Sound Transit has come up with five route alternatives for the Federal Way Link extension and is taking public comments. I attended an open house Wednesday in Federal Way; there will be a second open house June 26th in Des Moines. In September the ST board will choose which alternatives to study in the EIS, and in early 2015 it will select a preferred alternative. The study area is from South 200th Street (Angle Lake station) to 320th (Federal Way Transit Center). Construction to 240th is funded in ST2. Construction further south would require additional funding, presumably in an ST3 vote. The alignments under consideration are:

  1. I-5 Mixed West Side ($1.5 billion). Elevated and at-grade along the freeway.
  2. I-5 Mixed West/Median ($1.6 billion). Elevated and at-grade along the freeway.
  3. SR 99 Elevated Median ($1.8 billion). Elevated in the middle of Pacific Highway.
  4. 30th Elevated West Side ($1.8 billion). Elevated on a minor street to 240th, then switching to one of the other alignments to 320th.
  5. SR 99 Hybrid ($1.8 billion). Elevated along Pacific Highway, crossing over the street several times. The north part would be in the median, 240th on the west side, 272nd on the east side. Then it would switch to the west side again and then cross east to the FWTC.

Ridership (23,000 riders) and travel time (14-15 minutes) are the same for all alternatives. This allows us to calculate a Westlake – Federal Way travel time of 55 minutes. That’s based on the existing travel time from Westlake To SeaTac (38 minutes), plus two minutes to cover the gap to 200th. For comparison, the 577/578 express bus is 37 minutes, give or take traffic conditions on I-5 and in the transit tunnel.

Seven stations are being considered, at 216th, 240th, 260th, 272nd, 288th, Dash Point Road, and 320th. ST2 envisioned only three stations (240th, 272nd, and 320th), but the others were added after public input. That doesn’t mean all these stations will necessarily be built, but some of them might.

My assessment is that SR 99 is better than I-5 due to the potential for TOD, direct transfers to RapidRide A, and proximity to Highline Community College. Its main hurdle is the $300 million cost over the cheapest I-5 alignment. Fortunately, travel time and ridership are not hurdles this time, unlike with the Lynnwood extension where the 99 alignment turned out worse in ST’s calculations.

To assess walkability, consider this: Going from west to east, first we have Highline Community College, which is on the west side of 99 at 240th. Then comes 99 itself, then 30th, then I-5. 30th is three blocks east of 99 at 216th, and one block east at 240th. North of the college, I-5 is 0.3 miles from 99 (7 minutes walk). South of the college, the distance between 99 and I-5 gradually widens, reaching 0.8 miles (16 minutes walk) at 320th. (Distances and walking times are based on Google Maps.) I don’t know much about the 30th Avenue alternative or why it’s being considered, but Google Street View shows a wide residential street lined with houses, apartments, motels, trailer parks, and car repair shops.

If you’d like to submit official comments to ST, go to the project page and click “Provide comments” on the right side. There’s a survey asking for feedback on the project’s need and goals, on which alignments to study, on which stations should be useful, and on what environmental factors to include in the EIS. The deadline is July 15th.

84 Replies to “Federal Way Link Planning”

  1. I would concur; Hwy 99 is the preferred option even at a slightly higher price. The prospects for beneficial TOD are much improved, and I’ve never liked the idea of putting all your transportation resources on one route anyhow. Any sort of major police or fire action has the capability of shutting down all your transportation resources just from one event. Put your transportation resources in different corridors and this won’t be an issue.

    The station at Highline CC is a must – it would be monumentally stupid to put this station anywhere but at the college.

    The Dashpoint Rd station is an interesting option. My gut says this station would perform better and have better TOD prospects if it was at the high school on 304th, but that is getting fairly close to the station at the FW TC. I’m not sure if that is an issue or not, but it might help convert that sea of parking that is FW into something a bit more palatable.

    1. Maybe a combination could work, like at S 304th St. That way, it makes the walk to local FW businesses more bearable (including FWHS, which would be a 5 minute walk away), rather than S Dash Point road, where it mainly would serve residents.

  2. Have their ever been any studies on increased ridership due to people stuck in traffic being able to see the trains speeding by?

