ST2 Link Operations Plan (Source: ST North Corridor AA Summary)

A tidbit that Martin tangentially covered last week is the the current operations plan for Link once ST2 is fully built out. The North Corridor Alternative Analysis includes a discussion of the current operations plan. It identifies two lines, both operating at 8-minute peak headways, for a combined headway  of 4-minutes on the common segment from the International District Station to Lynnwood.

Figure S-10 illustrates the planned light rail system configuration once extensions are completed east to Overlake in Redmond, south to South 200th Street in SeaTac, and north to Lynnwood. As can be seen in Figure S-10, the system will operate with two lines, one from Lynnwood to South 200th Street and one from Lynnwood to Overlake. Both lines will operate at 8-minute peak-period headways resulting in 4-minute peak headways between the junction at the south end of the Seattle CBD and Lynnwood, and requiring every train operating in the system to traverse the segment between Northgate and Lynnwood. Ridership forecasting indicates that this level of service, at least south of Lynnwood, will be needed to accommodate forecasted demand in the future. As a result, any delays incurred in the segment between Northgate and Lynnwood will affect the operation of the entire light rail system. This problem becomes worse when the system is eventually built north to Everett, south to Tacoma, and east to downtown Redmond.

Previous planning documents from the Roads and Transit package assumed three 9-minute headway lines resulting in 3-minute headways on the common segment during peak periods, with two 10-minute headway lines during non-peak periods.

Line 1: Northgate to Tacoma Dome Station. 4-car trains running every 9 minutes peak, 10 minutes off-peak.

Line 2: 164th SW/Ash Way to Kent-Des Moines Road. 3-car trains running every 9 minutes peak only.

Line 3: 164th SW/Ash Way to Overlake Transit Center. 4-car trains running every 9 minutes peak, 10 minutes off- peak.

This change results in less frequent service in the overlapping segments by 1-minute, but more frequent service on the non-overlapping segments by 1-minute. At 4-minute headways additional service can be added if additional capacity is needed, with fewer impacts on reliability. This change also simplify operations for ST because trains go out of service only at the end terminals, not midway along the segment north and south of Seattle. This benefits users as well and eliminates confusion between peak and off-peak operating patters.

106 Replies to “ST2 Operations Plan”

  1. Has ST formally given up on making past S 200th St with ST2’s money?

    Don’t show that map to the City of Federal Way…

    1. No, but they are still in the process of south corridor project realignment so it looks like this discussion is premised on what ST knows will be built.

    2. the ST Board told staff in May to proceed with developing “the next light rail segment” south of 200th. This means building an additional segment and station at Kent-Des Moines Rd. Look for this project to surface in the 2012 budget.

  2. This is great news!
    This still leaves room for trains from Ballard to enter the tunnel at CPS every 4 minutes, and diverging to W.Seattle after SODO, bringing DSTT up to it’s designed maximum of 2 minute headways. That comes close to the 22,000 riders per direction in the peak often quoted in the press when Sound Transit was formed.

      1. Yea it would also be very expensive to build (bridges). I just dont think that ridership/cost will be worth it. I think BRT would be better for this corridor since you could implement it quicker and cheaper. Rapid Ride is a start, but it is not really BRT. Hopefully in the future we can make improvements to Rapid Ride and make it close to BRT. Then if the Demand grows and overwhelms the BRT, we could start talking rail.

      2. But why @zefwagner? Even a casual glance at the West Seattle Junction would tell you that it already outpaces the density of some neighborhoods that already have link service. Sorry, but even after $200m+ for a tunnel and station on Beacon Hill, the area around the triangle and junction are more dense than Beacon Hill is even willing to consider being in 20 years. I-5 & 145th? Pretty sure the Junction is denser. And, since you seem to be a fan of Human Transit please refer to this post on choke points.

        A mere two station extension of of link from SODO with stations @ delridge and near the Junction would allow for the truncation of all of these routes: B Line, 21, 21x, 22, 54, 54x,55,56,56x,57,116,118,119,120, & 125. The combined downtown-bound ridership of these routes is fairly high and would compete well with say the Rainier Valley, and even the rest of the system on a riders per mile basis.

        Yes, it would be challenging, but please don’t dismiss an entire quadrant of the city as not dense enough. It is at least as dense as neighborhoods that already have light rail service, and has been at least (and more) willing to density than other neighborhoods along ST2 alignments.

    1. It create an interesting possibility of trying to route a Ballard/Fremont/Eastlake/SLU rapid streetcar through the tunnel. Don’t know if it would even make sense but there now is some additional capacity in the DSTT.

