Future Link riders are mostly in Seattle

163 Replies to “Future Link riders are mostly in Seattle”

1. Yes, it seems the system was designed for commutes to and from downtown Seattle, with little thought for commuters who travel to other cities from Seattle or between other (out of Seattle) cities for their jobs.

   1. except for the Issaquah portion…

   2. The system is designed perfectly well to serve the other major regional cities. The only reason the models predict roughly no one will use it for that purpose is most other cities (except Bellevue) have 99% of their homes and jobs laid out in car-oriented development. That means someone stepping off a train in Everett or Issaquah or Redmond is still miles away from 99% of the jobs and homes of people in those areas. Those last few miles then require a difficult-to-time local rambling bus transfer trying to cover office parks and cul-de-sac neighborhoods, or paying for a car ride, either via park and rides (which don’t scale well) or private hire.

      Any of those cities can (and some are planning to!) change their fate by revising land use near their stations. Here’s hoping!

      1. Redmond is actually quite an outlier here – downtown Redmond has been redeveloping for a few decades now. It has tons of new apartments and retail, and feels far more comfortable to walk/bike in than Bellevue. Also, a good portion of the ~40,000 jobs at Microsoft will be within the walkshed of Redmond Technology station.

      2. While the system is largely designed for moving people to/from and within the center, it still could function pretty well for travel between the less central nodes. Land use patterns really will drive ridership in these places, so to assume the status quo here is to miss the potential.

         It would be terrific to see this same modeling done for different scenarios, including ones more akin to the high-density TOD patterns seen in the Vancouver area. Not showing such comparisons predetermines to a large extent what value people will put in the system we’re building. By showing the alternatives, however, it makes it clear that we actually do have choices on how we develop, and how our ridership responds.

      3. That’s not really true. Everett has a pretty compact and walkable downtown for instance. Of course, the light rail station isn’t placed there…

         Similarly, the Issaquah line won’t have a station in downtown issaquah.

      4. @Brendan, the city of Everett has plans to fill in the area between downtown and Everett Station with a lot of upward density, i.e. condos, retail, etc.

      5. It isn’t always existing density that matters but can be more about what the potential density is. Too many people get caught up here and fail to see that rail stations encourage the creation of density. Also an important factor is transportation corridor density. I think this should be given as much consideration as urban/neighborhood density when designing systems. Given a high corridor density P&Rs can play a pivotal role in station utilization and be effective in taking rubberized vehicles out of the equation

      6. They *can* encourage density. If they do or not depends on a number of factors. Tukwila Internetional Blvd Station?

      7. “rail stations encourage the creation of density”

         That’s “encourage” not “cause”. Capitol Hill, Mt Baker, and Beacon Hill haven’t increased much because of zoning restrictions. East Bay BART stations are in the same situation. The highrise clusters in Vancouver and the large TOD buildouts at stations in in Arlington and Alexandria, Virginia, are because the government encouraged it. In contrast, if a developer wants to build a 10 or 20 story building at Capitol Hill or Mt Baker the government hinders it. And in Mt Baker’s and Beacon Hill’s case it also hinders 6-story buildings two blocks from the station. So they don’t get built.

      8. @degnaw — Agreed, Redmond is not bad for an eastside satellite city! The chunk of apartments near the future station is a good start. While the parks/trails like Marymoor are lovely amenities, they do wipe out a huge chunk of the walkshed from the possibility of frequent trip generators like more apartments. The plan still seems to be for the train to first stop at the no-mans land “SE Redmond station” before a 30-degree sharp dog-leg turn to the northwest to “downtown Redmond”. The latter has a decent few thousand apartments in walking distance, but also a crap-ton of surface parking. I’d like to have the optimism of some other responding comments and say that parking is ripe for future redevelopment, but the skeptic in me worries that won’t ever really pan out. We only have to look at SE Seattle to see the snail’s pace of densification around stations that can happen here. We shall see!

      9. Does anyone know if ST did some scenario planning to arrive at this projection? To present just this forecast makes it look as if this is the inevitable outcome of spending billions of a regional transit system, when of course in reality many possible alternatives exist, particularly if we aggressively rezone for TOD.

         Having alternative maps that show what could happen under such scenarios (e.g., what ridership would look like if we followed more of a Vancouver, B.C., approach), would help decision-makers and citizens better realize that the outcomes of our transit investments aren’t predestined to turn out like this…unless of course we maintain the status quo. Just seeing the one map doesn’t give this impression at all.

   3. @Brendan – incorrect on both cities.

      Everett station is a bit off center but still seems very much within the intended urban core: https://www.theurbanist.org/2018/11/28/everett-adopts-rezones-topping-out-at-25-stories/

      Issaquah won’t have a station in Olde Town, but the station will be in Central Issaquah, which has much higher zoning (8~10 vs. 2 stories).

      1. OK, that Everett upzone is larger than I thought it was. That looks good.

         Ridership will depend partly on density and partly on how much the businesses and services in the area attract people from further south on Link. I’ve gone to Everett occasionally for hockey games, band performances, an MMA school that used to be there, and just to look around. Will the new businesses and organizations offer more unique things that give me a reason to go there?

2. It’s so obvious there needs to be a line from Ballard over to the UW.

   1. It’s at least obvious there needs to be high-capacity transit with as much dedicated right-of-way as the City can offer, taking priority over a few dozen parking spots along the route. If it ends up only being a stub route from Ballard to UW, and not crossing over to Kirkland, then station access time may wipe out the time savings of grade separation. Good luck getting Fremont and Wallingford to put up with major street closures over several years to build stations and maybe guideways. UW-Ballard light rail won’t happen in most of our lifetimes. Ballard-UW-UVillage-Children’sHospital can happen in a matter of a few years, if prioritized, and with the right Seattle administration to not cave to neighborhood obstructionists. The existence of such obstructionists is not a terribly good reason to spend billions for grade separation.

      1. “Good luck getting Fremont and Wallingford to put up with major street closures over several years to build stations and maybe guideways.”

         This kind of thinking is a huge problem here and part of the reason why we can’t get anything done (not your thinking, per se, but the sentiment you express). For too long we’ve allowed individual neighborhoods to hold veto power over infrastructure that benefits hundreds of thousands, if not millions of people. It’s a symptom that is enabled by having a weak transit board peopled with local politicians who don’t want to get yelled at, and thus repeatedly take the path of least resistance, ALWAYS at the expense of transit riders and people who live across some imaginary line on a map.

         You want transit to go where it it will be most used, support reforming how our board is selected.

   2. yup. But we were told that the Ballard riders would overwhelm the system (when coupled with the hordes from Lynnwood) and there was no maintenance facility available.

      The funny thing is when ST used to did a poll of possible routes they offered 1 West Seattle route and a few Ballard routes (Ballard to downtown (albeit without the “Big Tech” stops and Ballard to UW). While the West Seattle route got the most votes (since there was no “Green” or “Libetarian” alternative for West Seattleites, the Ballard to UW route got the second most votes (beating all the Ballard to downtown routes).

      1. Does anyone actually have a quote, or a document saying that Ballard riders would overwhelm the system? It doesn’t make sense to me (as I’ve mentioned many times before). People who know a huge amount of transit systems (people like d. p.) find it hard to swallow. It is quite possible that ST was never worried about it either. Yet is also seems like an easy rumor to start.

         I don’t think the unfounded rumors about capacity are why ST went with Interbay Link. I think it was to serve South Lake Union and Lower Queen Anne. Well, that and to make for a really nice looking ‘X’ on a map. The case for West Seattle rail is far weaker than the case for Interbay Link, and the case for “the spine” is even weaker than that. But they all look great on a map and sound great for those who think mass transit systems operate like freeways.

      2. Sound Transit is a regional agency. They are not looking to build intra-city lines within Seattle. They are not thinking about people riding light rail short distances within Seattle. They were caught off guard by the large number of people riding light rail between UW and Capitol Hill.

         Ballard-UW doesn’t fit into the future expansion of the Interbay line, it would have to be done as a standalone line. Once you think about the line as an intra-Seattle line, it makes no sense to go toward Lake Washington or even Children’s Hospital. After the U-District it should have a stop at U-Village and then turn north at 25th Ave NE and pop out from underground at Lake City Way and 25th. The last stop would be at 145th and Lake City Way where there is more space to build an operations/maintenance facility and the terminus on the other side would be in the real center of Ballard. The Interbay-Ballard line would continue north to Crown Hill as drawn on most maps. The Aurora Line as drawn by Seattle Subway would be the last piece–then you have 3 lines serving all of North Seattle.

         Ross, the article says 32,000 will use West Seattle Link and 34,000 will use Interbay. How does a difference of 2,000 riders make the case for one line ‘far weaker’ than another?

      3. “I think it was to serve South Lake Union and Lower Queen Anne. Well, that and to make for a really nice looking ‘X’ on a map.”

         It’s because of downtown bias. Most people in Ballard and West Seattle want to go downtown, don’t they? And they’re suspicious that a detour through the U-District would be as time consuming as the current north-south bus routes. That’s where the opposition is coming from. Ballard-downtown is also driven by the need for a second downtown tunnel to deal with increasing capacity needs downtown. SLU was not part of it because it was added to the line late in the process, long after McGinn tilted the expectations toward Ballard-downtown and Murray didn’t counteract it. But with the late realization that SLU needs high-capacity transit, Ballard-downtown became insurmountable. Interbay is a minor bonus, driven mostly by the lower cost of no tunnel north of Uptown. The other alternative was a Queen Anne/Fremont tunnel. And the remainder was because 15th Ave W looks like a wide expressway that could accommodate elevated or surface trains, so it would be a shame not to leverage that. It’s the same legacy attitude that led to Northgate and 145th stations being where they are: the transit center/P&R/mall is at 1st & 102nd, and 145th has an existing P&R under public ownership and the adjacent I-5 entrance must be good for Link somehow. (Although that’s a false assumption because 145th Link riders won’t be driving from I-5 north and south; that’s highway thinking.)

      4. @Joe — You have to look at time saved per rider. The bulk of the ridership from West Seattle Link comes from terminating the express buses. That means that a rider (on a bus like the 120 or 21) has to get off the bus and wait for the train, right when the bus was about to get on the freeway (with the next stop being downtown). This might be welcome during rush hour, but it will be a big pain outside of it. At the same time, there is very little in the way of new improved connections. When Link gets to Northgate, one of the biggest improvements will be Northgate to Capitol Hill, which will go from being a very time consuming two seat ride to a quick and frequent trip. West Seattle Link will have none of that.

         With Ballard Link much of the ridership comes from Lower Queen Anne and South Lake Union. These are trips that don’t involve transfers. Instead of walking to a bus, you walk to a train. The train, meanwhile, will get there much faster than the existing buses. In terms of new connections, Ballard to South Lake Union (a relatively common trip) will be a lot faster.

         To be clear, if they manage to actually skip Ballard, then all bets are off. At that point, the two lines are a wash, and remarkable for how poorly they perform, given that neither leaves the city limits. If someone has to catch a bus from Ballard to even get to the Ballard station, then it starts resembling West Seattle rail. Folks were better off when the bus just kept going.

         As to the main topic, I completely agree with what Deborah wrote: we obviously need a Ballard to UW subway. Furthermore, it obviously would have been a better value and thus should have been what we built next. Again, a big reason is time saved per rider (which is huge). It would be faster, *at noon* to take the train instead of driving, let alone taking the existing bus. It connects much better to existing (relatively fast) transit, and provides a much better network. It should be clear by the data (as it was to anyone who had studied such things) that urban systems — especially those that can have good stop spacing — are better values.

         Based on the numbers (presented here) and the geography, the case for a combination of a Ballard to UW subway, a Metro 8 subway and the WSTT would have been the best value. The only thing we are likely to get is the Ballard to UW line and most likely it will simply stop at the UW.

      5. Yes, they could have added a bus-only exit from the WSB to 99N and accomplished the same thing as a 1.5 billion Link line. But the southern end of the downtown tunnel had to go somewhere and it made political sense to send the line to West Seattle. Unfortunately it also made political sense to go up the hill to the Junction which is a mistake for future expansion of the line. Not to mention the current engineering reality of sending an elevated light rail line up a giant hill along Genesee.