  3. Ridership (23,000 riders) and travel time (14-15 minutes) are the same for all alternatives. This allows us to calculate a Westlake – Federal Way travel time of 55 minutes. That’s based on the existing travel time from Westlake To SeaTac (38 minutes), plus two minutes to cover the gap to 200th. For comparison, the 577/578 express bus is 37 minutes, give or take traffic conditions on I-5 and in the transit tunnel.

    One assumes that peak-hour express bus service from Federal Way to Seattle would remain in place, but that off-peak riders would have to use the train.

    If service were provided along Highway 99, how many current RR A users could use the train instead? Would there be any savings from reduced local bus service?

    1. Current 577/578 headways combine for 30 minutes off-peak nearly all of the day, 7 days a week. With an 18-minute time differential, Link becomes only worth considering if you just missed a 577/578 bus.

      Of course, that is only for people who are already at the transit center. If you are coming from somewhere along SR-99, a little bit north of the transit center, a direct route to downtown on Link would likely be faster than taking a connecting shuttle that backtracks to FWTC, then waiting for the 577/578.

      Of course, the real irony here is that the off-peak period is when the time savings of the 577/578 over Link really matters. During the peak, much, if not all of the 18-minute differential, is likely to get squandered in traffic, particularly along the final approach into downtown. Only off-peak, would the bus actually be consistently significantly faster than Link.

      1. Off-peak, I doubt the 577/578 will be retained as is. The 577 could be dropped, and the 578 rerouted to Kent to serve as a Sounder shadow.

        Of course, there’s always the (I think, obvious) option of having the 594 make a Federal Way stop off-peak.

      2. Former 577/578 commuter here.

        The bus does not spend much time on its final approach to Downtown. The majority of the time waiting is spent at the light at 6th & Seneca except for those rare occasions where the operator catches the green.

      3. I talked to one staffer who was described by another staffer as an ST Express expert. The claim turned out to be slightly exaggerated, but not the staffer’s fault he was described that way.

        I asked him about having the 594 stop in Federal Way. He was very, very cold to the suggestion, since Federal Way to Tacoma is already served by the 574, Seattle to Federal Way is already served by the 577/578, and Seattle to Tacoma is already served by the 594. I brought up how improving frequency from 30 minutes to 15 minutes reduces wait + travel time more than the time added by stopping in Federal Way. He didn’t seem to think frequency was relevant, and then got hung up on the notion of having all routes stop in Federal Way, including the 590. He was also unfamiliar with the 512 restructure. The conversation didn’t get as far as suggesting a different path for the 578.

      4. That’s the issue: it boils down to whether you think 15 minutes is a better frequency target, or 30 minutes to different parts of the corridor is sufficient. I think 15 minute frequency is important, and would gain more riders than the extra 594 stop would lose riders. Because people make marginal decisions whether to take transit, and if they know a bus is coming within 15 minutes they’re more likely to take it than if they know they’ll have to wait 25 minutes, or if they’re not sure until they get to the bus stop whether they’ll have to wait 25 minutes.

      5. And, even more so, the 594 could easily make up the time lost by stopping in Federal Way simply by skipping SODO and going straight into the center of downtown, like the 577 and 578 do.

        If you think of this way, it is not really a tradeoff of a slower ride than runs ever 15 minutes vs. a faster ride that runs every 30 minutes. Rather, it is a tradeoff between having direct service to SODO every 30 minutes, or require a connection and a short backtrace to reach SODO, but have service run every 15 minutes.

        Given that the number of people traveling off-peak between Tacoma and SODO is near zero, this becomes a no-brainer.

    2. I’m not even talking improved frequency. I just want to add coverage to Kent, which is very poorly served by the 150, and which could be done at no (or very little) cost by rerouting the 578. And of course no one wants to make peak-hour Tacoma expresses stop at FW or cancel dedicated peak-hour FW-Seattle service — that service already runs at capacity.

      Judging by the unfamiliarity of that staffer with the network I think I need to write ST a detailed letter.

  4. I say number 3, although 4 and 5 are fine, but 1 and 2 are simply downright unacceptable. If we are going to be talking I-5 here, then there is really no advantage to this other than an alternative to interstate 5 for express service. Where is the business? SR99. If we don’t build there, then it will take years to build the business where the rail is, when we should be building rail where the business is. Why should Mr. D. M. Resident get a link station right next to his front door, but all of the businesses are at least a 10-15-20 minute walk?