      1. I feel like I’m the only person who is wholly unsatisfied with this “Rapid Streetcar” idea. The streetcar is comfortable and convenient for short trip pairs. It fits more people carrying more things. For longer trip pairs, either the bus, with its forward-facing seats, or grade-separated light rail, with its gentler curves and much more constant speed, works far better. Long-haul trips up Westlake Avenue don’t take advantages of any of the streetcar’s modal advantages.

      2. Kyle: The HCT long-range plan goes to lenghts to make the distinction between “local streetcar” (given example: SLU) and proposed “rapid streetcar” systems. It states that these corridors would have more aggressive signal priorities, higher speeds, more extensive use of exclusive ROW, and also that they would use different, larger vehicles.

      3. …and still get mired in traffic in Fremont, Ballard, and Downtown, where the traffic really matters. And still get stuck waiting for the Fremont bridge to open. And still deliver a slower, more stop-and-go ride than grade-separated light rail can offer.

        I’m all for a streetcar to Fremont. But it should be a local streetcar up Dexter.

    2. Mike, the schedule variability from Rainier Valley and Bel-Red make that impossible.

      And even if it were possible, it’s a TERRIBLE idea. As demand increases, we’re going to want to run the two lines at 7 minute, and eventually 6 minute headways.

      We really do need another tunnel downtown. I’m sorry to have to sound like a broken record, but that’s reality if we want more grade separated service downtown!

      1. But we are digging another tunnel downtown… oh wait, it’s not for the trains or buses…. And the state is short $2Billion dollars which happens to match the funds being spent on the tunnel (so far)

      2. The only TERRIBLE idea here Ben is your fascination with Northgate ever coming close to needing more than 4 minute headways with 4 car trains.
        Look closely at reality in the charts provided by Bruce
        https://seattletransitblog.com/2011/09/30/ridership-patterns-on-central-link/
        Even during the peak, average car loading is only 50 riders per car on 2 car trains, at 8 minute headways. That’s 25,000 riders per day.
        Now double the train to 4 cars.
        and double the frequency to 4 minutes. That’s 100,000 riders per day.
        Now triple the car loading from 50 to 150.
        You can do the math from here, but your pipe dream of needing double that number someday is a pipe dream.
        Put the pipe down, and nobody has to get hurt!

  3. I was reading somewhere that alternatively, ST was considering two lines with 6 minute headways and 3-car trains. This, IMHO, is a much better plan because it provides the same capacity, but forces people to wait 25% less time. My experience in other big cities is that subway lines often have sub-5 min headways, even when there are multiple lines on the same track. Be should be targeting a similar level of service. So ideally, we could do 4 minute headways with 2-car trains, but I don’t know if ST has the money to do that.

    1. It costs more money. When the economy gets better and ST has some to spare, it’s a good idea. For now, it’s 25-33% more expensive to operate (at peak times).

      1. But decreasing the average time that people spend will provide a stronger incentive to ride transit, which will increase ridership, something ST should be strongly invested in. I hope that ST can afford that by the time the the east link is built.

      2. Unfortunately decisions like this aren’t motivated by good policy like encouraging transit ridership nearly as much as they are motivated by “what can we possibly fund without attracting the ire of Tim Eyman’s selfish cronies.”

        Unlike other states, which allocate funding based on leadership, we allocate funding based on fear.

  4. How wiil people transfer from East Link to Airport Link easily? Right now to switch directions at the ID station means going up and down a bunch of stairs. Maybe they could add a center platform.

    1. The center platform would be neat. They would have to add new stairs and ecalators/elevator to it. But it could help trasfers. Once East link opens i expect the tunnel to become Train only. So the additional center lanes in each Station become pointless.

    2. In systems that I’ve ridden (Tokyo, London, Taipei, Paris, etc.) I’ve always had to go up the stairs to transfer. But I agree, a center platform would be better. They should have done this during the tunnel retrofit.

      1. This is an interesting case because your frequent trasfer will be to a line going in the opposite direction. I have not seen too many lines like this elsewhere.

        Many Tokyo lines do use center platforms for frequent transfers, but those trains are usualy heading in the same direciton (like express train to local train trasfer).

      2. See Hong Kong’s MTR. They designed the system for timed transfers between trains on different lines going in different directions.

        Also see BART on the East Bay/Oakland side.

