a couple of minutes for another bus/train knows that it quickly becomes second nature.

        Any trip involving a Z pattern would be helped immeasurably by the vast improvement in cross-town travel this line would provide (getting cross-town by any mode in Seattle has always been a cluster****). Crown Hill to Capitol Hill, Lake City to Ballard, Fremont to the airport, etc etc etc — all would find this line extremely useful, and an urban-focused rail network would have it on the list.

    2. Technically, ST has not committed to how it will ultimately operate the system at the end of ST2 or the end of ST3. I don’t see a demand logic of high frequency trains all the way to Everett, Tacoma and Redmond. Branches could be added to at-grade sections somewhat easily, and maybe even to aerial sections — but the need for room to allow fast, grade-separated junctions could be difficult if it isn’t designed for that. We blew it woth the lack of a junction in the U-District or north of Northgate but it’s not too late to enable them for places like north of Lynnwood, south of Federal Way or north of Interbay or even SLU. The Issaquah-Kirkland line will require the addition of crossing tracks so Eastlink could be branched more easily in the future.

      The root problem is that ST does not design for possible changes to long-term operations or expansion well. Things that are strategic like tail tracks or future grade-separated cross-over designs are often not anticipated or they get axed in value engineering. It’s a cultural blind spot with ST staff and board members. I wish they could see the light.

      1. Yeah, I agree. There has never been much of a long term plan, other than complete the spine, which is not only dubious, but vague. I can’t but think that both the failure of the first proposal (which included the spine) as well as the financial failures of the second proposal had a lot to do with it. Folks basically took the “get a foot in the door” approach, and cut corners at every point. Mount Baker station — arguably the most important station from a bus integration standpoint — is a complete disaster. No work was done to eventually accommodate a line from Ballard to the UW, even though it is obvious, and folks have been talking about it for years (it was part of the original proposal as an “if we have the money” project if I remember right). There simply isn’t any long term strategy other than completing the spine.

      2. ST3 includes what was going to be ST4, so I don’t know how much Sound Transit really expects an ST4 or thinks it’s worth planning for now. It still has two phases of projects to build which will keep its hands full for a long time. The only indication of what it might want in ST4 is the studies in ST3, and did it designata any corridors? That and the lingering expectation of Everett CC and Tacoma Mall. The Long-Term Plan seems like not really “We will do these” as “We might want to do these”.

        The lack of a transfer facility at U-District Station is really odd, The 45th line is in ST’s Long-Range Plan. I’ve been telling ST for years that the station should be designed for transferts to a future east-west line but it hasn’t done anthing about it. The only reason I’ve heard (from back during the station open houses) was that “We can’t spend money designing a transfer to a line that hasn’t been voter-approved yet, and it hasn’t been decided whether the alignment would even use the station” (he pointed to UW Station as an alternate meeting point in some alternatives).

      3. How many rounds of ‘but voters haven’t funded it’ do we have to endure? There is benefit to building rail sidings and crossover tracks even if a new line isn’t funded. Voter approval is not required.

        A new referendum to improve a plan is also easier to pass if it doesn’t require extra funding. The logic that rail lines are fixed is a very lazy attitude. ST has a few years and ample guaranteed funds to explore and improve system options over the next few years. Then they can decide if it’s in the scope of ST3 or they can easily go back to the voters to redirect the dedicated funds.

      4. I think ST views UW Ballard as a part of the north lake Washington crossing (given it lumped them together into one study), so any U District transfer would be perpendicular and grade separated, like Westgate with the 2nd tunnel.

        I think ST has show as much interest in Ballard to downtown via UW as it has in Metro 8. In other words, the “obvious” solution for the commentariat the staff views as not even worth planning for.

        Similar to Metro 8 being rendered unnecessary with the ST3 Ballard to downt via SLU line, a Ballard-UW spur could become unnecessary with a Ballard to Bellevue/Redmond line. Looking at the tea leaves in the LT plan, I think that’s where ST is leaning.