    Also, the stations need to be more numerous and closer together; then someone will actually be able to get where they are going from the stations. For example: the had originally planned a S 240 St station and a S 200 St station. What if you are going somewhere on the 22000 block of SR99? Sorry, 20 block walk for you. That’s a 1.3 mile walk where the train goes by anyway (unless, of course, I-5 is chosen). It is wasteful of time and resources if I need to ride link and transfer to the A-Line to get where I am going, when they could make stations more frequent and cancel the A-line. If they could make Link a viable alternative to the A-line, then Metro could save so much money by cancelling it, and using it to provide service on other routes.

    1. Unless you’re prepared to make Link stop every quarter of a mile, I don’t see how you could get away with getting rid of the A-line. And if we did, it would take way longer than 55 minutes to get from Federal Way to downtown.

    2. “Also, the stations need to be more numerous and closer together; then someone will actually be able to get where they are going from the stations.”

      It makes sense to me to have fewer stations, keep RR-A where it is and use the bus to fill the gaps between stations. Each additional station adds time to the trip for deceleration, dwell time and acceleration.

      I’m not familiar with the area, but I get the impression it’s not very dense. ST should perhaps identify the locations of potential future stations and build them if and when the area can support them. With the exception of the community college and the transit center, wouldn’t almost everyone be driving to the stations?

      1. I’m not familiar with the area, but I get the impression it’s not very dense.

        You’re mostly right. It’s not very dense. However, it is pretty much the densest corridor on the east side of the valley. Major intersections on International Blvd generally have a good amount of retail at the intersection and stretching north/south along 99, and some apartment/condo complexes along the intersecting streets. In this part of the county, it’s the best we can hope for.

      2. You guys all understand that Link’s best estimated ridership numbers are terrible, right?

        AW is correct: this thing will not be useful. Better to spend a fraction of that money making RR what it was advertised. At least the few who do exist in the walkshed of 99 can all walk to RR. Link will (badly) serve a handful of select locations with not nearly enough cumulative demand to justify its existence.

      3. Actually, it was Alex who correctly pointed out the uselessness of Link. AW went to bat for the ridiculous concept of medium-frequency (yet still empty) trains overlaying a medium-frequency (yet still slow) bus.

  5. While the bus may be almost 20 minutes faster in theory, at least with the light rail you can pretty much bet that you’ll make the trip in 55 minutes, rain, traffic, snow, protests, presidential visits, etc.

    Now if they could just figure out some sort of Rainier Beach bypass…

    1. The real slow part of Link is not actually Ranier Valley, as everyone claims, but SODO and downtown.

      If we want to make things faster, we should be focusing on fixing the signals at Royal Broughm so the train can go through faster, kicking buses out of the tunnel, fixing the security barrier so the train doesn’t have to stop and wait a minute to enter the tunnel, nor crawl through the maintenance south of the ID station at 5 mph, and closing stadium station during the 99% of the time that nothing is actually going on at any of the stadiums.

      While each of the above improvements sound small in isolation, taken together, the savings adds up to about the same amount of time as a Ranier Valley bypass would provide. But it would do so at a fraction of the price of miles of new elevated track. And, it would improve service to the Ranier Valley as well, rather than taking such service away.

      1. I completely agree. The aggravatingly slow parts of Link is not the RV it’s the segments you point out. The ironic fact is that the places where ST/Metro have completely control over operations are the slowest parts.

      2. I was more thinking about distance and speed. By cutting off the RV deviation and raising the speed limit from 35 to 55, you would definitely save a good chunk of time. Couple that with some SODO improvements and you got yourself a heckuva LRT.

      3. For once, I agree with ASDF completely. Complaining about the part of the route that goes where people exist, while giving a pass to the insane operational fails at the tunnel entrance, is nonsense.

        No need to close Stadium station, though. The minimal dwell time of a single station at least has the benefit of offering added walkshed and permutations of destinations. 5mph tunnel crawls offer nothing.

        (And dwell time itself could easily be shortened; Boston and Montreal dwell times are frequently as short as 8-10 seconds.)

        RapidRider: A heckuva useless LRT.