         One thing to keep in mind is that a Ballard-UW subway line isn’t going to come by itself. If there is going to be a tax within Seattle there will need to be ‘subarea equity’ within Seattle and underrepresented communities in south Seattle are going to want their share of the light rail. Metro 8 will have to be on the proposal as well and the Aurora line will need to be saved for “ST5” or later.

      6. For Ballard-UW line, which I’d really like to see, geology could be a major decider. If the ground is good, deep, straight bore could be done in a couple of years.

         If it’s ok to keep using the trains with light rail capacity, meaning surface running past Children’s Hospital and across the lake at Sand Point, definitely do Ballard-UW-Kirkland.

         Would make Ballard-Fremont-“Gasworks Park”-UW a surface streetcar line. It’ll carry a lot of people.

         But again and ’til I get noticed about this: for every paragraph about the “turf” divisions of the ridership, let’s have a page on the “dirt”- and rock- in front of the cutter.

         Because with our landscape, the terrain is a major stakeholder- and not just the ones for surveying.

         Mark Dublin

      7. “But the southern end of the downtown tunnel had to go somewhere and it made political sense to send the line to West Seattle.”

         That’s putting the effect before the cause. West Seattle was not because the Ballard-downtown tunnel had “nowhere else to go”. It was because West Seattle contains Important People and People Like Us, and they didn’t get the Monorail, and it’s a whole quarter of the city. So West Seattle Link was going to happen whether Ballard Link did or not. ST’s first proposal was West Seattle Link and a Ballard streetcar, which went down like a lead balloon. The Ballard corridor and the downtown tunnel were combined into one corridor study, so that’s what started thinking of them together. But Ballard really needs a downtown tunnel because surface in Balltown and downtown won’t cut it, and the elevated monorail got a lot of pushback because 2nd Avenue businesses didn’t want it running in front of their windows. If there were no Ballard Link then it’s an open question what West Seattle Link would do downtown because ST never addressed it.

      8. Mike nailed it. ST really wanted West Seattle rail, putting it ahead of other projects (like Ballard to UW rail) that performed better. Projects like a Metro 8 subway simply weren’t studied. The push for West Seattle rail lead to the desire for another tunnel, which in turn made the numbers for Ballard (Interbay) Link look better. Ballard to Westlake doesn’t perform as well as Ballard to UW. But if you add a second tunnel, then you get additional trips within downtown, and thus the numbers start looking better for West Seattle. When I write “numbers”, I mean ridership. Time saved is likely much better for Ballard to the UW, no matter how they try and cook the numbers.

      9. It’s always amusing to come over to STB and see commenters confidently proclaiming why the system is being built a certain way and why West Seattle is getting light rail without having any clue what they’re talking about and no evidence to point to… ST has never been an unbiased agency and will never have as its main focus to build the best transit system for our region. So you really shouldn’t be surprised when the Board makes decisions that don’t primarily address that goal.

   3. Um, “No”, it isn’t “obvious” at all. The 31/32 run every 15 minutes between them, even at the rush hour. The 44 runs seven buses in the 7:30-8:30 AM rush hour and 4:00-5:00 and 5:00-6:00 afternoon rush hours. Other than that it’s six buses per hour.

      Between the two lines that’s about 11 buses per hour or somewhere between 500 and 700 riders per hour in the peak direction. And that’s assuming you capture all of the bus riders, which clearly won’t be true with ST’s planned single station in Upper Fremont and single station at 45th and Wallingford.

      Even in Ross’s Valhalla of every downtown-bound rider on the 5, C, 28, 62 and 26 transferring at 45th, you’re still in the two to three thousand riders per hour range, WAY below the threshold necessary for underground rail.

      Of course that could change if the blocks within two north-south and three east-west of 45th and Wallingford, 46th and Fremont and 14th/15th and Market were all upzoned to twenty stories. That is unlikely, however, to happen.

      1. Nobody rides those buses because they are SLOW. I can beat the 44 to Ballard on a LimeBike. I can get to Redmond faster than Ballard from the U-District. Fast connections between urban villages will induce demand. Especially if you have a quick transfer to the red and blue line at the U-District.

         The demand at all stations within Seattle will grow as you add more connection points to the line. Especially if you reach the point where you can get to all major urban centers in Seattle on a subway or rapid bus line.

      2. Ballard currently has a half-hour overhead to get to the region, which means getting to the U-District or downtown transfer points. (Or to a lesser extent Northgate.) And the 44 between 3 and 7pm can get up up to 45 minutes. You could solve that with transit lanes, a subway, or something slightly less than full transit lanes. You could also give a thought to the car backups at the I-5 entrance. Wallingford businesses have too much clout to take away their street parking, although we still haven’t seen the RapidRide negotiation yet, so it may be better than worst-case. The east-west streets are also narrow, so strategies that would be feasible on wider streets and with more through streets aren’t feasible there. 45th on the west side of Phinney Ridge is three lanes, so if you take two for transit that leaves only one car lane going in one direction, and there aren’t any neighboring streets that go through without a big detour. Likewise, an elevated line would go right next to people’s houses, with more impact that it would on MLK or Pacific Highway.

      3. What Joe said. When transit is faster than driving *in the urban core* lots of people take it. Not only for single seat trips, but combinations that are a real pain right now. For example, Northgate to Ballard.

         By the way, TT, how do you justify any part of ST3, based on buses per hour? The 44 runs *more often* than the D Line (six trips per hour instead of five). The 512 (to Lynnwood and Everett) runs four times per hour. Even parts of ST2 perform poorly based on that, which is just silly. The 41 runs “only” every ten minutes, and the 512 is the only all day service to Lynnwood.

         It is just wacko the way that people think the trains will be crush loaded because of suburban demand, while urban trips aren’t that popular. Every bit of evidence out there as well as past experience suggests the opposite. Build a better urban network — a network that provides for lots of trips that are faster than driving — and you get plenty of riders. Extend the subway lines out farther than New York, Paris or London has, and you get bupkis.

      4. “And that’s assuming you capture all of the bus riders, which clearly won’t be true with ST’s planned single station in Upper Fremont and single station at 45th and Wallingford.”

         Beware of assuming that the current bus riders will be the only light rail riders. Link will be faster, more reliable, and more off-peak frequent in the Ballard-UW segment in question, but it will also give a one-seat ride to all the other neighborhoods along the line. (Which is unclear at this point because we don’t know what the other neighborhoods beyond Ballard-UW will be, but already with Central Link you have a massive improvement for Rainier-SeaTac, Capitol Hill-Rainier, UW-SeaTac, and soon Rooevelt-Rainier and Roosevelt-Bellevue.)

         Seattle Subway has recommended extending Ballard-downtown north on Holman Road to Northgate. ST’s long-range plan already has a line from Northgate to Bothell that it could interline with. But ST doesn’t seem interested in the Ballard-Northgate segment.

      5. Well, Ballard-Downtown will capture not just traffic from the D line north of Market, but also from the 15X if there is still local service on 15th NW, the 17, the 18X, and the majority from the 40.

         But the main justification for the Green Line is South Lake Union and Lower Queen Anne. Truncate the damn thing at Expedia if money needs to be saved, but that part of town needs a subway. Wallingford doesn’t because the comfortable burgers [sic] who live there would raise Holy Hell if the high rises necessary to support a three mile subway sprouted there.

         Don’t be a Polonius claiming that Lynnwood Link is only justifiable because of ridership from the 512.. Snohomish County ridership to downtown Seattle is a HELL of lot more than the 512’s. There are 86 buses in the hour between 6:30 and 7:30 that cross the King-Sno county line southbound to Downtown Seattle. And that’s today; in five years when Link gets to Lynnwood, it’ll be at least 100. Snohomish County is growing rapidly.

         Now of course those buses each have only three or four people on it, because they’re from the ‘burbs, right?

         Of COURSE ridership will be highest closest to downtown. Hello, that’s where people are going to work and sorry to tell you this, but that’s what Link is for: to get people to work and back.

      6. Joe, of course what you say is academically true, BUT, a subway between UW and Ballard would have at most three stations between the end points: Meridian/Wallingford and 45th, 46th and Fremont both in subway and 8th and Market, probably elevated. Unless the nearby hinterland of every one of those stations is allowed by the City to be built to twenty stories there will simply not be enough people in walking distance to fill a subway train.

         It’s not practical to serve Wallingford and downtown Fremont, so either the biggest job center in the corridor would be skipped or the bus intercept for the Aurora buses would be eliminated.

         Will people please stop obsessing on Ballard-UW. It should come after a Metro 8 distributor around downtown.

         You’re not in school at UW any longer nice as those days may have been.

         All the people between those stations will still be riding the 44 and so won’t be getting any advantage from the subway. Or they’ll be “walking up” the time they save once they’re on the train vis-a-vis the 44.

      7. that’s what Link is for: to get people to work and back [during rush hour]

         So, basically, you are saying we are building the most expensive commuter rail line in North America. Great. I guess it is time to remind you that every commuter rail line carries far fewer people than the corresponding city wide transit system, even when the city wide transit system covers a subset of the urban core (like San Fransisco Muni buses) or runs extremely slowly (like San Fransisco Muni buses).

         Ballard-Downtown will capture not just traffic from the D line north of Market, but also from the 15X if there is still local service on 15th NW, the 17, the 18X, and the majority from the 40.

         Why do you assume that the majority of the 40 riders will switch to Link? During rush hour, you would take the 18 if you are headed downtown. Even outside of rush hour, the 40 slogs its way towards downtown — the only reason to stay on it from Ballard to downtown is if you just don’t like making transfers (to faster, more direct buses). My guess is the majority of the folks are from places along the way, since it is the fastest way to get from Fremont to Ballard, or Westlake to downtown. It is also one of the better ways to get from Fremont to Westlake or downtown. None of this will change. A transfer will still be a transfer, and those other trips will still be faster via the 40.

         You are also missing the point. You are using midday data for the 44 (“six buses per hour”) while ignoring the fact that service to Lynnwood is similar. Of course there are additional buses during rush hour, but still nothing like the obvious ridership of buses like those that went from the UW to downtown, or buses like the 41. The 510 carries less than 2,000 people a day. The 511 carries 2,000. the 512 carries 4,000 and the 513 carries less than 1,000. That means ST carries less than 10,000 people a day from Snohomish County into Seattle (or less than many Seattle buses). Community Transit doesn’t release data per bus, but the entire system carries only 33,500 people a day. Swift has the highest ridership of any bus, carrying about 10% of the riders. Even if you are being really generous, and assuming that a full third of the remaining ridership is to Seattle, you are looking at 20,000 riders for Snohomish County.

         The entire system is based upon the assumption that more people will ride if service improves. Of course it is! That is the whole point. I have no doubt that more people will ride from Snohomish County, because now trips to places like the UW or Northgate will be much faster.

         But the increased ridership caused by speed improvements along that corridor will be minimal compared to that along the Ballard to UW corridor. The speed difference is bigger, the connections are much better, the distances much smaller (i. e. the trips take way less time) and the areas much more urban. Sorry, but there just aren’t thousands of people who will suddenly ride transit because the trip from Ash Way to Mountlake Terrace is faster.

      8. a subway between UW and Ballard would have at most three stations between the end points

         So what? It is an extension. Holy cow, the U-Link extension added only *one* station between endpoints. It added only two new stations, with the one at the UW horribly flawed. Yet ridership has more than doubled as a result of those two stations! Ridership at just about every station went up. That is a network effect caused by faster travel — which is common in urban systems. The only stations where ridership didn’t go up were in Tukwila and SeaTac, the result of people shifting to Angle Lake. That is because the network effect is weak for suburban stations.

         It’s not practical to serve Wallingford and downtown Fremont, so either the biggest job center in the corridor would be skipped or the bus intercept for the Aurora buses would be eliminated.