        Not to say there aren’t unforced errors. Not planning for 135th is a good example. But I don’t think U-District bramching is an “obvious ” failure to plan ahead.

      5. I think UW Ballard will actually end up being Redmond to Ballard. A line like that will be easier to fund because it also gives the east side better access to the UW. funding a major subway only with Seattle dollars would take a long time.

      6. @AJ >> Similar to Metro 8 being rendered unnecessary with the ST3 Ballard to downtown via SLU line,

        That is nonsense. The Ballard to to downtown line covers only part of the Metro 8, and it doesn’t even do that very well. It has the Uptown station, and the station in Aurora, but that is about it. The other station (on Denny and Westlake) is way too close to the Westlake station, and way too close to the Aurora station to be of much value. The walk share for the station is minimal as a result. Without a doubt it is a good stop if you happen to be on that line. But if not, it isn’t worth the transfer unless you are very close to Westlake and north of Denny. If you walk a bit east or west, you save about five minutes (more or less), but you lose that in transfer time. If you are headed up to say, Fairview and Harrison you are better off walking while keeping an eye out for a bus. If you are headed up to Fairview and Mercer, it makes sense to take the bus from Westlake. This is all in contrast to a stop at Fairview and Thomas. That stop (plus the one on Aurora) means that every place between I-5 and 99, south of Lake Union and north of Westlake is within easy walking distance to a station.

        But again, that isn’t the biggest reason this isn’t as good as logical Metro 8 subway. The big problem is that this only covers half of it. There is nothing for the eastern side, which is the most densely populated contiguous area in the state. It means that First Hill and the rest of the Central Area is simply left out. CHS is as great station, but it is not a great station for transfers, and is simply inadequate for covering the region. For example, let’s say you want to get from the U-District to TT Minor (my alma mater). Minor sits in an urban setting, adjacent to a census block that is quite high for Seattle (34,000 people per square mile) There are restaurants and shops nearby and the area (like most of the city) is growing. So how do you get there? If you take Link, you have about a 15 minute walk from the station. You can ride all the way downtown (to Spring), then walk a couple blocks, then take Madison BRT all the way back up the hill again. Maybe you take the train, then transfer to the 8 a few blocks to 16th (cutting your walk in half, but meaning an extra transfer). It could be that your best bet is to simply take buses — first the 48, then a transfer to Madison BRT. This is an area that is more densely populated than every census block south of I-90 or north of 65th. It sits within a mile as the crow flies of the station, but since it is diagonal to it, Link is just to cumbersome for it. This is just one of the stops that would make sense for a Metro 8.

        There are stops further south as well. Every stop is high density (for Seattle) or has a connection to another line (or both). It would not only add numerous high density stops, but would connect areas (Capitol Hill and South Lake Union) that are extremely difficult to connect right now, and will be even when ST3 and Seattle are done with all of their projects.

        >> In other words, the “obvious” solution for the commentariat the staff views as not even worth planning for

        I agree, which is the root of the problem. Focusing your subway on high density areas or considering the role that buses play are not radical ideas. It is standard practice. Sound Transit has taken a radical approach and simply denied all available science, experience and common sense when it comes to designing their system. It isn’t that they researched this line and then found it to be surprisingly cost ineffective, it is that they didn’t even bother to study it! Meanwhile, they have studied (or will study) rail to places like Orting.

        Or a Ballard to Redmond Spur. That is another ridiculous idea, that Sound Transit might just propose (for ST4). It is ridiculous because every significant area on the East Side is already covered by East Link. It is ridiculous because Kirkland is a broad, not very dense area. Instead of taking a bus from Juanita all the way to the U-District, you take a bus to South Kirkland, then wait for a train. There are no stops along the way that are being added, so you haven’t added any value. The bus is fast — traveling in one of the few HOV 3 areas — so even if you time it just right, you have gained nothing versus fixing the bus connection to Husky Stadium station (which can be fixed much more cheaply and sooner).