      4. The real slow part of Link is not actually Ranier Valley, as everyone claims, but SODO and downtown.

        If we skipped the RV alltogether, Link wouldn’t be so damn slow.

      5. I agree with RapidRider, its both. From a transit perspective, the train is no slower through downtown than a bus. Even driving is very slow through downtown (although not that slow). On the other hand, a bus is much faster getting to Sodo than link (since a bus can go 60 mph on a freeway, but the train has to go 35 on the street). This is why lots of people still prefer their express buses and the airport shuttle vans still do plenty of business.

        You are right, though, we should definitely try and improve the speed starting with the changes you mention, since it is much cheaper than adding new rail (or building a new tunnel, etc.).

      6. Stadium Station is where operators get on and off to head to other stations to make field driver swaps.

        The fact of Rainier Valley being at-grade adds 2-3 minutes to trip time. The fact that Link takes that path at all adds about 7 minutes. The twists and turns through Tukwila add about 3 minutes. The lack of tail track at Airport Station adds close to a minute for the approach, so we can look forward to a little bit of speed-up, mitigated by the fact that the trains will spend a more unreliable amount of time at the station as everyone and all their baggage can’t board until the train pulls up.

        I’m not sure if each station adds 40 seconds to trip time or if that’s how long they expect a train to be stopped at the station on average, meaning the added time might be significantly longer than 40 seconds.

        Link is suffering from the death of a thousand curves. Can we please not repeat these mistakes in South King County?

      7. To D.P.s point about dwell times, EVerywhere I’ve been except seattle has like 5-20 second dwell times, when I am back in Seattle trains stop for 30-50 seconds it feels like, what gives? Far more people get on trains in cities with 10-20 second stops than ever get on the 30-50 second super-stops that link does. I’ve always wondered why on earth they waste everyone’s time like that?

      8. Some of the station dwell time might be signal-related. If a traffic light happens right after a station, it doesn’t make sense to trigger the train’s TSP before the train stops at the station – otherwise, you end up with cross-traffic sitting at a red light because of the train that isn’t even there. So, when the train is ready to go after a station stop, the light ahead might be red. The train can either close the doors and inch forward to the light, or wait for the light at the station with the doors open. The latter option wins, as it at least allows a few latecomers to board who would have otherwise had to wait for the next train.

      9. Nope. Even our standard, off-hours, nothing-is-ahead-of-us-in-the-tunnel dwell time is arbitrarily long for no reason.

    2. The Rainier Valley overhead is 10 minutes. The problem with building a bypass is that Seattle has several more critical priorities for its transit dollars. It would be nonsense to give SeaTac and Federal Way travelers two Link options while Ballard and West Seattle have no Link service at all.

      1. @Mike Orr Agreed. Express train runs would be great, but we have neither the ridership nor the coverage yet to justify it. We can worry about a south end express AFTER we actually have something that covers the whole city.

        An express route makes a lot more sense when you have a much longer stretch (say Northgate to the Airport) to make an express route out of. In that case, fewer stops downtown would also be nice (in which case you would take a local to the collection station and transfer to the faster train).

        Really though we are talking about a new line at this point… and after we have enough ridership on the system to justify it. Cars always stand a chance to be faster, but at least you don’t have to park a train or deal with backups on I-5.

      2. Of course, Federal Way could “donate” money to ST to build the bypass, the way IT is buying a bus extension to Olympia. But South King County is going to have enough trouble even building this extension, much less anything more. ST2 was supposed to fund construction to Federal Way, but south King County got hit worst in the recession, and it doesn’t have many billion-dollar companies or upscale malls down there generating tax revenue. So I’m still wondering if south King will really take the Link extension bite now, or ask for more ST express routes and BRT instead. (Kent – Seattle and Kent – Federal Way come to mind, as well as upgrading the 169 and 180.)

      3. We already have a Ranier Valley bypass, and it already runs during only period when the ridership is there to justify it. It’s called Sounder.

      4. @Charles B, Mike Orr,

        There is a great opportunity to build a very cheap, mostly at-grade routing between Airport Way and the freeway, with perhaps a stop at Boeing Field if there’s enough demand for it. But that option could be fatally foreclosed if further development happens in the strip of land through Georgetown. ST should buy the property with a lease-back to the current occupants for fifteen years. Nothing will happen before then, if ever.