      3. @Murray

        A perfect example is the Stockholm Metro at Slussen, Gamla Stan and T-Central. At Slussen and Gamla Stan you have a cross platform transfer between the Green and Red lines in the same direction. Most people transfer at Slussen though because it is a covered station a less windy. Between Gamla Stan and T-Central the lines become “stacked” so that a SB red line train stops across the platform from a green line NB train.

        This was intentionally done because anyone north of T-Central, which is almost all of the north central city, has to make this transfer if they want to get somewhere else north of downtown on the subway. I would do this when I went from Tekniska Hogskolan to St. Eriksplan for example.

    3. A center platform wouldn’t meet ADA requirements without an elevator. If there’s room, maybe!

      1. Why would a center platform require elevators when wheelchair access would still be available via the Mezzanine?

      2. Ryan, if you get off a train, it leaves, and then Link stops operating… you need a way to get off the platform.

      3. I wonder if IDS could be modified so the tracks could run where the platforms are now, resulting in a very large center platform with elevators in an appropriate location.

      4. Ben,

        The solution to stranding wheelchair riders is trivial: The last train back to base wouldn’t leave IDS until the center platform is clear.

      5. Morgan: Unlikely, because it would be difficult to fit four-car trains at a tangent platform and still accomplish the track shift necessary to get into the tunnel bores at the north end. Yes, it’s possible to build platforms on curves, but the practice is avoided on new design projects.

        Brent: What about emergency responder access to the center platform — isn’t an elevator necessary to get a stretcher down there? I really don’t think someone who just had a heart attack on the platform is going to not file a lawsuit against ST because they had to ride a train to the next station merely so the EMTs could load them on a stretcher and whisk them off to the hospital.

      6. Guys if they go through the effort of building a center platform adding an elevator will be obvious.

      7. DWHonan,

        I don’t think we’re talking about the same kind of center platform. The suggestion is to leave ID Station pretty much as is, and build a simple, no-frills platform in place of the gap between the trackways, once the buses are kicked out.

        The only purpose of the center platform would be transferring between East Link and South Link. The most time anyone should spend on the center platform is the time it takes to exit one train, walk a few feet, and enter the other.

        There should be a simple plan for emergency access, perhaps a really strong plywood board that is stored nearby and can be laid across the tracks, or perhaps a couple retractable planks.

      8. And by that I mean, “The most time anyone should spend on the center platform is the time it takes to exit one train, walk a few feet, and enter the other” only if…

    4. I don’t see myself ever transfering from East Link to Airport Link. The ride is WAY too long. If I recall correctly, the ride from Seattle to SeaTac is something like 37 minutes? If the ride from Bellevue to Seattle is 30 minutes, you’re looking at over an hour to get to the airport. I usually get to the airport in about 45 minutes on the 560 bus, with no transfers.

      I might be wrong on the numbers by a few minutes but that doesn’t affect my main point: East Link to the Airport offers no benefit over a bus (or taxi!) unless you particularly enjoy train rides.

      1. The 10-minute headways mean you can leave when you’re ready rather than when the bus comes, and if you miss the train you’ll only be ten minutes late and can still catch your flight. The train is significantly faster if you’re coming from anywhere other than Bellevue TC: Mercer Island, Bel-Red, Redmond, and Rainier stations. Bellevue-SeaTac is not a primary market for Link, it’s just an alternative the network makes available. Lynnwood, in contrast, has no direct bus to the airport and is quite excited about Link taking “only” an hour to get to SeaTac. If you get off a plane at 7:30pm and can choose either to wait 10 minutes for Link or 60 minutes for the 360, the longer seat on the train looks better than the cold windy seat at the bus stop.

      2. Also, don’t underestimate the appeal of one simple map and schedule. If people are visitors, or already use Link for other trips, it’s nice to have one simple map and schedule — the one they already know. Especially if they never have to wait more than ten minutes (15 after 10pm).

      3. AP:

        – The ride from Seattle to Bellevue is supposed to be about 22 minutes.
        – The ride from ID to Sea-Tac is about 31 minutes. 37 minutes is from Westlake. And that’s going to be tightened up in the next decade.

        Average transfer would put you under an hour, total. That’s a little slower, but it’s so much more reliable that it beats the 560.

        You’re forgetting to price risk, which is part of the reason rail gets far higher ridership than buses, and part of why people erroneously think a cheap bus option is just as good!

        Also, the 560 has nothing like the headways you want, which means you end up with an inflexible schedule, adding time.

        Link will end up taking most 560 airport ridership.