        No, ignoring a spur junction in the U-District is just as much an obvious failure to plan ahead as omitting the station at NE 130th, and both are part of the same problem. These folks don’t know how to build a subway line. They ignore bus integration. They ignore density. They predict ridiculously high numbers for suburban areas and ridiculously low numbers for urban ones (and then trumpet their success when they finally — finally! — build the most urban section). So either they ignored what every transit consultant would consider a decent long range plan, or they knew about it, and simply avoided planning for it. I’m not sure which is worse.

      7. @Brendon — As far as funding is concerned, what your saying doesn’t make sense. Seattle pays for Seattle stops. So if you build a line from Seattle to Redmond, Seattle pays for Ballard to UW. So you are simply adding an additional cost — a new rail line that is completely unnecessary — to be paid jointly by the East Side and Seattle.

        A rail line like that would be completely unnecessary and expensive. You don’t add any value for a crossing — you simply force a transfer. You are much better off simply fixing the section between 520 and Husky Stadium (which should be fixed now, as the freeway is rebuilt). Fixing that should be a joint operation, that both Seattle and the East Side pay for. It should have been part of ST3, but for whatever reason (denial, perhaps) it wasn’t.

      8. Maybe the reason so many Sound Transit stations are so bad for transfers is that they refuse to build the stations under/over the major street where there can be entries/exits on both sides of the street for bus transfers… Capitol Hill (all south of John and east of Broadway), UW (all east of Montlake Blvd), U District (all south of 45th), Roosevelt (all north of 65th), Northgate (well south of Northgate Way). At least they did build the little foot tunnel under Broadway at Cap Hill Station.

        Lets hope they dont take this approach with the new downtown subway and avoid building the stations under the street.

      9. That was probably done to avoid closing Broadway for years of construction. The DSTT construction downtown was not that popular, and trhat”s one of the reasons Link is not on Rainier. (The other being that Rainier was considered too narrow and busy.)

      10. I feel your frustration, RossB.

        Still, I don’t want to miss the root problem with failing to put in junction capabilities. Another example: A junction around Westlake would have been a system game-changer for Ballard, either between University and Westlake or at the Convention Place land. Branches have been proposed around there in many plans from decades past. A branch to West Seattle has also long been anticipated, but there was no engineering for it in SODO and now the ST3 design challenges will be more difficult since there is now an existing line that operates there that must be constructed around. The U-District missing junction track is just one of many omissions.

        As far as infill station mistakes go, we paid lots of money to rebuild MLK and the Graham Station was always a possible in-fill station — yet the design never reserved a place for platforms and station elements! Another ST thinking-ahead omission.

        If you want to look at another looming mistake, look at the ST3 end of line proposal in Ballard. Because of demand (greater than Link in DSTT today) the train frequency will be so high that turning around trains there is going to be a major mess as soon as the tunnel opens. Now, ST could branch the line and have two or more stations on two lines north of the Ship Channel (North and west? North and east? East and west?), easing the impending backup at a single Ballard station — but NOOOO! No one is thinking ahead. It’s a cultural mindset of this is what we’re building under the current funding plan and we don’t care how it can operate or expand. Then, we the problem becomes apparent, there won’t be a proposal to fix it because ‘that’s the way it was built and we are now stuck with it’.

        It’s like planning a major house expansion room-by-room based on only last year’s extra income. It can be done that way — but it’s a horrible and ultimately more expensive and disruptive way to do it.

      11. Good point, Al. You are absolutely correct. Junctions at Westlake and SoDo, along with better headways through downtown (which is likely possible) and you just saved yourself a bundle of cash by avoiding the construction of another tunnel.

        Likewise with Ballard. What is the plan there, anyway? They definitely want to build Ballard to UW. They’ve already planned for it, did the study, and it was a better value than anything else they studied (and will build). So that is pretty much a given. Which means at a minimum a station that serves two lines, if not a branch. Same with the UW.