        If nothing comes of it, or ST decides on a west side elevated bypass with a stop at the air museum — a roughly five time the price — the land could be sold.

        In either case, a flying junction already exists at the maintenance facility that can be used to merge back into the main line there.

      1. The 577/578 also save quite a bit of time (usually) by skipping SODO and taking I-5 right into the center from downtown.

        However, it seems really arbitrary that people from Federal Way deserve this benefit, but people from Renton (101), Tukwila/Kent (150), or Tacoma (594) don’t. Having local buses, like the 124, run the SODO busway to Spokane makes a lot of sense. But I highly doubt forcing express buses down that slog has the ridership there to justify it. Also, even if buses like the 594 can be made to skip the SODO busway during normal operations, they can still use it as a backup approach into downtown during times when the normal I-5 approach is heavily congested (for instance, before a Mariners’ game).

  6. I was also at the display at the west end of the ridiculously anti-pedestrian anti-local-bus-convenience Federal Way TC (where you are supposed to cross north to get out of the TC before heading south to the Commons, just for starters, and the buses from 320th have to do a VA-style knot to serve the TC via 23rd Ave S on the east end of the TC, albeit without sitting in a parking lot SOV line).

    I asked about the possibility of keeping Federal Way Station on Highway 99. The various ST employees indicated that that is technically within the scope of the EIS, since it was within the box of possible station locations. A major hurdle is that ST got a lot of federal funding to build Federal Way TC, and might have to return the money if it ceases to be a transit center. So, the transit center will have to stay, and have at least some buses serving it, but those buses could also serve a station a few blocks away on 99. A little redesign could allow the buses to enter and exit the transit center on the west end.

    They also pointed out the preponderance of buses that will be coming to Federal Way TC to transfer to the train, as well as buses coming from the TC to head to Seattle. The first part, I presume, is riders coming from Tacoma. But Link will likely reach Tacoma some time shortly after reaching Federal Way. So, this is an interim ridership group. The permanent ridership group is local bus riders coming from the neighborhood spine of Federal Way: 320th St (served by the 179, 181, 187, and 197). Those riders would best be served by a station entrance at 320th and Highway 99. If those buses have to keep doing the FWTC knot, that adds about five minutes of travel time each way for the majority of Federal Wayers catching a neighborhood bus to the train.

    In terms of parking costs, that extra five minutes will add demand for more parking, as fewer Federal Wayers will but up with those circuitous bus routes. If taking the bus gains back its five-minute advantage, that will positively impact demand for parking.

    I also plugged the necessity of having a station entrance adjoining Central Washing University – Des Moines Campus / Highline Community College.

    Indeed, I’m not sure any other stations serve a purpose that justifies their cost. Part of the cost of any additional station is 40 seconds of additional trip time. One staffer pointed out that if more stations are added than originally planned, then more capacity for storing traincars will have to be found than is currently planned for with the new maintenance base.

    For me, the question comes down to purpose. Is Link going to be used by riders expressing between Tacoma, Federal Way, and downtown Seattle? If so, then I would suggest cancelling Redondo Heights Station and just having Highline and Federal Way Stations. If not, then I don’t see much harm in adding stations that have TOD potential, other than the cost of building the stations.

    However, if adding a station is the political cost of keeping the line on Highway 99, the jog over to I-5 and curviness/slowness in the path might have killed an additional 40 seconds of trip time anyway. I realize the EIS can’t weigh “political costs”, but we can.

    My bottom line is that there needs to be a station entrance at Highway 99 and 320th St, and on the west side of Highway 99 next to the CWU-DM/HCC campus, and whatever else can be done to minimize trip time, do it. These goals are intended to maximize the carbon pollution reduction effect of the project (which is important to mention when commenting on an EIS).

    1. “the transit center will have to stay, and have at least some buses serving it,”

      So that’s a perfect opportunity to have a separate “downtown” station from the “P&R” station. I’m not sure how close 320th & 99 is to The Commons entrance, but I expect something could be done with the station area.

      1. Don’t forget the Federal Way P&R. Even though the garage at the TC fills up, the P&R still sits mostly empty. Definitely has 60+ cars using it weekdays, so not completely useless, but still has plenty of capacity.