         Nonsense. It is simply a matter of prioritizing one or the other. If the train intersected Aurora at 45th then the transfer would be easy. If it happened in lower Fremont, it would require more work, but it still wouldn’t be that hard. You just need to add bus stops at the north end of the Aurora Bridge, and put one entrance there. The subway line could look like this: https://drive.google.com/open?id=1V7SVqymYwyy29rdVEz5Spw0XZPouoD7y&usp=sharing, sharing both Ballard stations. One end of the Fremont station connects well to the western side of Fremont (at Fremont Avenue). The other side connects with the buses that run on Aurora (the 5, the E, 26, the 28). You wouldn’t really need a stop at 8th, because the 28 would still intersect Link in Fremont. You would keep the 44, but it would carry fewer people (it would mainly be used for shorter trips). It wouldn’t be super expensive to build because you could reuse the stations for Ballard Link (assuming the train really did cross at 14th and then turn). This takes way more imagination than ST has ever shown (ST’s motto seems to be “close enough”) but that is no excuse for not building what should be built.

         Regardless of the particulars, you are completely ignoring the network effect that comes from improving travel times in an urban environment (even though it lead to dramatically improved ridership here, as it has all over the world).

      9. “But the increased ridership caused by speed improvements along that corridor will be minimal compared to that along the Ballard to UW corridor. The speed difference is bigger…”

         … and it’s a cross line. A line that meets in a + or a T facilitates more unique kinds of trips than a parallel line or an extension, because it facilitates not only north-south trips and east-west trips but also north-west trips and west-south trips like Roosevelt to Wallingford and Fremont to Capitol Hill. And if it continues east to U-village and Children’s then it supports north-east and south-east trips too. So you get both people who could take either of two north-south lines and people who wouldn’t be served at all if there were no east-west line to areas between the north-south lines or outside of them. And 45th is the highest-density east-west corridor in north Seattle and probably south Seattle, and where UW is, so it’s not like it’s some arbitrary location like 85th or 130th.

      10. “that’s what Link is for: to get people to work and back [during rush hour]”

          “So, basically, you are saying we are building the most expensive commuter rail line in North America.”

          Link is for several things.
          1) To provide full-time connections between urban villages.
          2) To move the peak-hour crowds in the biggest bottlenecks, particularly downtown-UDist, downtown-Lynnwood, downtown-Bellevue/Redmond, and downtown/federal Way.( Everett and Tacoma are subject to debate.)
          3) To move the crowds at ballgames, parades, 4th of July, etc.
          4) To provide more all-day capacity, frequency, and reliability in its corridors.
          5) To take people to nontraditional jobsites at nontraditional hours. (e.g., all the janitors and hospital workers working in scattered neighborhoods on swing/evening/night shifts)

          The political rhetoric focuses mostly on trips to work because that’s how people make money to feed their families, how businesses can operate, and how the government can generate tax revenue. In short, it makes the economy go round. It focuses on peak hours because that’s when the biggest bottlenecks are.

   4. Actually, to me this map says the Ballard Line should be extended north to connect with Central Link at Northgate. Southbound trains could branch at Northgate, with half the riders traveling via Ballard, to reduce the crush in the most-heavily traveled segment: U-District to downtown.

   5. $1Bn for a West Seattle tunnel? Use it for a Children’s Hospital – UW – Ballard streetcar with a separated two-way cycle track instead. And by Ballard, I mean we still need to get from the Link station at 14th or 15th Ave. NW to *actual Ballard*.

      1. Exactly, B, though you need to call it a tram to avoid Ross’s wrath. Also because five section trams are a lot more efficient than the current tiny streetcars.

         The NP right of way is wide enough for both the trams and the trail in most places. You’d have to cut-and-cover through “downtown” Fremont but the rest of the route should work and goes very close to the workplaces to which people want to travel from all over North Seattle.

         Swing north in the median of 14th and turn into Market with center running. Yes, it takes a lane from cars on Market, but if it’s a paved shared ROW, the 44 trollies can safely have their poles offset toward the center of the roadway, clearing the collectors on the trams.

         Chad,

         It’s unlikely that ST would go for that, because of the very high flying junction that would be required just south of Northgate Station.

         It’s an interesting idea, though.

3. Another way of looking at this is to realize that if the operations plan hold, we will be paying for lots of empty train segments.

   1. Are you referring to the operations plans that have East Link turn back at Northgate, Lynnwood TC, or Everett?

      Issaquah Link can be decently full just by running it as one-car consists. But the most serious density in Issaquah is way out at the Highlands. There is several blocks around there that is at least built denser than the area around Capitol Hill Station will ever be.

      If East Link is itself a ridership bust (which I don’t think it will be, given the connection to and between downtown Bellevue and Redmond Tech Center), then I hope ST will have tracks that enable it to be through-routed to Ballard, and let the higher-ridership north and south ends of the lines stay together, with operators transfering instead of trainloads of passengers transfering.

      1. Given the relatively short distance between Redmond and the Spring District, I think the outer East Line won’t be too much of a cost drain because it’s only about 15 minutes.

         It’s the time consuming crawls between Tacoma and Federal Way, and between Downtown Everett and Mariner Way (even with the turnback considered) that I’m alluding too. I’d also note that the Issaquah -Kirkland line looks so low that it makes me wonder if a different operations plan or technology (Driverless? Rubber tired? Single track sections? Two East Link branches?) would be better.

      2. Northgate will have a turnback so trains can turn around, but that’s not the current plan. A Northgate turnback is needed because there’s no turnback between Stadium and Northgate, and not having the flexibility of terminating at Northgate is going too far. The earliest plan had East Link going to Lynnwood peak and Northgate off-peak. The current ST2 plan has East Link going to Lynnwood full time. The ST3 plan has East Link going to 128th Street south of Everett. That will be driven by ridership demand. If demand is different than predicted, East Link might terminate at Northgate or Everett Station at some times. (Or Ash Way if it has a turnback, but I assume it won’t since 128th is so close.)

         The whole idea of Everett riders going to Paine Field is ridiculous. Part of the motivation for Paine Field is to convince drivers from Marysville and further north to park at Everett Station and take Link to Paine Field, to relieve congestion on the Casino freeway. That was in Everett’s or Snohomish’s’ reasoning for the detour. It’s dubous because it’s such a short driving distance in an exurban area with wide streets, that I can’t see a lot of drivers taking up the offer. And its primary benefit is to people outside the ST tax district. So I’m skeptical about it.

         But there’s another benefit of Paine Field that hasn’t gotten much attention. It will be good for people riding from Lynnwood/Mountlake Terrace to Boeing, and a major improvement for people riding from King County to Boeing. (And yes, a shuttle from 99/Airport Way would be just as effective as detouring to Paine Field. The point is Link reaching that far north.)

      3. The Paine Field stop could also serve Mukilteo pretty well with a bus connection. It would be a much shorter bus ride from Mukilteo to Paine Field than Lynnwood or Everett, which theoretically means that the same number of buses could provide more frequent service. So, instead of having a 45-minute bus ride to the train that comes once an hour, they can get a 10-minute bus ride to the train that runs 2-3 times per hour.

         The quicker the shuttle ride from Mukilteo/Edmonds to Link, the stronger the case for eventually eliminating the North Sounder money sink.

      4. Part of the motivation for Paine Field is to convince drivers from Marysville and further north to park at Everett Station and take Link to Paine Field, to relieve congestion on the Casino freeway.

         Hard to see how that will work. The drive from Everett to Marysville is horrible, as there is no HOV lane. That really is one of the big things they need to add. That would speed up the HOV lane for everyone (it gets bogged down now because it essentially ends at the bridge).

         Likewise it seems like they could improve the Boeing Freeway (what you called Casino Freeway). There is one little section that has an HOV lane, but that is eastbound to southbound (and even then I think it ends in the regular lane).

         Anyway, I think the big reason Everett took the approach they did was optimism. While everyone complains about the extra time it will take to get from downtown Everett to Seattle, I give them credit for rejecting that. It is just too far, and you can’t expect a lot of riders to (or from) Seattle. But if you build good transit connecting two employment centers fairly close together (Paine Field and downtown Everett) along with a few stations along the way, it might revitalize the city. Nice idea, but I just don’t buy it. I can’t think of any city that saw their fortunes change because they built a light rail line. Plus there just aren’t that many stops. From Lynnwood to Everett, there are only 8 stops (inclusive) and that includes the provisional stations, and stations that will have giant park and ride lots next to them, right across the way from the freeway (not exactly places poised to be the next Ballard). But more than anything, it is just a huge amount of money in the hopes that the city will incur massive growth — something that isn’t happening now despite Boeing doing extremely well and other Puget Sound cities booming.

      5. “Part of the motivation for Paine Field is to convince drivers from Marysville and further north to park at Everett Station and take Link to Paine Field, to relieve congestion on the Casino freeway.”

         “Hard to see how that will work.”

         I don’t believe it either but it was part of Everett’s reason for wanting the Paine Field detour. I don’t know what the drive in Marysville is like since I don’t have a car and don’t go there peak hours, but I have a hard time believing people will drive 75% of the way and then take Link the other 25%, when that area has wide roads and free parking. And what would they do at Seaway Station? Take a shuttle? That’s a three-seat ride, and it’s not an area like downtown Seattle where people would be more accepting of that.

4. An obvious point for the future: there should be no ST4. No more expansions to places with little to no density.

   Any expansions in the future should involve a Seattle-only plan and vote. Ballard to UW, Metro 8… whatever. Don’t waste money on an extension to Snoqualmie Falks or whatever far-flung destination some politician thinks is important.

   1. Sounder expansion maybe?

      1. There is a big bucket of money in ST3 for Sounder improvements. Not saying what they are for should improve ST’s hand vis-à-vis the for-profit BNSF. With a little better cooperation from some suburbs (I’m looking at you, Kent City Council), third- and fourth-tracking could enable schedule expansion much more cheaply than buying more trips on the current two tracks.

         This would also help reality-based higher-speed Cascades trips.

         I would surely love to have Sounder reach downtown Olympia some day, but Olympia has to want it to happen, or it won’t happen.

      2. ST3 has a budget for Sounder South improvements. We’re still waiting for ST’s negotiations with BNSF to conclude so we can find out how much it will buy. It may allow hourly trains with some evening/weekend service and track improvements. Or it may end up with nothing and ST will have to come up with a non-Sounder strategy for additional service. The result of that will basically determine what’s feasible in ST4. If ST can’t get all its desired ST3 service within the budget, then in ST4 it will want to fill in the remainder or forget about Sounder expansion.

   2. Agreed. At best you could cobble together some sort of system of bus improvements for the suburbs along with rail where it is effective (in Seattle). That type of system has proven way more successful than what we are building. Unfortunately, while it would have made sense for ST3, it would be difficult to come up with a proposal without admitting that ST3 projects for the suburbs were a bad idea. BRISK (https://seattletransitblog.wpcomstaging.com/2015/05/06/brisk-making-it-fast-frequent-and-reliable-alt-2/) makes way more sense than Issaquah to South Kirkland rail, but is hard to build once we are committed to the latter.

      I think you are right in that our best chance at improvement will involve Seattle doing it alone.

      1. I don’t think Redmond Link or Issaquah Link killed BRISK. It was more NIMBYism that killed BRISK. You know what will really be a dud on the eastside? Freeway running “BRT”, er STRIDE.

      2. ST killed BRISK. They rejected the report from the independent transit consulting group that Kirkland hired to look at the issue. Kirkland faced opposition from NIMBYs about running buses on the CKC, but was ready to just move ahead on the issue. Sound Transit, on the other hand, wanted rail (even though they didn’t have a study to support the idea). As a result, they punted, and went with the rail line and the “BRT” line, both of which make little sense.

      3. The suburban subareas already have some small goals. Pierce wants a Link extension to Tacoma Mall. Snohomish wants a Link extension to downtown Everett and Everett Community College. East King probably wants to extend Issaquah Link to Totem Lake, although downtown Kirkland will continue to be a dilemma. (Bypass it and lose the riders, or use the Cross-Kirkland connector and draw the wrath of “Save Out Trail”.) There are also concepts for a 522 line, a Ballard-Redmond line (or variations terminating in Kirkland or Issaquah), or a UW-Bothell-Kirkland line. If the suburbs go small with the first two and TBA BRT, then North King wouldn’t have much money for another subway. If they go big, then it’s a different picture. I have trouble seeing them go big, especially when Everett’s, Tacoma’s, and Issaquah’s ridership turn out to be disappointing, along with the Tacoma Link extensions. And in a sense we’re already doing ST4 because the 25-year plan has part of what was expected to be in it. I can’t see ST4 being as big as ST3, and maybe not even as big as ST2, or even at all.