        Here is a little diagram: https://www.flickr.com/photos/67869267@N07/9152772373/in/photostream/. Now imagine that station had been built just like that. Just the station, as well as the dotted line (that was built). The lines going east-west go a few feet and then stop. That east-west platform is empty. Now imagine ten years after that station is built, we decide to spur like shown. No problem. What if we don’t? What if we decide that we want to go east to U-Village? Again, no problem. What if we never run an east-west line — what if we have a local recession and simply run out of money. The money we would have spent on that extra capability would have been minimal. Far less, and way less disruptive (and likely far less convenient) than when we try to shoehorn the thing later.

        The Ballard station has the same issue. 24th NW is a fine stop — better than most of the stations in our system — and it makes sense to go that way. Like the UW, it has potential as a branch, a three, or a four way intersection. It makes sense to incorporate that in the design, instead of trying to figure out how to build it later without disrupting everything.

        It is obvious we don’t plan for the future, but Vancouver does (even if they build things in the wrong order).

      12. If ST thinks we need a 2nd tunnel because Lynnwood link is going to max out the tunnel capacity, why would they entertain the idea of a Ballard spur – any decent service allocation to the Spur would significantly reduce frequency north of UW (given it’s zero sum once the tunnel is tapped out), and that would be unacceptable, both operationally and politically.

        A Ballard UW spur coming from Ballard could work, given that tunnel does have spare capacity. RV becomes the limiting factor, but that can be solved separately (another spur? Push excess frequency into ballard- east link? Etc.)

        Any criticism of the Ballard station is premature – the sketches so far are for illustrative purpose only. Plenty of time on advocate for a Spurs there.

      13. Assuming you do have a capacity problem, you wouldn’t have a capacity problem all day long. So building something like this: https://www.flickr.com/photos/67869267@N07/9152772373/in/photostream/ would make sense (the description explains the concept in more detail). It would have made more sense to build that first with a Ballard to UW line.

        But they built things out of order (Hey — we do have something in common with Vancouver!). While it would be nice to have that spur junction for all of those people who are headed that way (Ballard to the hospital or Capitol Hill, Wallingford to downtown, etc.), what is essential is an easy way to transfer between the two trains at the U-District. That hasn’t been considered, even though they haven’t even opened the station (and it is obvious we will want it at some point).

      14. Two points:
        Can we expect the people to start spending more time where they work and live? Assuming we have people shifting from the suburbs into the downtown, are we building the right pieces to live there? PSRC is focused on transportation, if it focused on the urban village and keeping people’s use of transit shorter would be have a better picture?

        Sound Transit is missing their boundary options and is doing a poor job of using buses to feed the central corridor. Using rail we are beginning to see 2 ways of getting between I5 and I405 (Lynnwood and I90) which feels too far/too long apart. Getting rail from Redmond to UW to Ballard (underground and dedicated corridor) would make use of a lot of the 520 environment.

  3. Given that here in the Seattle region we are talking about lines 25 years out, I’ve never heard what comes after the Broadway subway, Surrey LRT lines and maybe Arbutus LRT (sounding pretty Save Our Trail-ish lately), anyone know?

    Extend Skytrain east from Waterfront Station in the Hastings corridor? North Shore rail transit line? SFU link? Further Skytrain extensions in Surrey? Subway to West End neighborhood?

    1. Good question. The Broadway/UBC connection is the big one. After that you could make the argument that you are done, other than improving surface buses. I would make that argument for Seattle if we had built the WSTT, the Metro 8 and the Ballard to UW subway.

      But it is still worth thinking about options. Looking at a density map, nothing looks urgent: http://blogs.ubc.ca/maps/files/2013/07/2011_Population_Density1.jpg. In fact, unlike Seattle, it looks like they pretty much covered everything (or will cover everything once they get to UBC). I would look into adding some stops for the West Coast Express, the commuter rail that goes to Coquitlam, especially now that they have a subway there. Maybe a stop at Main, as well as Nanaimo (McGill) which would keep the express nature of the line, but add a couple other options.