    2. “where you are supposed to cross north to get out of the TC before heading south to the Commons”

      No one ever does that. People cross the bus area SE of the platform, and no one ever gets in trouble. It’s even safer than crossing a one-way street, because it is limited to transit traffic. But if you are heading to the commons, then you can make a quick transfer to 903, 182, 500, 501, 402, which all go by the commons. You have 6 chances per hour off peak then (7 on peak), and if you count 181 and 187 and don’t mind walking from S 320 St, then you get 9 chances per hour, 10 on peak.

      I don’t mind the knot so much, but I would like it if (going eastbound) they would turn left at 21st Ave S, and enter the transit center from the west side, but then they would have to reassign bay arrangements for the 181/187. I hated when they started making the A-line go all the way to 23rd before entering the transit center, where it used to take a right on 21st. It has made RapidRide even more of a misnomer.

      1. What is the point of a “transit center” if you then have to catch another bus to get to any real destination, especially one a quarter mile south? It’s just an expensive park&ride.

        What would make a lot more sense is to have the A Line stay on Highway 99 and continue down to S 348th St. As Joshua Kelley points out elsewhere, there isn’t much market for catching the A Line south to FWTC and then transfer to a 577/578 north to Seattle, but we’re serving that nonexistent market at the expense of riders who just want to get to their jobs at the Commons without having to wait for and transfer to another bus to go that last quarter mile.

  7. I really wanted to go to this open house, but wasn’t going to drag along four young kids with me. I went to one a few months ago, and made sure to include my input that an I-5 alignment is a really bad idea.

    As one who lives within a block of 99 not far from 272nd, Link all the way would definitely be faster than A-line > Tukwila > Link > Seattle or A-line > FWTC > 577/578 > Seattle.

    TOD is a huge potential and there are many apartments and neighborhoods much closer to 99 than I-5.

    As far as the potential stops, 216th would likely be strong, and 288th MIGHT be strong. I’m not so sure about 260th. I’m not sure about Dash Point; there are quite a few apartments there in addition to the middle school, so it might do decent.

  8. Hint for DP and Mic: you can go to the survey and click “Strongly Disagree” on the need for this extension. I still believe that this extension is not “necessary” in the way that Northgate to SeaTac is vital to Seattle’s future. But at the same time, if south King County is ready to go with it, we mustn’t stand in their way because a seamless subway system across the metropolitan area benefits everybody. But its biggest benfit depends on the cities realizing the vast redevelopment potential along 99. There are tons of underused parcels and tired old big-box lots that could easily be redeveloped to mixed-use buildings at 4-6 stories without blocking any single-family house’s view.

    30th also has potential for redevelopment, and that may be why it’s being considered. I predict two futures for 30th. Either it’s already being planned for upzone, or the single-family homeowners will successfully defeat this alternative. There’s no “need” to put it on a residential street when a wide commercial street is a few blocks away.

    1. Thank you, Mike. As always, you’re a gentleman with those known to disagree with you on particular points.

      But honestly, “a seamless subway system across the metropolitan area benefits everybody” is a The Great BART Fallacy, already disproven on the ground.

      When vast stretches of the system serve underwhelming single-use areas with no possible appeal or purpose for those elsewhere in the system, those parts are near-valueless to the whole. (Put another way: whereas maximized urban coverage benefits all urbanites, and whereas links to major regional destinations benefit the region as a whole, spindly exurban coverage benefits only a very select few.)

      Repeating that fallacy is dangerous: bad lines get planned, urban segments get shafted, idiots suggest bypassing the R.V. for the further gain of a tiny unidirectional minority, and North King gets directly billed for segments built to others’ specifications. It’s a faulty, fraudulent, very hairy logical path.

      1. The billing of Seattle for the Ranier Station comes down to considering the alternative, which is Link going straight from downtown to Mercer Island, without stopping at Ranier at all.

        From the perspective of the suburbs, this no-build alternative would actually be preferable, as it would save them half a minute to get downtown and if they want to go to somewhere along Ranier – well, not many suburbanites will be going there anyway and those that do can just hop in their cars.

        Hence, the argument goes, if the marginal benefit of station vs. no station is entirely Seattle, the marginal cost of building it should be paid for entirely by Seattle. While east Link as a whole may provide more benefit to people who live in Bellevue than people in Seattle, Ranier Station is still mostly useful to people who live in Seattle.