      4. Speaking of Lynnwood, you can get creative now as I mentioned in a previous post. If your plane comes in anywhere from about 1:00ish to 5:00ish, you could take Link north to the ID and transfer over to Sounder or Amtrak for one of the 5 trains scheduled from 4:05pm to 6:50pm.

        Granted, you would have to get off at either Edmonds, Mukilteo or Everett on Sounder or Edmonds or Everett on Amtrak, but then you could have someone pick you up which is a lot easier than taking the bus up there or having someone drive all the way down to pick you up and drive back.

      5. Yes, but that bus is hopelessly mired in traffic jams all day. Much better to have a much longer train ride because it’s always reliable and very cool.

  5. I am am running late this morning, so I didn’t have time to write a few paragraphs on this thought I had, but I did write the first sentence of it. So please imagine I wrote a very insightful comment on the danger of putting too much weight in increased headway, at the expense of convenience.

    One common mistake transit planners (and transit fans) make is fetishising headway.

    (Please imagine a well-written paragraph or two here).

    Bye

    1. But increased headway = increased convenience, they aren’t at odds. I’m having trouble imagining your paragraph.

      1. I think I figured out the end of it, we just need the bit in the middle:

        Don’t fetishise headway.
        ???
        Profit.

      2. Thanks for the thought provoking statement. What’s the headway between trams on the the Mattmobiles.

      3. The gondola community has been calling headways LTM, for “less than a minute”, but it’s an acronym I dislike. For a high capacity system you’ll likely see cars leaving at least every 30 seconds.

        Oh, and to get us mildly back on topic: A Seattle – Bellevue gondola system would only take 15 minutes, and could have about the same capacity as an 8-minute headway Link.

        I honestly don’t mean to be anti-rail. I’m a rail fan. But I’m having trouble finding a problem gondolas can’t solve.

      4. I picture you as the ‘caped crusader’ of Mattmobiles ™, and I say that in a kind voice.
        If moving masses is the goal and not breaking the bank (oh, I forgot, it’s already broken), then I’d pay for 20 of your lines before I ever voted for another damn tunnel project in Seattle.

      5. Matt, the problem with gondolas is that they’re point-to-point. Yes, I know, there are a few weird systems with intermediate stops, but that’s ferociously complicated and expensive.

        Accordingly, gondolas are terrible if you want to serve a whole string of destinations. Rail, on the other hand, is great for that.

        If, on the other hand, you just want a single link to go to ONE station on the wrong side of difficult terrain, a gondola is fine. So, map out where you’d like to see stations, and if there’s one just sort of sitting out there by itself on the wrong side of a waterway or mountain, *that’s* your gondola-ready location. If on the other hand there’s a long string of places you need stations, build rail.

      6. [Nathanael] I think you’re trying to put gondolas into too tight of a box. Rail is point-to-point as well, and their stations are probably close to the cost of a gondola station (or much, much more if the station is underground). The machinery is cheap for an intermediate gondola station – most of the cost is in physically moving people around (platform area, elevators, etc.), and these requirements are roughly the same for rail or gondolas, varying mostly by station capacity.

        The difference is that rail costs a lot more between these stations. If underground you need boring machines or at least major earthwork, if elevated you need massive pre-cast concrete sections, and even if surface run you still need significant groundwork. For a gondola you need some towers and some wire. Sure, there’s earth work with the towers, but that’s limited to at most a dozen spots per mile.

        So cost wise, gondolas should beat any rail system for the same number of stations. That said, if you can get light rail headways that are more frequent than 8 minutes with 4 car trains, then gondolas probably can’t keep up in terms of capacity.

        [Morgan] I don’t think I fetishize gondolas at all. I just think others dismiss them too quickly (see [Nathanael]’s gondolas are only for mountains bias). It’s like walking around in a wet, rainy city where everyone wears socks without shoes because shoes are only for walking up hills. Am I fetishizing shoes if I imagine they can be used almost everywhere?

      7. To put a fine point on it: designing gondola stations is not rocket science. Check out this station. It could be any at-grade Link station, maybe with a roof over it and some gondola machinery. This is an end station – intermediate stations are even simpler.

      8. Matt,

        Whenever I advocate a certain idea, I try to ask myself, “what’s a situation in which this idea isn’t good?”.

        If I can think of something that would make my idea less than optimal, it’s a good sign that I’ve put a good amount of thought into it. Conversely, if my idea seems like the perfect solution to any problem, then I’ve probably missed something.

        As you yourself said:

        But I’m having trouble finding a problem gondolas can’t solve.