      Another possibility would be to to run an east-west line on 41st, starting at Kerrisdale. I don’t see the point of going past the other line, though (at Joyce or Patterson) so it would effectively be a split, even though I doubt it would function like one. You might also go on 49th/Imperial instead, which means a split farther away from the city center and a more likely possibility of going elevated for much of the way. As long as it curved around to 41st at the end (on East or West Boulevard) it looks like a winner. That would not only serve a dense area, but make for some more connections outside the core of the city (e. g. Surrey to Richmond).

      Speaking of East and West Boulevard, there is an old rail line there. I don’t know if that could be leveraged for a rail line (elevated I assume) and then another tunnel to downtown. That is probably excessive, but it would mean a fast one seat ride from Kerrisdale to downtown, as well as lots of good connections.

      1. Absolutely the UBC connector is a huge issue. I attempted to get out there a few months back, and even though bus route 99 is running articulated buses every 2 minutes I had to wait for two to go by before I could get on one, and even that was a crush load that require I get on the back door, tap the compass card at the rear door, and keep getting off and back on for the next 20 minutes to let other passengers out.

        After the last two days of power outages and service disruptions, there is no way I can compliment TriMet’s implementation of light rail, but it sure beats the experience of dealing with buses on a route that should have had higher capacity transit 10 years ago.

      2. Yeah, but I would rather have that problem, since the solution is obvious (build what you planned on building in the first place). It is similar to light rail in Boston — yeah, sure, maybe it would have been better to build heavy rail, but at least you are getting your money’s worth. I would much rather have that than what so many cities put up with, which is trains that suck money because they are empty most of the day.

        The takeaway, of course, is that transit works best in an urban environment. This is true everywhere. You simply have more riders. So while the Vancouver system is much better than its Northwest neighbors by every conceivable measure (rail ridership, transit ridership per capita, fare box recovery, etc.) they failed to learn from their own example.

        To a certain extent, they are victims of their own success. They did everything else right. The bus is reasonably fast and connects well with the subway. They didn’t realize that it was going to be so popular. They kept improving it (running buses every two minutes) and the numbers kept growing (because it was running every two minutes). With the discounted transit passes for the students, it doesn’t shock me at all that it is so popular.

        What is bothersome is that a line that carries over 50,000 people a day is put off while they focus on one that carries 10,000. I realize that the new line will be faster, and that there are other improvements over the current system, but that is certainly true of a bus that carries 50,000, while literally leaving people at the curb! Holy smoke, I wonder how many people just walk. As mentioned above, the problem is politics — suburban versus urban — and not what is cost effective (or even what might be more useful for those suburban riders, many of whom are in the same boat you were in).

    2. I’m not sure whether I’ve actually seen this or if I’m just imagining it, but I’m thinking the future Arbutus line would be LRT instead of elevated skytrain? Presumably the line would go into downtown Vancouver, perhaps via the Burrard Street bridge (unless they decide to create a subway for the portion downtown)? I would think it makes sense once the train is downtown to turn it east on (under?) Hastings and extend the line east to at least the PNE. Once outside of downtown Hastings becomes wide enough for a surface or elevated structure.

      1. I was wondering if an Arbutus line would pick up that 2009 LRT line between Granville Island and Science World/Main Street that must have the distinction as the short-lived LRT line in the world.

        Seems to me also, a short extension of the Expo Line from Waterfront Station to Gastown would be fairly easy and cheap as it could stay above ground in the railyard.