      2. @asdf: Billing North King for Rainier Station is reasonable, for something like the reasons you state. Billing North King for the track segment is unreasonable, by pretty much the same logic.

  9. 55 minutes to Federal Way-reminds me of recent posting assessment of transit in Seattle being quote “pathetically slow” unquote.

    Recent BART travel confirms above impression. There is definitely something in the makeup of our whole region that honestly feels that “fast” means dangerous, expensive, and politically impossible.

    Nothing against LINK through Rainier Valley as a first leg. But a few years back when Alaska Airlines was considering moving to Boeing Field, I really hoped we were going to get the motivation to put an additional line connecting the two airports.

    Know the reasons, have heard the arguments. But an hour from Downtown Seattle to Federal Way is just too damned slow.
    Exactly the kind of thing that makes a difficult political sell into an impossible one.

    Mark Dublin

    1. BART takes about 40 minutes to cover a similar distance. And that’s in spite of streamlined switching, arrow-straight running from park-and-ride to park-and-ride, and miles between the stations. Not all that impressive, especially considering the total lack of impact BART has had on the built environment and the paucity of actual destinations it can be used to reach.

      Could it be that spontaneous travel over wide expanses of sprawl isn’t a sustainable model of rapid transit? And that what BART and Link try to do in the first place is the impossible sell?

      1. Quick answer…NO.

        “Stations seeing a strong uptick in usage compared to last year include: San Bruno (up 15%), Colma (up 14%), Ashby (up 12%), SFO (up 11%), and 19th street in Oakland (up 10%).

        For the first time since the opening of the San Francisco International Airport BART station, ridership reached and exceeded 4 million annual passengers in the fiscal year.”

      2. Yay for P.R. spin!

        If you suspect a transit agency is trying to pull wool over your eyes, look for self-referencing comparisons (rather than raw mode-share) and annualized ridership (rather than daily numbers that are easier for the reader to parse).

        Fact: Outside of rush-hour, mode-share on BART’s sprawl branches is negligible. Even for peak commutes, mode-share is in the pathetic single digits. Subsidies at outer stations crunch to $32 per boarding.

        Fact: SFO had 425,000 aircraft operations during that same annualized period. That means that fewer than 10 passengers, crew, and/or assorted airline/airport/tangential staff arrived or departed by BART for every flight movement. And direct SFO-to-city trips is one of the few things BART is actually good at, in terms of comparative speed, access, and convenience.

        If you’re okay with building multi-billion-dollar lines used by only a handful of people for a single purpose at $32 subsidies each time for all eternity, then I guess the BART model is for you. But by most objective measures it is a failure.

      3. I mean… yeesh!

        Mike is forced to write that all of these options will cost $1.5-$1.8 billion and garner only 23,000 boardings (11,500 users), and we’re all so inured to that insanity that no one even addresses it directly?

        This is ridiculous, and those who argue “regional benefits” make themselves look ridiculous.

    2. This line is slow for the reasons mentioned above (starting with RapidRider’s comment). The main reason is that we were (and are) cheap. The other reason is that folks in Rainier Valley had no interest in an elevated system. There aren’t that many stops, so a system with grade separation as well as the improvements asdf mentioned would be much, much faster. The improvements asdf mentioned sound fairly cheap and easy — building a tunnel wouldn’t be.

      For what its worth, the North Link will be much faster, and I predict, a lot more popular. We should have started with this (of course) but better late than never. When it is built, it will be the fastest way to get from any one of the three largest urban centers in the state of Washington to another.

      1. It also has to do with when the decisions were made relative to Link’s opening. When the Rainier Valley alignment was decided, there was no light rail anywhere in the region so it was all theoretical, and most people had not been to Portland or Vancouver or San Francisco to see their systems, and even if they did it was still theoretical how it would interact with the Seattle landscape. Now Link is on the ground and successful, and people in Des Moines and Ballard can ride it and decide what they think of its surface segments and elevated segments, its speed and frequency, and its impacts on the neighborhood. So now it’s more concrete: “This is what you would be getting.” Its success is also a recommendation, because Federal Way is quite keen on getting it. All this means that more things are politically feasable now than when the Central Link alignment was decided. And indeed, mistakes like the First Hill streetcar compromise are not likely to be repeated. (I.e., the trolley-BRT alternative might have gotten a better hearing now than it did in ST2, and put a full streetcar somewhere else.)