        You’re an engineer. You know that engineering is all about tradeoffs. So what are the tradeoffs associated with gondolas? If you can’t think of a single one, then forgive me if I question whether there’s something missing.

        As far as your socks/shoes example goes, I think that’s a misleading comparison. Most people agree that walking on pavement with bare feet or socks is no fun. Similarly, it’s clear that motorized transportation provides mobility that walking cannot. But there are a huge variety of shoe types and styles. Gondolas are no more the solution to every transportation problem than sneakers are the solution to every don’t-get-your-feet-wet problem.

      9. [Aleks] I was being mildly facetious with the “finding a problem” remark. Situations where gondolas probably aren’t the best solution:
        1. Corridors that deserve > 6,000 pphpd either now or in the foreseeable future. (should be light rail or metro)
        2. Corridors where frequent stops make sense for your fastest form of transit. (should be streetcars or buses)
        3. Medium or long distances (should be buses, commuter rail, or heavy rail)

        That said, there are a whole lot of urban instances where these three constraints don’t apply.

  6. Having light rail between the eastside and the northend would be great once ST figures out how the service is going to work across the floating section of the I-90 bridge.

    1. On other forums, we call that “concern trolling,” you know. It’s repeating an anti-light rail talking point and pretending you legitimately don’t know the answer, when you actually do.

      1. John, I want to say I know better, but it almost sounds like you’re accepting light rail isn’t so bad after all. :)

    2. Implying that adding an HOV lane in each direction doesn’t completely replace the existing, outdated, reversible lanes, or solve the current “reverse-peak” bottleneck.

      1. Formally, the Record of Decision for adding in the new HOV lanes in each direction on the I-90 bridge (R8A configuration going from 3 to 4 lanes in each direction of the outer roadways) is a stand-alone approved project. R8A is documented in the environmental record as valuable with the Center Roadway continuing in its present format.

        For example, the I-90 bridge operating as six road lanes (and no trains) in the peak direction, and four in the reverse direction, would improve the peak travel time for ST REx buses in the I-90 corridor.

        The two Federal Records of Decision for (1) East Link light rail, and (2) for the I-90 conversion from road to rails, are not yet issued, because (speculating here) the preliminary design for certain aspects of East Link is not complete.

      2. WSDOT has been of the opinion that the reversible lanes were a mistake for a while now. They don’t fit in with modern traffic patterns, and overcompensate for what has become an increasingly small difference in traffic volumes. And there’s been plans to use that section of ROW for rail since before the 2nd bridge was even built.

        I will freely speculate that the reason it’s taking forever to finalize the design is the Bellevue city council stonewalling, even though their voters already approved the concept years ago.

      3. Is that also true with the I-5 reversible lanes? I have lived in Seattle for 3 years and have literally not once used the I-5 reversible lanes. Personally, I think we should just add 1 lane to the express lanes and put a divider in the middle so it can be in both directions. Then we could run all of the busses in those lanes in both directions and that way we could have far more reliable bus service.

      4. it’s taking forever to finalize the design is the Bellevue city council stonewalling,

        The pontoon bridges are between the cities of Mercer Island and Seattle. How is the Bellevue City Council in any way even tangentially affecting the design of this segment? There’s been zero opposition to this section of the route from the Council. Kemper Freeman yes, but not the City Council and even his opposition should have no effect on design work. Really the only entity outside of ST that has anything at all to do with it is WSDOT which has provided it’s full support up to this point and is why the R8A project in on schedule.

      5. Bernie: I was speaking of the whole project, not just the bridge segment. Wasted time and manpower on non-issues in Bellevue is slowing the whole thing down.

        Stephen: This is especially true of the I-5 express lanes. They consider those are far worse than the I-90 lanes – I-5’s entire design from I-90 to Northgate was based on the flawed assumption that the primary users would permanently be mass commuters from North Seattle to Downtown. It was a old plan for a standalone toll highway that had been scrapped in the early 50’s, which was just dug up and integrated wholesale into the I-5 plan to save time and engineering money when the federal Interstate Highway system came into existence a few years later. The connection to a continuing highway south of Downtown (or north of the 522 interchange and Northgate) was just kind of shoehorned in.

        But anyway, the express lanes on I-5 are harder to make bidirectional than I-90. Simply splitting them down the middle is less than ideal, because there are exits and onramps from both sides, and in most places no room to build new ramps. Suddenly, the only northbound ramps are Mercer, the U-district, and Northgate. Everything else is southbound only. And at the very south end, I’m not sure the ROW is even wide enough for a lane in each direction.