      2. OK, yeah, that makes sense. I just realized that the old railway I mentioned is the Arbutus line (duh). Based on what I can tell, you are right. It is a quiet path, and they probably won’t run an elevated line above it. There are intersections along there, which means that this would be unlike the rest of the system (it wouldn’t be automated). It would essentially be a streetcar, but with large streetcars (presumably). Density looks good along there, and while the addition to the rail grid isn’t great (it doesn’t connect that much) it still looks like a worthy project. A streetcar wouldn’t have to deal with traffic, so with a little bit of signal priority, it could carry a fair number of people.

  4. So why was the tunnel controversial? Cost? What was the alternative?

    In general, this looks like a dubious extension, but someone else (who knows more) can correct me (or at least help me with the trade-offs). So from I can tell, this looks a lot like the 97-B Line, one of a handful of special buses in the area. That bus carries 10,000 a day, or nothing special. It is hard to see how demand can justify the switch over to rail (unlike the 99-B Line, which will eventually be replaced by a rail extension on the other end). Is the bus really slow? Couldn’t they widen the road, or otherwise make a cheaper set of improvements?

    Even with the extension, I’m not sure why they went farther than Coquitlam Central. That would have made a good terminus, from what I can tell. That would provide good connections to the commuter rail, as well as bus service. I’m not sure why they extended it out to Lincoln and Douglas (other than the fact that the stations are easy for Americans to remember).

    1. The tunnel was controversial due to a dispute over safe work practices and numerous sink-holes that formed during construction in difficult soil (similar glacial till to what’s seen in much of Seattle). There was also some early controversy over alignment and environmental concerns.

      While existing bus service on the 97 B-line is unimpressive, there’s several more core bus routes will be significantly relieved by the Evergreen Line including the 143, 160, and 169, in addition to providing a frequent alternative to the at capacity West Coast Express. Each of these bus routes also had to balance local accessibility against their role as a regional connector, thus the 97 B-lines, detour through Port Moody.

      Furthermore, this line is well placed for inducing demand. Coquitlam, Pitt Meadows, and Maple Ridge had been difficult to reach from the rest of the region, with the above variety of either infrequent or indirect buses providing service. The new line provides a rapid connection to the entire north-east part of the region. Any additional bus service, even with widened roads, wouldn’t have been able to provide the combination of speed, frequency, and local accessibility that the new line has.

      While Coquitlam Central is a wonderfully positioned intermodal connection, it has an attrocious pedestrian environment. The short extension to Guildford Way brings the line within the urban environment of Coquitlam, next to a college campus and surrounded by residences. It also provides an easier connect to local buses in the northern part of Coquitlam, without having to cross the congested Lougheed Highway.

      1. I was thinking out loud that the tunnel could swoop further to the west under SFU and have a deep station Washington Park MAX style.

    2. I recently rode WCE to Coquitlam and then took the 97-B bus to Lougheed which ran via LaFarge Lake/Douglas. Yes the bus is slower (or was at it has been eliminated effective on the opening of the Evergreen Line) and it had to traverse the steep hill that the SkyTrain now tunnels under. The Towers that have been built already along this line will be joined by others now that the line is indeed open leading to a surge in population and ridership. There is no doubt in my mind that the 97-B bus would have soon maxed out operational capacity much as the 99-B certainly has.

    3. This line is expected to have 70,000 boardings per day in five years, so this will be a useful part of the transit system. Already around 45,000 different people have tried the system out as paying customers, and naturally some of those will have used it many times. And at around 1.3b CAD (1b USD) this isn’t insanely expensive.

      The tunnel wasn’t actually “controversial”. There were a few geotechnical problems with sinkholes, but in the end they didn’t amount to much. I suppose there was “controversy” about whether a single bore is superior to twin bores, but this was among the engineers. (Suspect that they will go back to twin bores.)

      The Broadway Line will be a much more important part of the system, but it will also be much more expensive. Could be 3b CAD which would be the most expensive transit project to date. Total bus ridership is over 100,000 per day on Broadway, the 99 isn’t the only bus, so a Broadway Line to UBC would probably have 150,000 boardings on the first day and the city predicts more.