         The ST district and the state will eventually have to confront the increasingly-diverging wishes of North King and the other subareas. The suburbs wanted big initially to get closer to Everett and Tacoma, and in ST3 they wanted big again, but maybe they’ll have second thoughts about any of the Kirkland proposals or 522 or Burien-Renton or 405 south of Bellevue. Seattle will definitely want another subway or two. ST can’t charge a different tax rate in the different subareas because it’s a single tax district. So it may be necessary to split the tax district. That means different boardmembers would be eligible to vote on different subeares’ proposals, or the board might have to be split too.

      4. Sure, there are places that have projects, but my guess is ST4 would lose if it included the suburbs. ST3 failed in Pierce County. It passed in Snohomish County probably because a lot of voters forgot that Lynnwood Link was happening regardless. In any event, the whole idea was that this was a big spine — a once in a lifetime chance to connect Everett to Tacoma and all places in between. Sure, it doesn’t actually go to downtown Tacoma, but again, Pierce County isn’t likely to suddenly embrace the thing.

         With the spine “done”, I just see a big disconnect between big suburban projects (which will always have really bad numbers) versus urban ones (that are the opposite). You could promote better bus service in the suburbs, but that has lost in the past. I think the only way we will add more rail is if Seattle does this by itself.

      5. The lack of the Spine will make it harder to sell a regional package, but on the other hand, there will be a much larger portion of the region actually using Link & therefore interested in a better system.

         And an ST4 can always include bus investments for the suburbs. For example, ST4 could pay for new bus bases for PT and CT, a 4th SWIFT line for Snohomish, or even road improvements like HOV lanes & direct access ramps. The package doesn’t need to be limited to just Link spending.

   3. At least we agree that a Seattle-only system should involve a vote of Seattle-only taxpayers, and tax increases only in Seattle. It is probably a more high-priority need than ST4, but if it only serves Seattle, then Seattle needs to cover it with their own tax base.

      I’ll disagree on ST 4. You consider Tacoma far-flung. Guess what? I consider Seattle and Shoreline far-flung. It’s perspective. I hate going “way up there” for meetings. Seattle’s the sticks. Millions of people in South King, Pierce, SnoCo, and the Eastside need to get to and from work. ST really needs to serve the entire UGA, and ST 3 gets us pretty close to that, but with a single spine that misses a lot of dense locations. Extensions within and between areas already served, plus some limited extensions further out are what I envision. In 50 years, a spur off of MLK to Link out to Renton? Possible. Link extensions out from Bellevue north and south? Why not?! Fife-Kent via the old Interurban? If the density increases, sure. Something more to serve Everett? Of course. As the region grows, and the “rural”ish pockets between suburbs infill, we will need transit throughout the burbs. The big question is whether we’ll make the investment to build something efficient like Link, or whether we’ll continue adding lane-miles to roads, let them get clogged in traffic, and have buses that are no more efficient than just driving yourself. I’d prefer to infill with more light rail.

      1. We may be getting a Seattle-only investment much sooner than that: some new city taxes to fund making West Seattle Link and Ballard Link lesser-ridership by hiding them underground, and to make the ID Station transfer really awful with a several-story elevator ride, because impacts brought up by neighborhood associations are more important than making transit work for riders.

         Whether such a plan passes or dies, it will pretty much kill any faith Seattle can properly execute a Seattle-only ST4. I will be voting against ST 2.9 should the City give us a say in the matter. I hope a more pro-transit candidate will run for D1, and that Mike O’Brien not cave to the tunnel advocates (mostly the Port?) in Ballard. Oh, and also that Ballard Station be built within the walkshed of Ballard.

      2. according to Erica B’s twitter feed, Mike O’Brien is still on the fence about running (he’s filed the paperwork , but his polling wasn’t great if a serious candidate like Sara Nelson got in the race)

      3. I don’t think you get it. Running subways out to low density suburbs doesn’t work. It hasn’t worked anywhere. Even in the Bay Area, where they have huge suburbs, with millions of people living outside the urban core, it doesn’t work. Even with extremely fast trains that make very few stops it doesn’t work. BART was an experiment and it failed. They should have built a better subway for the urban core (San Fransisco/Berkeley/Oakland) and then improved express bus service (and commuter rail) to the suburbs.

         As I’ve written before, there are no studies to support ST3. We are in the process of spending an enormous amount of money for very few passengers. That is because (for example) not that many people want to spend an hour and a half getting from Tacoma to Seattle every day. Instead of pretending that there are, we should do what every successful transit system in the world does for the suburbs, which is to run express buses and commuter rail (re-using existing tracks) while providing a good subway system in the urban core. Doing so means that folks from the suburbs can actually get around after they get to Seattle.

         Perhaps an example will help. Let’s say I live in Everett and work in Fremont. Right now a rider takes a series of buses, or they just drive. In a few years, they will be able to take a bus to Lynnwood, then Link to the UW, then another bus to Fremont. This is a very big improvement. Obviously the weakest link — the slowest and least reliable trip — is the last one. When ST3 is complete, they will be able to take a bus to one of the stops in Everett, then Link to UW, then that same bus to Fremont. Nothing has really improved. The train from Everett may actually be slower than the express bus from Everett to Lynnwood.

         Now imagine a subway line from the UW to Ballard, stopping in Fremont. You’ve dramatically improved that trip, along with any that start in Snohomish County. Oh, and way fewer people commute from Everett to Seattle than from Lynnwood to Seattle (because Everett is just farther away and more independent).

         Now imagine the destination is First Hill or Belltown. Or imagine that someone started in Tacoma, and Tacoma had express buses running in HOV-3 lanes (or HOT lanes). It is pretty easy to see why what we are building (a BART style system) has failed every place it has been attempted. It is focused on a relatively small set of riders, and adds very little for them.

      4. To say that BART failed is unfair. After all, after these nine high-volume stations there are many more stations with at least 70 percent of the activity of ninth-highest one like Walnut Creek, Fremont, Bay Point and Dublin.

         The biggest BART “failure” is the Oakland Airport Connector (an airport with 800-1000 percent more flights than Paine Field will have). Of course it’s inflated fare keeps it from being used by nearby non-airport employees. That station is not far out in the suburbs.

      5. The Puget Sound has a long distance commute bus network that is far more productive than any other N.A. city. I think our metro is unique enough, and has well established trip patterns, that long-distance, high frequency suburban rail backed by strong feeder bus networks will work in ways that it simply hasn’t work elsewhere.

         The problem with BART isn’t BART. The problem with BART is the poor land use around suburban stations. If Seattle’s suburbs can build good activity centers around most suburban stations, ST3 will be a success.

         Look at Caltrain – the line runs exactly where it should, through historic town centers, because the ROW was established well before the automobile age. Ridership is solid for commuter rail, but it would be much, much more if Silicon Valley allowed even modest midrise growth. Caltrain long term plan is to basically convert it to Link – electric trains, high frequency, and long end to end running times. Whether than plan will work as little to do with transportation and everything to do with how the Valley decides to grow.

      6. Ross,
         I don’t really think you get it. Federal Way Link isn’t about Seattle. It IS about Federal Way, Highline, Tacoma, Fife, and Seatac. If Kent and Federal Way rezone extensively around Link, then there will be a corridor of high-capacity land use, and not in astronomically-priced Seattle. What will eventually happen is serious business development around each suburban station.
         Tell me, what is the land use around each of the BART stations? Hayward is medium density residential and low density business district as one example. San Leandro? Walnut Creek?Federal Way’s zoning allows 200 feet height in the city center, and Light Rail hasn’t even arrived yet. I wonder what an upzone would look like, especially once big money is wanting to invest, pay property taxes, etc, and Link is under construction. Bellevue 2.0. Tell me that there is no density in Tacoma, either? Federal Way Link is Central Link v.2. More downtown cores to the airport (with extensions to distant Seattle and Bellevue).
         Based on Ross’s assessment, you’d think that Federal Way Link was going to a little town called Mayberry. Federal Way and Tacoma, pal, have over 300,000 people, not counting the other little suburbs around them. Seattle’s population was 560,000, when ST1 was under development just after the 1996 vote. Let that be perspective for you. South Sound will be a few million people in a few years. The freeway systems are (fortunately) underdeveloped, so we need to get the transit infrastructure in place now, before it becomes a massive car sewer resembling Seattle (520, 90, 99, 599, Mercer Mess, etc).

      7. … some new city taxes to fund making West Seattle Link and Ballard Link lesser-ridership by hiding them underground…

         What the what? While I can’t comment too much on West Seattle, how does burying Ballard Link cause “lesser-ridership”? I think the “lesser-ridership” concern for Ballard Link is almost solely focused on station location (15th v 14th), of which both have tunnel and bridge alternates.

      8. Then you need to convince your Representatives and Senator in the State Legislature to support giving Seattle the right to determine it’s types and rates of taxation in order to pay for it.

         Even Democratic legislators from outside King County have been unwilling to do so, let alone the covetous Republicans from Wheatsylvania.

      9. @ Al and AJ — I didn’t say that BART is a complete failure. I said the BART experiment failed. It was a reasonable hypothesis. Build a brand new subway as you would a commuter rail line (with very big stop spacing, few urban stops and very fast trains). The idea being that long distance commuting would be even more popular than urban travel, since the time saved (versus a crowded freeway) is even greater.

         But it doesn’t work that way. It doesn’t work that way *anywhere*. There is no place, anywhere, where commuter rail has more riders than the corresponding urban transit system. Even in San Fransisco, where the buses are extremely slow (the slowest average transit speed in the U. S.), more people ride the Muni buses than the entire, 112 mile BART! Just let that sink in. BART serves San Fransisco, Oakland, Berkeley, Daly City as well as a bunch of large suburban cities. It provides the essential connection between East Bay and San Fransisco. It does all this at ridiculously fast speeds. Yet it still carries fewer riders than the slow as molasses *buses* of *just* San Fransisco.

         I think it is obvious to everyone involved that it would have been better if they had built a traditional subway system with traditional stop spacing. Add a bunch of stops and lines in San Fransisco, Oakland and Berkeley. Build solid bus intercepts in the suburbs in places like Daly City. Then improve the express bus service and commuter rail to the suburbs and call it a day. If they had done that, then there would be much higher transit ridership and transit mode share would be much higher *everywhere*.

         To be clear, things would be better if the areas around the stations had more density and were bigger destinations. But even when you do that, you still don’t get the ridership that you do with areas that are closer together. Proximity matters.

      10. @Engineer. You’re argument makes more sense, but it is ridiculously presumptuous. You are basically suggesting that a brand new city will be born (or set of cities) with neighborhoods centered around each station, despite the fact that most of them sit next to the freeway. Places like Star Lake (https://goo.gl/maps/S9TnKxRExB92) aren’t likely to be major destinations, nor are the surrounding neighborhoods about to have the density of Ballard, let alone Belltown. There are plenty of people who live in the greater South Sound area, but they live along different corridors, too diverse to serve well with one (or even two) rail lines. None of them are even close in terms of population density to make that make sense. The area is just too small on its own to justify rail and too far away from the big city to justify anything more than express buses and commuter rail.

      11. Engineer, I don’t think anyone agrees that only Seattle should pay for Seattle segments in the future. If that is going to be the case, sales taxes raised in Seattle can stay in Seattle, instead of flowing to areas of need on the other side of the mountains. While we’re at it, federal income taxes raised in wealthier parts of the country can stay in those areas too, instead of flowing to New Mexico and Alabama.

          See where this is going? For some bizarre reason, everyone agrees that taxes should flow to the areas most in need, except when it comes to transit. You are basically making the case that zero state dollars should ever pay for transit. That zero state dollars should ever pay for education and that schools should be funded solely through levies raised in their respective districts.

          Why the two standards? The need for expensive capital projects is in Seattle, and the region can chip in to pay for them the same as we all chip in to pay for stuff other people need.

      12. “The biggest BART “failure” is the Oakland Airport Connector”

          The biggest BART failure is not serving northwest or northeast San Francisco. The Geary corridor is one of the densest in the region, and BART would have served northeast SF if Marin County hadn’t opted out of the district.