      6. Bernie: I was speaking of the whole project, not just the bridge segment. Wasted time and manpower on non-issues in Bellevue is slowing the whole thing down.

        I call BS. Bellevue has presented probably the best TOD project of the entire Link Light Rail project in Bel-Red. The City has not only revised zoning but committed mega bucks to the infrastructure. Compare what is proposed in the Spring District to what Redmond is only now looking at at the Group Health site.

      7. Bernie: what about B7? What about the money that could have been spent on speedy construction and on mitigation amenities? East Link is delayed a year because of the council’s obstruction.

        As for the I-5 express lanes, they don’t connect to 520. That’s good for Northgate-downtown drivers but bad for everybody else, who has to go through the interchange bottleneck no matter whether they’re going to the Eastside or not.

  7. Is the link to the current alternatives document missing in the post? It’s really hard to scan for hyperlinks when they almost look black and have no underline!

    I’d suggest this to add to the css: .storycontent a { border-bottom: 1px dotted #000 }

    1. The Alternatives Analysis is here. I looked through a few chapters and did not see the headways. What are the off-peak headways projected to be?

  8. Dear god.

    4-minute headways to Lynnwood. WHY? What a monumental waste of service hours!!

    For the umpteenth time, Seattle Ether, the length of a trip is inversely proportional to the need for spontaneity.

    You’re going far; you’re going to be gone for a long time. The difference between waiting 4 minutes and waiting 8 minutes for a long-distance trip is negligible.

    Everywhere else on earth gets this. I dare you to find a commuter rail, an S-Bahn, or a “regional” metro anywhere on this planet whose frequencies are anywhere near its respective urban-transit counterpoint.

    Even BART, which hemorrhages money trying to perform a “metro” service at commuter-rail distances, has no headways closer than 15 minutes outside of urban SF/Oakland.

    Is our goal to be dumber than BART?

      1. Ok, my bad. The Richmond-Fremont line gives that whole corridor 7.5-minute headways.

        …combined headway of 4-minutes…to Lynnwood.

        Dumber than BART.

    1. d.p., I know you don’t have the whole story, but if you go off on ideas before you understand them, all you’re going to do is have a hard time here.

      We do need 4 minute service as far as Northgate. At that point, it makes sense to keep running the trains farther north, because it gives ST and CT a great opportunity – terminate buses headed to downtown instead of pushing them into I-5 traffic. There’s a significant cost savings, and it increases service reliability for those users.

      I really feel like you latch onto one little thing over and over and don’t pay attention to its impact on other potential choices. You’re just getting angry for no reason.

      1. At that point, it makes sense to keep running the trains farther north, because it gives ST and CT a great opportunity – terminate buses headed to downtown instead of pushing them into I-5 traffic.

        4 minute headways seems out of scale with the amount of demand that ST Express and CT can deliver to the corridor. Sure, we’ll have the capacity to deliver 4 minute headways to Lynnwood, and that might even make sense during the morning rush. But round-the-clock, I just don’t see how we can funnel enough demand to North Link to justify 4 minute headways at Lynnwood.

      2. Read my other post. You’re providing New York levels of capacity to Northgate. Unnecessarily. I think it’s time for you to stop claiming to be pushing “informed” math.

      3. …Or you could try to engage my challenge:

        Find a commuter rail, an S-Bahn, or a long-distance-service-oriented quasi-metro offering those kinds of headways. Anywhere. (Good luck.)

      4. Due to Seattle’s geography, Lynnwood is “closer” than the same distance in other cities. Also, it makes the service more legible and administration simpler if trains aren’t terminating in the middle of the line.

      5. While terminating buses in the 2020s that are no longer needed because of light rail is part of the expectation for the I-5 corridor north out of Seattle, I understand from hearsay lately that ST has no plans to encourage/facilitate bus-to-train transfers (AM) and the reverse (in PM) at Northgate light rail station. In the computer modeling, the trains get quite full enough without that intermodal connection.

        If anybody has evidence from the recent Central Link planning that I’m wrong about that, sound off, and I’ll be motivated to look at and report what the existing environmental record says about intentions for Northgate.

        It is also clear from recent PSRC documents that the regional government intent is to replace the Northgate Park & Ride with TOD-compatible housing.

      6. Lynnwood is “closer” than the same distance in other cities.

        No it’s not. 15 miles is 15 miles.

        (Your point might be salient in Los Angeles, or Houston. But Lynnwood is distant and knows it.)