      1. Yes. Let’s not forget that suburbs in Vancouver aren’t just a bunch of parking lots. Burnaby has a skyline that rivals many American downtowns.

        It would be interesting to see what happens to the Rainier Valley if height limits went away.

  5. Zach, for both Vancouver BC and Seattle, urban-suburban divide is now somebody’s forty-year old’s grandfather’s transit district. Vancouver’s and ours, regional transit and regions, we’re both barely getting started.

    My call is that as more people arrive, through highways, airports, and delivery-room doors, whether surrounded by woodlands and fields or (probably cracked) parking lots, coming generations now boarding trains in strollers will look at transit as freedom and prestige.

    Best if justice is in there too for a change. If present ill-conceived wage trends continue, income-related injustice will simply do a living-pattern 180. As has already started everywhere in the developed world, suburbs and sprawled development will be for the poor.Transit may be gateless, but passengers’ home communities won’t.

    Could be good PhD material, Public Administration or History (illustrations can be everything from etchings to videos, much cooler than graphs and algorithms) to compare our transit development with Vancouver’s since early-mid 1980’s ’til now.

    Both cities needed a Downtown subway with suburban connections. Vancouver lucked into an old rail tunnel under their CBD, vented for steam locomotives, so tall enough to hold a vertically stacked tube in each direction. And miles of railroad property wide enough for both freight tracks and SkyTrain pillars.

    And a Federal Government eager to pay for miles of beyond state-of-the-art rail transit as an exhibit for a massive 1986 trade fair.

    We lucked into the need to get started building, step-by-step, the system we couldn’t build whole. And the political leadership to approve it, the world’s top rail engineers to design an unprecedented two-mode subway, and the crew to build and run it.

    Two systems, two comparable cities, started roughly same time, 33 years ago. In 2049, we’ll be neighborhoods in the same city. Also quarrels about how bad the northwest transcontinental mag-lev vacuum tube line leaves us under-served.


  6. Are those long dwell times a result of driverless operation, or of scheduling requirements?

    And can we get at least one end of each new Link LRV to have no driver compartment?

    1. Just a guess, Brent. Could be that dwell time is set to keep trains at same spacing from each other. Which could also include calculation for boarding time, varying according to time of day, or other considerations.

      Portland MAX SR70’s each have a driver’s cab at one end, and a kind of parlor with a u-shaped bench at the other end. Each car being operated as part of a coupled set. Each car could operate individually if every line ended in a loop. As many streetcar lines used to do.

      Definitely more efficient arrival and departure, and taking less switching equipment than setup at the end of our lines. But requiring either land or looped tunnel or elevated at every terminal. For passenger loads like for LINK and MAX, minimum two-car “consist” makes the most sense.

      Though no reason individual single-cab car couldn’t work here, if any line had ended in a loop. I think the real question on most people’s minds here is why we need drivers at all, let alone seats and controls.

      Honest answer is that on any line that no vehicle can drive across, and no person can walk across without jumping or climbing that can’t be inadvertent, we don’t. Everett to Tacoma, absolutely no reason not to have walk-through trains, though.

      With a single driver carrying however many carloads will fit along a platform. Given the natural growth of this part of the country, no reason why every additional line we build can’t be automated. Also, when bigger and faster regional lines are built, the LINK LRV network will probably become like comparison of SR99 and I-5 as highways.


  7. You mention that the line is “suburban”, but Vancouver’s version of “suburban” is not like that of other North American cities. Yes, there are issues with “tower in a park” or “tower in a parking lot”, but the areas are urbanizing further. On one end of the new extension, the area around Lougheed station already contains many highrises, and is being redeveloped into this: http://www.thecityoflougheed.com/ On the other end, the area around downtown Coquitlam is already a dense urban area with many high rises, and will only densify further.