      13. “The need for expensive capital projects is in Seattle, and the region can chip in to pay for them the same as we all chip in to pay for stuff other people need.”

          That’s specifically what subarea equity was designed to prevent. Maybe suburbanites in the 2020s or 2030s will be more open to it than they were in the 1990s, but I don’t see any sign of that yet. Instead I see a movement of “Everett and Tacoma must be served now!” and “We’re thinking of opting out of ST’s taxes, especially car tabs.”

      14. Sorry, to pop your bubble, but BART is squeeze-only during the peak hours and it runs 10 75-foot car trains every two minutes during those peaks. It carries something like 35,000 people per hour during the peaks. It’s hardly a failure.

          Is it empty off-peak; not entirely, but yes mostly. Does it get most of its ridership in the San Francisco stations? Absolutely. But the vast majority of riders aren’t riding between San Francisco stations. They’re boarding at a San Francisco station to return to one of the other 60 some suburban stations. (Well, 50 suburban, 10 other urban).

      15. Yeah, let’s suppose just that you do. How do you like living in Everett? It’s pretty cheap, at least for now, huh? How many of your neighbors travel with you on your commute?

          Ummm, yeah. Zero, huh?

          Your hypotheticals are beguiling, but they are almost laughably improbable. The fact is that the people transit will attract in a place like Greater Puget Sound is the repetitive home-to-work-to-home trip. There are only so many of us transit nutters who will waste the time it takes — even in your ideal DesignedByRoss® network.

      16. RapidRider, Ross and his acolytes think that the stairs, escalators, and elevators required by subway stations are an oppressive disincentive to riders. They’ll stay on the surface and crawl along on a bus to avoid them.

          Yeah, right.

      17. despite the fact that most of them sit right next to the freeway

          Yeah, sort of like downtown Seattle, and we know how hard it is to get someone to build density down there!

      18. I think it is obvious to everyone involved that it would have been better if they had built a traditional subway system with traditional stop spacing.

          That’s certainly academically true, but it ignores completely that, like Seattle, San Francisco would never have gotten the taxing authority to build DART (“Dense Area Rapid Transit”) from the State legislature. And in any case, San Francisco could not have paid for the system alone or even with just Alameda County.

      19. huskytbone,
          Regarding tax flow and my thoughts on Seattle supporting their own transit, ask me about any other issue (highways, economic development, etc), and you’ll get the same answer from me.

          Yes, let’s please stop the flow of income tax from Washington (or New York or California) into the failing economies of Alabama, Ohio, or Iowa. Their cost of living is low, so they shouldn’t need a handout anyways. The only “national” needs I see for our Federal taxes are things like Department of Defense, Social Security & Medicare (while they continue to be funded), and Department of State. Since a certain political party has gutted almost every social safety net for human beings, all that is left is corporate golden parachutes, so let’s just cut all of it. Washington can fund Washington, and “middle America” can just fail or thrive on their own – their choice.

          That being said, for a lot of us in smaller Washington cities, we would be just fine not having Seattle nearby. Seattle brings in a lot of money, but it’s a two sided coin. We have lots of new fancy towers in a downtown I can’t afford, but we’ve also driven lots of family owned businesses to failure because their rent went up. We’ve driven lots of working class families from being able to live a few miles from work to enduring 3 hours of daily commuting. We’re now seeing Seattlites show up to bid up housing prices as far away as University Place and northern Snohomish County. Dear, God, please make it stop. I want to have a single mom or a firefighter or a teacher living in my neighborhood. I don’t want to live in a neighborhood made up exclusively of business executives and programmers. And, while these folks show up and drive up our housing prices, I also don’t see Bezos, Gates, Schultz, or the other moneybags showing up in Smokey Point, UP, Tacoma, Parkland, or Marysville, offering charitable programs to improve our quality of life. Those are reserved for Seattle, or for overseas places that make good press. Sorry to be so cynical, but I’ve seen the massive influx of money and people hurt a lot of hard working people who don’t earn big salaries, but are absolutely critical to our economy and our communities. Maybe my mind would be changed if Amazon created an employee clinic and paid the nurses and technicians twice what they can earn in the nearest hospital. Maybe I’d think differently if I knew that the janitors at Expedia and food service workers in Microsoft’s cafeteria were earning enough to cover rent on an apartment in Capitol Hill. That isn’t happening though.

          So, yeah, if Amazon, Microsoft, Expedia, Boeing, Starbucks, and UW want strong high capacity transit to serve their employees and clientele, I do believe that they should be on the hook to pay for it, not the people in communities who they have driven out. Again, sorry for the cynicism, but hopefully you now can understand where that comes from. I’ll happily pay for transit, as long as my community gets a fair proportion of the $$$$.

      20. But the vast majority of riders aren’t riding between San Francisco stations. They’re boarding at a San Francisco station to return to one of the other 60 some suburban stations. (Well, 50 suburban, 10 other urban).

          Holy cow man, didn’t you read the article you are now commenting on? Here, let me quote it for you:

          BART has comparable track miles to the fully built-out ST3 network, and a continuing program of suburban expansions. But just nine stations between San Francisco and Berkeley account for half of all rider on/offs.

          Ridership within the system is *not* driven by the suburban riders, despite the thing being designed to maximize service there. It is being driven by folks *within the urban core* of San Fransisco/Berkeley/Oakland. You could get rid of *most* of the stations (meaning the suburban ones) and still have over half your riders.

          Meanwhile, now are busy building straw men. I never said that San Fransisco should do it alone, nor did I dismiss Oakland and Berkeley. I clearly, repeatedly, wrote that they are part of the urban core. That is what Dan wrote, and it is what I wrote.

          By the way I also mentioned the value of a solid terminus serving as a suburban intercept and the role of (affordable) commuter rail and express bus service. But my point is that the suburban oriented BART experiment failed, because you do *not* have massive ridership from the suburbs, despite the system being oriented towards it.

          “I think it is obvious to everyone involved that it would have been better if they had built a traditional subway system with traditional stop spacing.”

          That’s certainly academically true, but it ignores completely that, like Seattle, San Francisco would never have gotten the taxing authority to build DART (“Dense Area Rapid Transit”) from the State legislature. And in any case, San Francisco could not have paid for the system alone or even with just Alameda County.

          What the hell is academically true? Either it is true or it isn’t. Either it would have been better to build a traditional subway system or BART. I guess you finally agree with me that it would have been better to build a traditional subway system, but you think it would have been impossible to do so for political reasons.

          First of all, that was never my point! I never discussed the politics — I was merely making the case that the BART experiment (building a commuter oriented subway system instead of a traditional one) was a failure. It doesn’t get the kind of ridership that they expected. You are basically arguing that the Iraq War was inevitable for political reasons, while I am busy explaining why it was stupid, and why we should avoid similar conflicts in the future.

          I have no idea what the political situation was in the Bay Area before, nor do I know it now. It is quite possible that it is politically impossible to build the most cost effective system, because doing so would appear to benefit the urban core more than suburban voters. But I do know that it is a crazy stretch to think that a more cost effective system couldn’t be built in Seattle. It wouldn’t have taken anything more than sensible leadership. If I was on the board one of the first things I would do is hire transit consultants. I would ask them to come up with what they consider to be the most cost effective transit system for the region. I would get a different set of opinions, because one group might favor one approach over the other. Then I would spend lots of time, in meetings, while we discussed the alternatives.

          I am quite confident they would have proposed exactly what I’m talking about. Rail, serving the urban areas of the city (complete with good suburban oriented bus intercepts) along with buses in the suburbs. They wouldn’t have come up with the crap that is ST3. Would that have been impossible to pass? I doubt it.

      21. “Federal Way Link isn’t about Seattle. It IS about Federal Way, Highline, Tacoma, Fife, and Seatac.”

          Do the residents of Federal Way, Fife, and Tacoma know this? Or is this just the hope of their governments?

          I can see 200th to Tacoma Dome being potentially useful for local trips, because Swift does the same thing on lowish-density 99 and it’s somehow the highest-ridership route in the county. It all depends on how much housing and business the cities allow around the staitons, and whether these businesses and amenities offer sufficiently unique attractions to make people take Link there rather than busing or driving somewhere else or going without. On 99 it can also serve as a faster alternative to the A, again like Swift, at least between the areas that have stations. That could be more or less useful depending on what exists around the stations.

      22. Ross – you cannot ignore politics when analyzing transportation infrastructure. Transit doesn’t succeed or fail in a vacuum; particularly for rail, local land use is critical.

          It’s very easy to look at the BART network and imagine a slightly different political environment where suburban BART stations look like suburban Vancouver.

          Again, the problem with BART isn’t BART. It’s suburban land use. ST3 is, fundamentally, a bet on Metro Seattle in 2040 looking more like Metro Vancouver and less like the Bay Area.

   4. Nah, there should be an ST4, and it should be a “big 4” ST4, building nearly everything in Seattle, Tacoma, Bellevue, and Everett, and hardly anything anywhere else. Seattle’s demand is obvious, and the other three will clearly have the demand in the future, but they need a network for their own communities. Tacoma, in particular, only has Link going to Seattle in ST3, and nowhere in Tacoma itself.

      1. Not true, Tacoma’s Freighthouse station will soon be a very busy hub.

         There is the Hilltop Extension currently under construction. There is the TCC Link and Tacoma Dome extensions.

         Add in the new Sounder and Cascade runs and I think Tacoma will be maxed out for short and long term.

      2. Isn’t the Hilltop Extension part of ST2?

      3. ST3 covers the extension from MLK and 19th to Tacoma Community College via 19th Ave. Hilltop Extension is an ST2 project.

         There’s also ST3 money for a planning study of an extension from Tacoma Dome to the Tacoma Mall area which is being upzoned and redeveloped into a walkable community.

      4. but they [Tacoma, Bellevue, and Everett] need a network for their own communities.

         No they don’t. Everett and Tacoma are very low density cities, and the only high density places in Bellevue are covered well with East Link. Small, low density cities don’t need rail. A city like Spokane does not need a light rail system — it needs good bus service.

         Community Transit has the right idea. A series of high end buses (Swift) along with an improved overall bus system is the best you can hope for. Holy cow, most of the buses in Snohomish County run every half hour, or worse, and you are suggesting that they build a train network? They will never need that kind of capacity. If they built it, the poor folks in those areas would be burdened with the cost, and left with a system that makes Sacramento light rail look great (Sacramento RT has frequency ranging from 15-30 minutes during the day and 30 minutes at night).

      5. Community Transit really does have the right idea. A limited-stop major-corridor network that puts the Eastside and South King County to shame, and premium fares on its long-distance express routes. There’s even a fare difference between express routes that go to Lynnwood and express routes that go beyond it. And it’s trying to improve the frequency of its local service, within the funds it has available.

      6. RossB: “No they don’t. Everett and Tacoma are very low density cities” okay, have you actually visited Tacoma or are you basing this on presumptions. I actually live in Tacoma and it’s a fairly dense city. We actually have a density of 4,000 people per sq ml. Which is pretty similar density wise to Bellevue. Along with the fact that I’ve seen Tacoma actually do a lot of work in upzoning the city to actually be a lot denser. And will likely see a lot more upzoning as the streetcar extension and Tacoma Dome extension are done. Tacoma isn’t some podunk suburb anymore, it has actually worked hard to change its image (in particular after Russell Investments left) and it will change as said infrastructure is built.
         You keep professing that Tacoma isn’t deserving of high quality light rail transit, when there is a an actual need for it. “The bus is faster” maybe, but when you add in frequency to the equation there is a case for time being saved using link. The bus gets stuck in traffic, at lights, and getting to the transit center can add time and that can add up pretty quickly. Link would also add needed frequency between Tacoma and Seatac, which is what the intention of light rail extension is for. Seattle to Tacoma is a secondary priority when it’s being well served by Sounder during peak.

      7. “Tacoma actually do a lot of work in upzoning the city to actually be a lot denser.”

         Funny, what I see on I street just north of downtown is just single-family houses, and I see alternating pockets of them along Pacific Street that we’re upgrading to RapidRide, including the part closest to Tacoma Dome. Will areas like these be upzoned?