        There are two arguments in play here:

        1. Ultra-frequency = spontaneity.

        I agree. But different frequencies suffice for this at different distances. At 3-5 miles, <5 minutes is ideal. At 15 miles, 10 minutes is fine. At 100 miles, hourly will do. And so forth.

        2. Northgate "needs" the capacity.

        If the 15,000 in 2030 estimate is correct, then no, it just doesn't!

      7. Lynnwood is closer in the sense that Seattle is hemmed in by Puget Sound and Lake Washington, so it has to expand further north-south than most other cities do. So rapid transit has to traverse a longer distance than it does in rectangular cities. Likewise, Bellevue is distant only because of Lake Washington, so rapid transit is necessary to cover the gap. You can argue that Seattle should be denser, and then Lynnwood and Bellevue wouldn’t be necessary, but that’s a land use issue, not a transit issue. The reality is that the single-family houses in Seattle aren’t going away any time soon, and we can’t just do nothing until they do.

    2. This is peak headways. And it says, “Ridership forecasting indicates that this level of service, at least south of Lynnwood, will be needed to accommodate forecasted demand in the future.” Lynnwood is analagous to Berkeley in our geography, not Pittsburgh. Pittsburgh would be an extension to Everett. Lynnwood is obviously not as big as Berkeley because there’s no major university there, but still there will be some people going to Lynnwood and other people going through Lynnwood on their way to elsewhere (Everett, Mukilteo, etc). BART has 10-car trains, Link has 4-car trains, so it’s not like they’re overbuilding.

      1. Absolutely. There are 1.5 million more people forecasted to arrive around these parts. We are not going to to significantly expand our freeway system in the next 30 years, and the price of gas is going to continue to rise forcing people to seek alternatives. This system will be essential to managing further congestion on the existing road systems including buses in that congestion. It will be seen as an attractive alternative to getting to major employment, commerce, and entertainment centers in the region given the cost of travel we’re going to see in the future.

      2. If you want analogies, look at SkyTrain.

        Northgate is nearly as far as New Westminster; Lynnwood is Surrey. (But with far fewer stops/far less stuff along the way!)

        Vancouver made this trade-off. They used higher frequency (automatic train control) to enable shorter trains. Their trains, at 3-minute frequencies, have roughly equivalent capacity to a 4-car Link at 6-8 minute frequency…

        And guess what? SkyTrain carries an order of magnitude more people than in Northgate’s wildest dreams!

        So sorry. The “ridership forecasting” cited by Adam and Ben does not convince me.

      3. p.s. Forgot to remind that BART’s mean per-rider subsidy is a whopping $6.17.

        There’s nothing in North Link’s ridership estimates to suggest we’re not overbuilding/overservicing, and plenty to suggest that we are.

      4. True, but they make up for it in volume. That’s why I don’t worry about Links current subsidy of $13 per rider.

      5. Huh?

        That’s $6.17 per trip. They’re not making anything up in volume. Their current 350,000-ish ridership (about their peak) is losing them $6.17 x 350,000 = $2,159,500 every single day!

        Link’s subsidy will definitely decrease from its astronomical $13/rider as it extends and its ridership increases. But if Link mirrors BART in everything it builds and the way it distributes service, don’t expect us ever to drop below $6. Which is still a money-hemorrhaging indicator of a poorly-conceived system!

      6. Metro’s subsidy is around $6 too. That’s what 25-33% farebox recovery means. Are you complaining about Metro’s subsidy too? Do you think enough people would use the transit system if tickets cost $7.50-10 per trip?

      7. Metro is grossly inefficient, as we all know. It’s nation’s-highest fares yet still-crappy farebox recovery rate reflects that.

        They’re also running buses, which are inherently less efficient to operate.

        In addition to speed, one of the major selling points of finally building real mass-transit is that economies of scale can push down the subsidy rate.

        If you do it right.

        So, no, $6/rider is not an acceptable target.

      8. Metro is grossly inefficient, as we all know. Its nation’s-highest fares, yet still-crappy farebox recovery rate, reflects that.

        It’s also running buses, which are inherently less efficient to operate.

        In addition to speed, one of the major selling points of finally building real mass transit is that economies of scale can push down the subsidy rate.

        If you do it right.

        So, no, $6/rider is not an acceptable target.

      9. Metro’s subsidy is around $6 too.

        That’s not correct. From Metro’s Facts page:

        Bus operating cost per boarding: $3.92
        Bus operations revenue per boarding: $1.05

        Now, $3.92 is still really high. But the poor recovery is mostly because of discounted fares, monthly passes, transfers, and fare evasion.