    The next Skytrain project will be an extension of the Millennium Line (which was just extended in the other direction with the Evergreen Extension) from VCC-Clark to Arbutus St. along Broadway with 6km of subway with 6 stations. This will connect it directly with the Canada line. The route already contains the busiest bus route in North America, and contains Vancouver City Hall, Vancouver General Hospital, and is generally becoming an extension of downtown in Vancouver.

    See http://www.translink.ca/en/Plans-and-Projects/Rapid-Transit-Projects/Broadway-SkyTrain-Extension.aspx for map and more details.

    After that, it looks like the next Skytrain or LRT projects will be south of the Fraser River in Surrey. An extension of the Expo line to Langley, or LRT in an L shape in Surrey. See http://www.translink.ca/en/Plans-and-Projects/Rapid-Transit-Projects/Surrey-Newton-Guildford-LRT.aspx – note that these plans are not up to date, and the technology (Skytrain or at-grade LRT) has not been decided.

    For more information on Vancouver transit projects, I read http://forum.skyscraperpage.com/forumdisplay.php?f=166

    1. The Coquitlam SkyTrain station has been built with a junction so that it will be easy to have a branch leave it to follow the CP Rail ROW which is presently used by West Coast Express. Vancouver has much experience in building along existing rail rights of way, from the aforementioned use of the downtown tunnel to the more recent use of the BNSF ROW for portions of the Millennium Line.

  8. >>The extension covers 6.8 miles in 15 minutes, roughly the same average speed as Link (27 mph).”

    Doesnt LINK average about 24 MPH. Not that thats bad. NYC subway averages about 16-17 MPH.

    1. Averags speed seems misleading to me because it’s not the trains’ actual speed. We don’t say cars traveled 16 mph when they actually traveled 35 or 55 when they weren’t at stoplights. Stops are just part of the overhead and the purpose of the line; they don’t mean the train “traveled slower”.

      1. What is “actual speed”? Do you mean top speed? Average speed by definition is more representative of the travel experience. A train may travel at 55 in some parts but that isn’t true of the entire line. If one were led to believe that “oh, the train’s top speed is 55 mph, that means Everett to Tacoma on Link would be an hour ride.” That is not true.

        We don’t say cars traveled 16 mph

        We absolutely should say that! Just like when you fly and after you consider the time it takes to get to and spend at the airport before you even take off, it might take as much time as driving or taking the train. Or when you drive and waste time circling for parking or stuck in traffic trying to get to the freeway.

      2. Yeah, what Oran said. Actually, for a subway line like this, I don’t think it matters, and is a bit misleading. New York is relatively slow because it is making lots of stop (adding value).

        Where you really want to figure out average speed is with a non-grade separated transit line. For example, the 44 averages less than 8 MPH. I think San Fransisco’s buses and trains (the non-BART ones) average around 7 MPH, which is ridiculously slow, and thus worthy of a major investment (especially since they carry so many people).

        How slow are our express buses (on the freeway)? What about HOT? How does that compare to, say, the 44 or 8 I mentioned?

        That is really the way to measure it, not what is the max speed.

      3. It is a trade-off between speed and access. The higher the speed of the transit line, the less access can be provided along it because of physics. Speed is mostly a function of the distance between stops and the quality of ROW.

        Three of the busiest bus corridors on the West Coast come to mind: Broadway in Vancouver, Geary in SF, and Wilshire in LA. All of these offer both local and limited-stop service at high frequencies and have hit the limit of what surface buses can do. They should’ve had rapid transit service already but for political reasons they don’t.

        The ST Express buses are pretty fast. In the PM, 44 UW to Ballard = 550 downtown to Bellevue TC in time but 2x the distance.

  9. I must say that your diagram of Lougheed Station makes much more sense than even the official TransLink diagram. Thank you for making it because it totally clarifies my confusion of how the platforms work and how the trains flow at that station

  10. Oran: Wilshire Boulevard has a short subway, which is being extended now, and will be further extended in phases. It sounds like the Broadway Skytrain is coming. Geary Blvd. is getting a BRT in a few years. Better late than never.

Comments are closed.