      8. Yes, Zachary, I’ve been to Tacoma. I have friends and relatives who live there. Many of my close friends grew up in Tacoma. I like Tacoma. It is a nice city. It just doesn’t have much in the way of population density.

         Here, look at a census map: https://arcg.is/1uyGWH. Tacoma is growing, of course, but at numbers that don’t alter the dynamic in any significant manner. There are no census blocks of over 25,000 people per square mile (ppsm). In Pierce County I count 17 blocks of over 10,000 ppsm. Their is really only one significant cluster, which happens to be close to downtown. Now zoom into Seattle (https://arcg.is/1zyHve). There are clusters comparable to the highest density areas of Tacoma in most of the city. The entire north end of Seattle is pretty much like this, as is all of Queen Anne, as well as areas like Bitter Lake and Lake City. There are also places with much higher density, like the greater Central Area (which includes Capitol Hill) and Belltown. Most of these places will never have light rail. Because even with the growth that has occurred — which has been much bigger in Seattle than in Tacoma — most of the city is not like Belltown. It just doesn’t have the density, and running subway systems is extremely expensive.

         Nor does Tacoma have an unusually high concentration of jobs in one place. If you look at the numbers (https://onthemap.ces.census.gov/)*, then you can see that while downtown has the highest concentration of jobs, there are plenty outside of there. Jobs are pretty spread out (it isn’t like Calgary). I’m afraid that it has neither the high concentrations of people (in more than one place) nor the concentrations of jobs (in one place) to make a light rail system work. Tacoma needs to focus its efforts on better bus service.

         * It takes some effort to see the job concentrations using that map (and there is no way to link to a particular map). If you search for Pierce County and then select it from the list and go with the defaults, it should provide you with a map. I found the Point Overlay option to be too cluttered (unless you zoom in a lot), while the Thermal Overlay to be a great snapshot of job concentration.

      9. “There are also places with much higher density, like the greater Central Area (which includes Capitol Hill) and Belltown. Most of these places will never have light rail.”

         There’s another factor. Areas like Queen Anne and Greenwood lose because there’s a higher-priority corridor in the same city. So it’s not just absolute density but relative to the other corridors in the city. Since Tacoma has no Central Link and only a stub Tacoma Link, the issue is where to put the first line. And I suppose Tacoma Link is in the densest part of Tacoma and Tacoma Dome is adjacent to the densest part. Whereas in Greenwood’s case — and Ballard and West Seattle before ST3 — they lost because they weren’t the highest-demand and biggest-bottleneck corridors in the city.

   5. I think the value of the ridership models is primarily is in comparing two or more competing projects using some kind of math (or “mathiness” to use a pejorative). Decisions have to be made without regard to how complete/incomplete the information we have is. Ridership forecasting models aren’t known for being particularly reliable. They’re always “improving,” but social sciences will never be accurate like STEM fields.

      If Puget Sound communities build new walkable communities and industry around the stations. Then they’ll likely blow these estimate away. If they bow to NIMBY pressure, then we might end up running “trains to nowhere.” That being said, it’s not all bad running a train that isn’t jam packed.

   6. ST4 is going to need to fix mistakes made in ST3. Things like adding capacity (longer platforms at stations, for example), putting in the Duwamish bypass or rebuilding the MLK segment (South King will want this), building “ribs” to the “spine” (Ballard-UW is the biggest but others could emerge including walkways and connecting driverless shuttle methods to big attractions), fixing operational design flaws and modernizing existing stations and systems . We also will need more money to complete ST3 because the budgets and contingencies are woefully inadequate (ST way overpromised what ST3 money can build).

      I think we really need a few years with the expanded system to hone in on the best investments for ST4. We need to live with the new reality a bit to get better clarity on what’s needed. Visionaries can identify problems and opportunities early but voters generally need to experience them. 2024-2025 will be a revealing, reality-changing year.

      1. Whatever we can achieve with longer platforms, I’m pretty sure we can accomplish orders of magnitude more cheaply and decades faster with open gangways, slightly longer trains where the first and last door are at the two ends of the platform, and less seating. Adding more platforms on the other side of the train would not only be easier than lengthening stations, but also reduce dwell time, minimum headway, and overall travel time and trip time including waiting time.

         Fixing the other elements that set minimum headway will also still be far cheaper and more useful than lengthening all the underground platforms.

      2. Brent, you’re right. It’s an inexcusable error on ST’s part to have ordered their next batch of trains with two driver cabs in each car. Imagine 7 expensive, unused, space-hogging driver cabs in each and every consist for years to come. Sometimes ST really makes stupid decisions.

      3. I’d agree that open gangways (fewer driver cabs) would enhance capacity as well as be potentially justified based on forecasts showing possible overcrowding. They also let riders move between cars, which also eases overcrowding.

         I’m also surprised that “paired vehicles” (cabs at just one end of each car and an open gangway at the other) weren’t procured.

         Will more capacity be needed? It may take the Ravenna area riders possibly griping about overcrowding after 2025 to finally force the issue with the seemingly “un-visionary” ST people making decisions. Why else would a deep transfer station (400+ steps) at ID still even be on the table?

      4. @Al — Some good ideas but …

         Bigger Platforms? What makes you think we need bigger platforms when our platforms are huge and ST plans on running trains every six minutes? Just run the trains more often.

         Duwamish Bypass? As this article clearly states, there just aren’t that many people riding in from the suburbs –certainly not enough to justify an entire new line that will only have one extra stop in Seattle (and a very weak stop at that). That sort of thinking reminds of folks that want BART to have express trains that skip stops (missing the entire problem with BART).

         Modernizing existing stations? If anything, our stations are overbuilt. In most cases we would do just fine with a hole in the ground, but instead build gigantic structures (my favorite is UW station, which has a deep tunnel platform, but a really big above ground structure leading to an overpass).

      5. Look a little closer at the map in this article, Ross.

         The Rainier Valley segment at 24m is the second highest non-Downtown segment in the system. The highest one is south of Northgate at 33m — and it’s going to have twice as many trains. It’s even forecast at 20m from SeaTac. Meanwhile, ST says that they will run twice as many trains to Lynnwood as they do the Rainier Valley, which is showing just 15m here.

         Look too at the bullet points. Westlake to Capitol Hill (twice as many trains) at 146,000 with twice as many trains as SODO to Beacon Hill at 86,000. That means that on a per-train basis, the SODO to Beacon Hill segment will be more crowded than Westlake to Capitol Hill is.

         In sum, the map and data both demonstrate that the potentially most crowded segment of Link is through the Rainier Valley.

         It’s pretty clear from the diagram and data here that any capacity increases outside of Downtown should go first to address the Rainier Valley. There is no way that ST can run three-minute trains through the Rainier Valley without a substantial and expensive and disruptive grade separation project — which quickly leads to the need to create the Duwamish Bypass.
         .

      6. First of all, none of that is anywhere near where crowding is a problem. Secondly, if crowding does become a really big problem, it will be because Rainier Valley (by the stations) has become very big. Increased ridership is not going to come from huge numbers of people coming from Tacoma (there are only so many people who will put up with three hour commutes). Third, it would be much cheaper to just cut and cover MLK than it would be to build a bypass. Doing so would increase the headways to three minutes or better (most of Link is designed for 90 second headways).

      7. You’re pretty naive about cut and cover costs, Ross. Most rules of thumbs are that subway tracks are 10 times more expensive than surface tracks and 3 times more than aerial tracks.

         Just look at how much more expensive the West Seattle extension would be if the last segment to Alaska Junction is in a tunnel.

         Surely a capital cost estimate would be needed, but it is pretty much a certainty that a Duwamish Bypass would be a heck of a lot cheaper to build than an MLK subway. I’d even suggest looking at a Rainier Avenue subway segment if we were going to tunnel because we would better serve SE Seattle.

   7. 15th Ave NW to Aurora Ave. up to Shoreline, and something to Renton, might be candidates for ST4. Metro 8 subway for Seattle, for sure. Perhaps even an express downtown – airport line with a station in Georgetown? If I-405 (pseudo) BRT proves popular, perhaps an expanded network of “premium express bus” service, or even actual BRT, might be in order. Perhaps have it start when ST2 taxes end to help appease the “no additional taxes” crowd.

   8. In year 2200, how do you know Snoqualmie Falls won’t be major manufacturing center- with Northwest Washington’s most important station on the Point Barrow-Tierra del Fuego supersonic line following fortuitous bed-rock underlying the Cascades from almost pole to pole? Maybe the line will share tubes with ore trains carrying the new uranium of the future.

      Remember how densely promising Seattle looked to the miserable travel-weary seasick shipload of new residents as they stood on deck in the pouring rain listening to their real estate promoter tell them that the reason he just named their new home “Alki” because it was the Indians’ way of saying it’d be just like New York City “In A Little While”.

      Still waiting for the corned beef sandwich, I mean Cleveland’s got them at Shaker Heights light rail station, but guess we gotta be patient. So while wishing you a long string of elections where you can vote negative, absolute number of ST-‘s is, thankfully, somebody else’s call.

      Mark

5. I wonder what the breakdown will be for the 10 East Link stations to the UW? At what point north will riders deem it better to take 520?

   1. Even from Downtown Bellevue the 271 only takes about 20 minutes to UW, while Link is projected to take 30. With the 541/542 expresses from Redmond/Overlake it’s no contest, especially once the Montlake HOV improvements are made.

      The only exception to this is going eastbound into Overlake in the mornings or into Redmond in the evenings, when buses consistently get stuck in heavy offramp traffic.

      1. So basically the last 3, ie, South Bellevue, Mercer Island and Judkins Park, could use Link to get to UW leaving 7 stations primarily for trips to downtown Seattle and within the east side area.

         I’m trying to figure out what the demand will be for a 520 or SandPoint Link to UW and beyond to Ballard. It looks viable.

      2. les, “No, it doesn’t.” And the folks surrounding the north end of the Lake would be up in arms.

      3. There’s no need for another Link crossing of Lake Washington. The express buses work fine. There is a greater need for light rail along the 372 route.

      4. Definitely 372, it would be my top route for ST4. But it would be interesting to see what the volume of bus riders across 520 to UW is. Otherwise I don’t see a justification for a Ballard to UW line, seems like too short of a line with low ridership for the buck.

      5. As noted elsewhere, it’s dangerous to use bus ridership to project light rail ridership, especially if the current buses (e.g. the 44) are slow and unreliable. It also doesn’t make sense to put light rail on 520 where there is a nice network of rapid bus lanes already available with major improvements coming to the Montlake area.

         People will want to travel between areas of population density (or proposed population density) and to major employment centers. Making those connections fast and reliable will induce demand.

      6. What Joe said.

         There are three big factors to building good mass transit — density, proximity, and speed relative to other alternatives. Ballard to UW scores big on all three. Kirkland to UW does not.

   2. “I wonder what the breakdown will be for the 10 East Link stations to the UW? At what point north will riders deem it better to take 520?”

      It depends. In general, I think the bus would win by about 20 minutes (for Bellevue) and 30 minutes (for Redmond), unless you’re already on a Link train when you get to the bus stop. But, if the Redmond buses get subjected to a South Kirkland P&R detour, as outlined in Bellevue’s master plan the bus edge over Link for Redmond riders drops from 30 minutes to about 20.

      There’s also a question of whether the bus option will even persist long term, or whether in an effort to save money, Metro and Sound Transit will try to force people traveling during off-peak hours to ride the train all the way around, so they don’t have to pay to run the bus. Rush hour, the bus is almost certainly safe. But, it is not clear yet what happens in 2030 at 9 PM on a Saturday night. I hope the bus will still be running then.

      But, at the end of the day, all of the Eastside->UW buses don’t actually increase the set of destinations you can get to on the transit system, given unlimited time (except for very-low ridership neighborhoods, like Medina and Yarrow Point) – they’re just shortcuts that get you between the same places the train does, but in less time. Which means, at the end of the day, if budgets get tight, they can become expendable, especially during hours when ridership is not super-high.

6. Crowding analysis should probably be focused on peak direction at peak times and not daily. It’s certainly possible that a low ridership segment could be mostly empty except during commutes. That’s the case with BART, where westbound trains are full by Walnut Creek or Pleasant Hill in the morning commute.