      1. Dennis, I didn’t see this reply until just now. Thanks for engaging the substantive question!

        Peak headways on MTR’s urban lines (Island Line, Tsuen Wan Line, etc.) are 105 seconds.

        As I said, long-distance lines never justify the frequency of their respective urban counterparts — even when the demand for both is very high!

  9. Why is it I have this nagging suspicion that the minute they start work on the I-90 bridge they will find some spec that is not adequate for trains?

    We’re re-building a new 520 bridge — this seems like the ideal time to add direct light rail along the 520 corridor to Redmond.

  10. What’s ST’s backup plan in case Eyman gets his way and convices enough people out in Ellensburg and Yakima to shut down the whole project? Even if Eyman loses this time, what’s to stop him from trying again?

    If, a few years from now, ST is halfway through building East Link and then Eyman passes an initiative to shut down rail on I-90, what happens then? Do we just have to stop building and accept the loss for the billions of dollars already spent?

    (I know it sucks to think that people who don’t live here and aren’t paying for it get veto power, but it seems we have to have some backup plan just in case).

    1. First off, understand that Tim Eyman craftily designs his measures to either fail or be rejected by the courts (that’s the best outcome for him). If they pass and are enacted he’s out of business. The former watch salesman is now in the business of selling initiatives. If an issue passes and is enacted he’s out of work. Second, once the go ahead for converting the center lanes to rail is adopted and work begins a Washington State Initiative can not stop it. Even though WA “owns” the bridge all that really means is we’re on the hook for maintaining it. The feds have total control over the Interstate Highway ROW. They say rail, we say yes sir. At stake is billions of dollars in federal highway funds.

    2. The backup plan for East Link is obviously BRT. There’s no reason to spend resources on it now when it’s just a theoretical possibility. Implementation is a simple matter of doubling the number of 550 buses and deciding what to do about Bellevue-Redmond. Or maybe they’d get three-door buses and copy Swift’s station design. BRT would be less frequent than Link (15 minutes) if it follows the North Corridor precedents.

      The thought of an initiative halting East Link in the middle of construction is too far-fetched to consider now. Eyman could just as easily cut the sales tax in half. We can’t plan for everything. Is ST supposed to build peel-away structures like temporary tape? What could the Monorail have done to plan for its own abolition?

      1. a simple matter of doubling the number of 550 buses

        My thoughts exactly with respect to the 2 or 3 Move Bellevue Forward folk that spoke at the last public hearing about how they’d ride transit more if it weren’t standing room only. Except of course they don’t actually use transit; you had to park up at the library and walk to the meeting because parking at City Hall was maxed out a half hour before the hearing started. Our intrepid HS student said it best, “it may not [won’t be] me using it but someone will”. Yeah, the old eastside transit logic is alive and well… “we need more transit so I can drive my car without being stuck in traffic.”

        In essence, the idea is if I vote for transit then I get to drive congestion free for a relatively low cost. Of course that’s a fallacy. Congestion pricing and/or HOT lanes (not really clear on the difference) is what gets that done. Of course it’s going to cost users a heck of a lot more than the sales tax ST collects since it’s collected over the relatively small percentage of people it affects instead of the shotgun ST taxing districts but there’s no free lunch.

    3. I don’t think the feds control it. The federal subsidy was just for the construction. Highway funds are voluntary and the states can opt out. It would be an interesting initiative campaign if the opposition said this initiative was an unstated withdwawl from highway funds, or if the initiative was forced to say that in its text. It would be tragic if people passed the initiative and only then realized the highway funds would go bye-bye. That would drive a wedge between Eyman’s anti-tax supporters against his pro-highway/anti-transit supporters, and probably mean the end of his initiative career.

      1. The Feds have ultimate control over the use of the I-90 corridor. They say rail/no rail then everyone tucks their hat under their arm and goes home. The funding stick isn’t just funding for I-90 it’s ALL federal highway funding for the State. Montana tried bucking the federally imposed double nickle speed limit and D.C. yanked every penny of their highway fund subside. That tends to get the attention of politicians at every level! Eyman is a side show. Vote however you want on his crap; makes no difference.

    4. IANAL, but the contracts clause of the state constitution (Article I, section 23) would come into play. Assuming everything goes to plan, the State and Sound Transit would’ve entered into an airspace lease of the I-90 Center Roadway, at which point you can’t pass a law to invalidate that lease (contract).

Comments are closed.