   I think the higher segments are picking up more non-commute trips, which puts more riders on trains at off-peak times. I don’t think it’s going to be proportional between high demand and low demand segments.

   1. THIS! Only between the Ship Canal and Dearborn will Seattle ever be a real “city” in the sense of all-day subway demand. Well, unless the City Council and Mayor get high behind and say “If you live within three blocks of a Link station you can own your home but it will have to be in the form of a condominium of homes at least twenty stories high.”

      Everybody else will be renting in similar buildings.

      1. Northgate – U District will never be another 3rd Avenue downtown, but it’s nothing to laugh at. Especially if the UW innovation district and the redevelopment of Northgate happen.

      2. The redevelopment doesn’t look too promising. The Northgate Mall lot is the only one zoned for 20-story buildings, and the last mall plan I saw only used three or four stories of it. The other lots will probably get built the same size as the other recent buildings. So most of Northgate’s growth is already there unless somebody steps up to improve it.

7. Sound Transit anticipates trains at six minute headways on each line will be sufficient to manage demand through 2040.

   The vital contribution of ST3 is core capacity in Downtown and South Lake Union.

   So, basically we are adding capacity but ST doesn’t think we need it until 2040 (if not later). Keep in mind, (according to ST) we could triple the capacity through the downtown core by running the trains every two minutes.

   1. Could you please link to the ST document that says we could run trains through the downtown tunnel every two minutes?

      1. From https://seattletransitblog.wpcomstaging.com/2015/03/21/capacity-limitations-of-link/:

         Although going below 3 minutes is possible, due to the variability inherent with human factors and surface operations it “wouldn’t give our ridership as reliable a service.” The small windows to fit in delayed trains might cause them to bunch up, delaying riders. Furthermore, it would likely require additional investment in Traction Power Substations.

         In other words, it might cost a bit more to get two minute headways and it might result in some bunching. Realistically, you wouldn’t need to have more trains from Everett to Seattle, but from Northgate to SoDo (which would be fairly easy to manage without bunching).

      2. Also from the same article (I should have started with this):

         As it turns out, Sound Transit’s signaling system is designed for a minimum 90 second headways.

         So, I stand corrected. We should be able to run trains every 90 seconds, not every two minutes.

      3. ST says the DSTT could accommodate 90-second trains with capital improvements. A proposal to support at least 2-minute service was in the list of potential ST3 projects. When ST decided to go with the second tunnel, it decided not to do the other project, at least not in ST3.

   2. “we are adding capacity but ST doesn’t think we need it until 2040”

      It takes ten years to plan and build an underground line, so even if we started today it wouldn’t open until 2029. If 2040 is the target, that means the latest we could start it is 2030. There’s only an 11-year difference between starting it now and starting it then, and only a 4-year difference between its planned opening date and 2040 (I think it’s scheduled for 2036). Given the three-year process to update the long-range plan and put together a package and schedule a vote (and get legislative approval if necessary), 11 years is not a lot of time. We should have a long-range plan for an Aurora line, and move on it fifteen years before we need it.

8. The estimates showing ridership fall off outside the core give me reason to believe the one valid-sounding argument I’ve heard for BART-style distance-based fares on Link may turn out to be false.

   Namely, some distance-based fare advocates believe ST will get more fare revenue that way than with flat fares.

   It could still, but I think we’d have to ditch the discount over local bus service that is supposed to encourage riders to choose trains over buses. I don’t think the discount really does that, except maybe in very rare cases. If it does have an impact, then there is something wrong with Link.

   The short-distance fare discounts could very well wipe out the marginal higher fares from the fewer suburban riders, and in the process push those suburban riders to push harder for keeping express buses because of the substantially cheaper bus fare. That last part could make distance-based fares a big money-loser. There is also the detail that RRFP fares will forever be only half of the lowest distance-based fare.

   I’m still betting that the suburban-dominated ST Board will move to flat fares somewhere along the way, in part because it will turn out to make fiscal sense, but also because social justice for the less-rich suburbs, and maximizing ST’s share of Business Passport revenue.

   And then there is the whining from those with passes that don’t completely cover the longest-possible trip. As Link grows, that will include more and more Seattleites heading outbound from downtown who have no intention of going all the way to Everett or Tacoma, but will get warned or fined when they forget to tap, even though the pass on their ORCA card covers the trip they were actually taking. Flat fares would enable FEOs to just remind them to tap, and keep a count, rather than read the forgetful passholder the riot act. Not that ST would take the opportunity to help themselves by letting passholders get a pass, but at least flat fares would enable them to have such a policy improvement.

   I’m glad I will never have to throw in the utter ridiculousness of double-tappers getting warned or fined, ever again. Thanks for the two beeps!

   1. I have my doubts about the Sound Transit ridership forecasts for a number of reasons:

      1. FTA requires that land use plans be fixed. It’s obvious to us that new TOD is attractive but it’s often not allowed to be considered. I’m not sure what the forecasts assume for Spring District, Northgate or Lynnwood but surely the most recent plans aren’t included. Even the infill at places like Columbia City, Capitol Hill and Roosevelt may not be fully included. Looking to the future, I expect the market to create unanticipated building booms at Federal Way and Smith Cove — and surely SLU is under-forecasted.

      2. The models are based on travel behaviors based on surveys from either 2010 or 2000 (also an FTA requirement). Those years (especially 2000) are almost in the horse and buggy days when it comes to transit riders. We didn’t have Uber/Lyft, widespread texting and messaging and bus arrivals on most of our phones in 2000. Places like BART are now showing as many riders getting dropped off as are those parking their cars at distant stations with big lots.

      3. These models assume feeder buses and these assumptions are often hidden. How buses serve Link will affect demand. What was the 520 assumption for example?

      So why the patterns are relevant, I wouldn’t think of the forecasts as worthy of debating too closely.

      1. Predicting life in 2040 is science fiction.

         Yet, ridership forecasts extend out the status quo.

         The reality is that no one knows how many people will live where in 2040 or how they will transport themselves.

         The official answer is to project population based on historic growth rates mixed with official zoning plans, and to assume transportation methods remain static.

         The real answer would be a wide range of outcomes within a probability threshold, based on radically divergent settlement and transportation scenarios.

      2. What the FTA requires is approved zoning plans, or at least close to approval, not speculations on possible future zoning. Seattle and the other cities were very late in the game in upzoning. That lost Lynnwood and Federal Way the opportunity to get some of the real-estate bubble development in the mid 2000s. And for Lynnwood Link it skewed the ridership potential of the Aurora alternative and 130th Station downward. Aurora had almost as many riders as I-5. It lost because it was 4 minutes longer, and ST estimated it would lose more riders in Lynnwood than it would gain on Aurora. But if Aurora had seven-story apartments and walkable retail destinations along large parts of it like Dexter Avenue does or the Spring District will, its ridership would go through the roof and swallow I-5. But that zoning wasn’t considered in time for the ST2 decision. (And it still hasn’t been considered yet. And even a less-ambitious zoning like Shoreline’s or Lynnwood’s urban villages is still lacking. You know your city is a doofus when Shoreline surpasses it, as it has done with RapidRide BAT lanes and urban villages on 99.)

      3. If you build a new highway now, developers will build sprawl around it by 2040. Similarly, if you build a transit station now, developers will build densely near it by 2040. (Assuming upzones happen, of course!). Both highways and mass transit have historically been built in rural areas on the periphery of cities, and guess what, those areas didn’t stay rural areas. In either case, assuming the status quo is stupid at best and blatant transit bias at worst.

9. “The Kirkland-Issaquah line, where crowding is not a concern…”

   Very subtle. This route perplexes me quite a bit. You can still accomplish what the 555 and 556 do by interlining Link with DT Seattle and asking the Bellevue riders to xfer at Mercer Island or a Factoria stop. And by interlining with DT Seattle you wouldn’t be inconveniencing all those Seattle commuters and airport trips by forcing a ridiculous xfer all the way up in DT Bellevue.

   Another head scratcher!

   1. It’s so far away that I expect the operations plan could change.

      The Issaquah part could be handled with a self-contained, driverless single track system (double track at end stations with a bypass double-track halfway) from either East Main or South Bellevue. As driverless, the frequencies could be high and a train with open doors could wait at the transfer station for the transferring riders from either Bellevue or Seattle.

      The problem with this is the one-station South Kirkland end. It’s notable how low the demand is for that station in the forecast. It really makes me wonder if it’s needed for anything other than political reasons. A new Sprint-branded line from Bellevue to South Kirkland to UW would seem to do better.

   2. Bellevue College will be a sleeper hit on the Issaquah line.

      The South Kirkland terminus is the result of a broken political process. ST couldn’t get consensus on a route through Kirkland so it terminated it at the furthest point before the controversy. The P&R and on-site (affordable?) housing also make it look like an attractive terminus, although the case for P&R riders is dubious. Will people really drive from Kirkland to the P&R and take Link two miles to downtown Bellevue, or take Link to Bellevue and transfer to Seattle? Not if there’s a 520 bus from the P&R, as is pretty certain with the 255’s plans.

      1. Continuing to South Kirkland P&R has some advantages. For Kirkland riders, you have more transfer options than you do from Bellevue, making for more frequent service. You’re close enough to DT Kirkland that a bike connection could take as little as 10 minutes, with no waiting. It would enable commuters from Kirkland to Bellevue College will be able to get off the bus and onto the train sooner, reducing exposure to traffic. And people could drive and park at the station to avoid having to park in DT Bellevue. Right now, it feels hardly worth the bother, since parking is generally free for most employees and customers, but by 2050, that could change. 30 years is a long time.

         It is also worth noting that once you’re already building the line from Issaquah to Bellevue, extending it to South Kirkland P&R doesn’t really cost all that much. It’s just surface track, and the grade-separated underpasses for all of the freeway crossings are already there. It’s really just a matter of laying down the tracks, clearing space for the trail alongside, and building a station platform.

10. “Perhaps Snohomish County should have prioritized an early extension serving commuters from Ash Way and Mariner, rather than waiting to open those stations along with lower demand markets at Paine Field and Everett.”

    “Snohomish County” did just that. The 2007 ballot measure (which failed) included two more stations with Ash Way as the northern terminus. Sound Transit stripped those two stations out of the revised plan in 2008’s ST2.

    1. OK, but that was a small part of a large three-county measure, and I think it was the highway-and-transit package, and it was 12 years ago, and a significant chunk of current residents didn’t live here ten years ago, and I-5 congestion has gotten worse, and the large tech employers in SLU weren’t there then. I agree with the article that ST should have considered two phases, and I disagree with the idea that the 2007 vote means Snohomish is definitively opposed to it. I think what the Snohomish government would say is it wants Everett and Paine Field as soon as possible, and doesn’t want anything to postpone their opening. So as long as an interim terminus doesn’t postpone them much, I don’t think they would be categorically opposed. I think the reluctance to do it probably stems from the fear that even if ST says it won’t lead to delaying Everett, they’re not sure if they can believe it. That’s similar to the Ballard-downtown issue, where people are hesitant to accept that a Ballard-UW-downtown trip wouldn’t take longer than a directBallard-downtown line.

       1. You missed my point entirely. My comment was meant to contradict the OP’s little bit of historical revisionism. SnoCo actually pushed for a northern terminus farther northward than the Lynnwood TC as part of the planning for the orginal ST2 plan (2007). Sound Transit unwisely stripped out the Ash Way and Alderwood stations in the revised proposal that was put before voters in 2008.

          Additionally, Snohomish County has reason to be skeptical of ST keeping its committments to the subarea. There’s a history here…..

          The following is an excerpt from Motion No. 1 passed by the ST Board on Dec 2, 1994:

          “NOW, THEREFORE IT IS MOVED BY THE BOARD OF DIRECTORS OF THE REGIONAL TRANSIT AUTHORITY AS FOLLOWS:
          The Regional Transit Authority Board intends that, in order to complete the planned connections to the four major centers, extension of light rail service to Everett shall be a first priority in Phase II.”

11. “West Seattle Link will carry 32,000 over the West Seattle bridge”

    I think a relatively cheap extension to White Center and on to Burien would do this segment justice and help justify the West Seattle expense.

    1. How is going to White Center from the Alaska Junction ever going to be “relatively cheap”? I guess if the elevated station between 42nd and California is chosen and somehow the overhead “wiggles” over to California it might be. But, boy, would there be Hell to pay when such a line was proposed. Whoooey!

       1. Your right, it won’t be as cheap as I thought. The study from years back has 3 billion..yikes:

          “extending elevated light rail from West Seattle Junction to Burien Transit Center along a jagged course. Its estimated cost would be between $2.7 billion and $2.9 billion for a projected ridership between 10,000 and 15,000. The train would cover the 9.1 miles in 22 minutes and take 12 minutes more to Downtown. The plan calls for six stations at Burien Transit Center, SW 128th St, White Center, Westwood Village, High Point, and Morgan Junction.”

       2. West Seattle Link will never connect to White Center. White Center will be served by a bus transfer at the Delridge Station. The Junction station will probably be along 41st Ave which is 25 blocks west of White Center.

       3. The equivalent comparison would be putting a Link station in Madrona and then coming back to Broadway.

       4. Joe, have you ever been to White Center? You’re acting like it’s right next to the Junction, when it’s almost 4 miles south of it. Between it and Burien there will be plenty of justification for an eventual southern extension if West Seattle isn’t built to preclude one.

       5. I have been on every street in that area. It makes zero sense. The way to get to White Center is on a new line via Georgetown and South Park, not by spending 3 billion dollars blasting through hundreds of residential properties and several steep hills into a land of zero density.

12. 1) I really question these projections. As more and more souls are priced out of the cities; as light rail becomes more and more popular; and as many Sound Transit bus routes are gaining ridership – I just think these projections underwhelm.

    2) As to,

    “The weak intra-suburban estimates bely the hope that ST3 would serve travel within the Eastside after ST2 had connected it to Seattle. Nor is Everett Link carrying many Everett residents to jobs at Paine Field. Snohomish County riders are mostly starting their morning trips south of Paine Field, and commuting to Seattle. Perhaps Snohomish County should have prioritized an early extension serving commuters from Ash Way and Mariner, rather than waiting to open those stations along with lower demand markets at Paine Field and Everett.”

    Part of the issue may be assuming folks will want to take squiggly Everett Transit neighborhood coverage bus routes to Paine Field. Part of the issue may be lack of proper transit connections from the light rail station to various Paine Field & vicinity destinations. Part of the issue might be underestimating demand to take light rail to Paine Field and transfer to a fast, frequent bus to the Paine Field Commercial Terminal. I just really question the math here.

    1. You are right about questioning the math, but the math is consistent with every other transit system. ST was built on the assumption that huge numbers of people were going to ride transit in ways that they don’t, anywhere. It was also built upon the assumption that Paine Field — despite all evidence to the contrary — was going to be a major, centralized employment center, and that Everett was going to grow much faster than Seattle. All of these are projections, of course, but the math is simply backing up what just about every expert anywhere would say, which is that the spine is a really stupid idea. Areas like Everett and Federal Way are too spread out (in terms of population and destinations) to create enough ridership on their own right to justify a rail line, and too far away from Seattle to justify anything but express buses and better commuter rail.

    2. The math was based on the then-approved zoning plans. Everett and Lynnwood could improve the ridership including intra-county ridership if they did something about that. Everett’s station zoning is a step forward, but it still leaves vast areas of downtown and Broadway at 1-2 story.

       Paine Field was always wishful thinking. The employers aren’t within walking distance of the station. Industrial facilities are so large that not many of them can be. Although Brooklyn manages to fit a lot of industrial space in multistory buildings on walkable streets. That’s another thing Snohomish hasn’t considered yet. But most of the motivation for the Paine Field detour was to attract employers to the county, regardless of how many workers actually ride Link or how they’d get from the station to their worksite. Because more employers means a larger tax base, and right now Bellevue and Redmond are kicking Everett’s butt.

13. Does anyone have estimates of weekday East Link Seattle to Redmond Technology Station (Microsoft), compared to a Saturday or Sunday East Link Seattle to RTS?

    1. Only my limited experience with the 545. Eastbound the bus is packed from 7am to 11am weekdays, and anemic outside that. Westbound I don’t know as much. You can see it in Metro’s schedule, buses every five minutes in the morning hours but dropping to half-hourly evenings and weekends. If you mean specifically RTC station, the proportion will remain as-is unless Microsoft starts increasing shifts at other ours. But if you look further to downtown Redmond, a different pattern emerges. Link will not only serve downtown-Redmond and Capitol Hill-Redmond, but also Bellevue-Redmond, Spring District-Redmond, and UDistrict-Redmond. And SeaTac-Redmond. All that will probably generate more weekend ridership, and it will all pass through RTC station rather than getting on/off there. That’s a great thing too.

14. Me thinks we are gonna need to keep buses running on the I-5 Express Lanes to supplement and complement Link for service from the North.

    Link won’t be able to carry everyone. Maybe we can add some freeway stations at NE 45th and Green Lake and route these buses direct into SLU, First Hill or Lower Queen Anne to give those areas one-seat rides while leaving Link to directly serve Downtown?

    1. I disagree. With 4-car trains running every 3 minutes, there will be plenty of capacity for the foreseeable future.

       If we do continue running buses down the I-5 express lanes, it only be so that people in some politically-connected neighborhood don’t have to transfer. But, even they won’t really benefit from the service. With trains running every 3 minutes, fighting downtown traffic on a bus would take far longer than transferring to Link. Already, there’s a long, sad line of buses stuck in SOV traffic at the Stewart St. exit from the I-5 express lanes every morning. It would be crazy to wait in that line when a quick and convenient transfer to Link exists as an alternative.

       1. The south end math will still favor express buses speedwise. I don’t foresee Federal Way and Tacoma electeds letting current express routes go away easily.

15. Just goes to show how train interiors are designed wrong:. Instead of maximizing seating space for regional trips with rows of seats that make exiting busy trains a difficult dance. We should have gone for subway style open interiors that minimize load / unload times in the central downtown core.

    1. The Siemens LRVs starting to arrive later this year will be much better on that score.

       But you are right that politicians didn’t think in terms of the LRVs’ layout’s impact on headway.

    2. Even the Kinkisharyo cars could be pretty easily adjusted. Just take out the two rows of two seats on one side of the low-flow section of each side of a car and replace them with three seats along the car wall. That would free up lots more floor space and speed boarding on crowded trains. That would only mean of a net loss of two seats (one on each car end) per train car.

16. Thanks Dan. All the more important that the ST3 alternatives in SODO allow for operational interlining to boost service in the new downtown tunnel. As of right now, the DSTT will have the crush loads and the higher frequency with the Red+Blue lines, while the brand new tunnel would have just the Ballard-Tacoma line at 6 minute headways. That’s a huge capacity waste of the most expensive per-mile investment in ST3. At-grade/E3 through Sodo would allow West Seattle trains to also access the new tunnel, SLU, and Ballard. Call it the Orange Line or the Silver Line or whatever. But it would give the core of the system more capacity for almost no money, no subarea equity concerns, no new capital package needed to strain regional politics. Just doubled service where it matters.

17. This is no surprise. I’ve always said the non-Seattle parts of ST3 are a total waste of money. We roughly doubled the track miles of the system with ST3 (IIRC), while adding some small fraction to the total ridership (and most of that is in Seattle, downtown).

    ST2 captured pretty much all of the valuable service areas, and I think if it had been in place before the ST3 vote, ST3 would have failed. I think people were voting for “more rail than we currently have” with ST3, not “more rail than ST2”.

    One major issue with ST3, given that it will mostly serve trips to Seattle, is that it is way, way too slow. You basically get almost the worst rush hour travel times, but you can get them any time of the day (hurray!). Train speeds should have been improved somehow, or the system should have been designed to allow for express trains (not sure how much time that would save).

    1. Thanks for saying that light rail speeds are too slow. It’s unnoticeable for trips under 4 or 5 miles like much of Seattle. It becomes increasingly frustrating after that, especially above 12 or 15 miles.

       It’s why I’ve observed that light rail is not the best technology for the spine. It’s slower and more expensive for going beyond Federal Way or Lynnwood. For the same dollars, we could have had a much longer mobile-powered rail system with additional single-track segments to many more places.

       Unfortunately, the decision has been made and it is irreversible. It’s a useless issue to discuss at this point.

       1. “It’s why I’ve observed that light rail is not the best technology for the spine.”

          Amen to that. (Actually, amen to your whole comment.) A spine extending from Everett to Tacoma using light rail was always a dumb idea. But this agency has never wanted to listen to the naysayers, tending to foolishly lump all contrarians into the same “anti-transit bucket”.

          Oh well. As you’ve stated, that (slow) train has already left the station, so to speak.

    2. “ST2 captured pretty much all of the valuable service areas, and I think if it had been in place before the ST3 vote, ST3 would have failed. I think people were voting for “more rail than we currently have” with ST3, not “more rail than ST2”.”

       I totally agree. There were only 8 years between the two votes and in that time ST still had very little to show for the twenty years since Sound Move was passed. Yes, U-Link had just opened earlier that year, but honestly big effing deal. It was delivered 10 years late and still didn’t reach the orginal northern terminus at 45th Street.

       I agree with your opinion that had the Sound Move and ST2 light rail projects (all of which will be delivered late btw) been further along at the time of the 2016 proposal for a third phase, then ST3 most likely would have failed to secure passage. I think Sound Transit knew this and decided to also take advantage of the timing of a Presidential election cycle to help their chances.

       1. I think that voters are more swayed by offering a sexy rail system than by looking at specific cost-benefits of the larger system. It’s also a repudiation of using buses — even though buses often offer comparable travel times.

          The general electorate is not yet at a point where they can discern a difference between an effective and usable system and a lighter-used and less user-friendly system. They trust that ST knows what they want. At some point, this will change — probably about 2025 when Roosevelt and Beacon Hill riders can’t fit onto trains, 145th riders get stuck in a localized station traffic jam, many station neighbors complain about pick-up traffic issues, Westlake riders experience overcrowded stairs and escalators and thousands of Bellevue riders cross a busy 110th to transfer to any bus.

       2. Oh… I pretty much agree with both of you!

18. A more accurate title would be, “Load Factor is Highest in Seattle”. There’s a difference between where the most people are present on the train and where they all live. ST taxes are based on where you live, and those are the riders to whom the benefit accrues to. The fact that the middle is crowded comes down to the intrinsic nature of trains. When I take Amtrak from Seattle to Chicago or Seattle to San Jose, when it leaves Seattle the train is only a quarter to a third full and there are empty cars. That’s because it will pick up people in Tacoma, Portland, Salem, Spokane, Montana, etc. If it were three-quarters full when it left Seattle then there would be no room for the riders in the middle. If you look at Link from the perspective of where all riders’ homes are, you’d end up with a fatter tail and thinner middle. If you live on Capitol Hill, the presence of all those suburban riders passing through is not a benefit, it’s a drawback, because it leaves less capacity for you.

    1. That’s a great observation, Mike.

       The analysis that is most needed isn’t million passengers over a year; it’s the average loads on each segment at peak hours in the peak direction (and standard deviations) given the limited size of four train cars. More specifically, how many people that get left at the platform at each station when a train pulls out.

       We put lots of energy on STB debating loads by trying to interpret this missing information. It even creates internal disagreements in how to interpret this very generalized data.

       We should all come together and push ST for a map like the one in the article but for peak load factors. It would be very useful!

19. It looks like ST found a way to get taxpayers in Gig Harbor, Issaquah, Lakewood, Bonney Lake, and Everett to pay for trains for Seattleites. Collect $1.00 in taxes, provide 15 cents worth of express bus service, and apply the balance to light rail to transport Seattle residents to and from their jobs. A nice con that they have made work.

    1. Well, no. Issaquah got a train. Everett got a train. In fact, the budget was stretched thin to pay for Everett’s train.

       Snohomish County got more than its fair share of the dollars. It just won’t be spent very productively.

    2. I don’t think Gig Harbor is even in the ST taxing district? I thought they just paid for STX service directly

Comments are closed.