        Our suburban situation is not as bad as the Bay Area’s, so I think BART II would not have the same impact. That doesn’t mean Des Moines will start looking like Brooklyn, but I think some of BART’s extreme examples like Dublin and Pittsburgh can be avoided. Distances are shorter here: most people travel half the distance than they do down there, the main cities are closer together, and if you drive 60 miles you’re in freaking Olympia or Mount Vernon, not San Jose. The longest Link extent proposed is Tacoma and Everett, which are the traditional second and third cities in the region (now third and fourth due to Bellevue). Nobody has proposed extending Link to Du Pont or Marysville (except John Bailo), which would be the equivalents of Dublin and Pittsburgh, the hinterlands beyond the principal cities.

      2. 1) It’s silly to blame democracy for the decisions that are supposed to be delegated to experts. It is those experts’ job to understand precedent and to overrule parochial ignorance.

        2) You would actually have expected BART to do better given the Bay Area’s “situation”: very large numbers of people in extensive sprawl causing very bad congestion much of the time. And yet BART still didn’t work.

        Here we have an even sparser form of sprawl, with not quite as many people in it, and as bad as our highways can sometimes be, we’ve got nothing on the Bay Area as a whole. Link looks even less favorable in comparison to driving most places most of the time.

        Notes: Everett and Tacoma are 28 and 33 miles, respectively. Fremont, California is a similar distance from San Francisco. Fremont is also the 15th most populous city in California, with a population that dwarfs Tacoma’s and especially Everett’s. Although Everett and Tacoma each have a downtown, neither is especially populous nor the primary activity center for its city. The preponderance of land use in Everett and Tacoma is about on par with Fremont. A whopping 7,000 use the BART terminus at Fremont on a weekday.

      3. Driverless trains also depend on no buses in the DSTT, and no surface segments like SODO. In the beginning ST was trying to put surface segments in Tukwila to keep the capital costs down, and wanted to keep the option of surface segments in extensions. This was before any Link was on the ground, when passing ST1 was more iffy, and when people were more focused on capital costs than on anything else.

  10. I provided feedback. 99 or nothing at all. It’s a clear no-brainer. Although, ST is likely to fight for I-5 desperately for no apparent reason. With many infill stations, RR-A could simply be reduced to local service to supplement intermediary stops and provide funds to other routes in the area.

  11. I think it is worth mentioning that I believe this system will mainly serve Sea-Tac area workers (and fliers) as opposed to folks getting to downtown. If you look at the existing rail line on a map, it is easy to assume that most folks riding this are coming in from the suburbs to work downtown. But the numbers don’t bear that out. The Tukwila station is a nice, suburban station. It has a very big park and ride that makes John Bailo happy. It is the only station for a long while. You would assume that it would have big boarding numbers in the morning (heading north) and big departing numbers in the evening (coming from the north). But that isn’t the case. Lots of people depart in the morning (from the north) and not that many people board (heading north). The same sort of pattern is repeated in the evening. In other words, people are using link not as a classic “suburban bedroom community to downtown” system, but so they can get to their job in Tukwila.

    My guess is that most of these folks take another bus to get to their job. My guess is that the job is low paying (motel or restaurant work). I would also guess that lots of these people don’t drive at all (or drive very rarely). This suggests a couple of things:

    Since this will be grade separated for much of the way, this could (unlike the current system) provide for a good way to get to the airport. Some of these folks work at the airport on a daily basis, so using a park and ride (for them) makes sense. However, lots of folks getting to the airport will be gone for several days, so using the park and ride is out.

    Lots of low way workers will continue to commute to the Sea-Tac area. Some of these folks have cars, and will drive them to a park and ride, but a lot don’t.

    All of this suggests that stations interfacing with buses on this route, or serving big apartment complexes (or both) is critical. Providing a few good park and rides would certainly add some ridership, but not as many as one might assume. As lots of folks have mentioned — riding Link to downtown will take a very long time, so lots of people won’t want to do that.

  12. Is there really no way to get link trains to go 60+ miles per hour in the space between Tukwila and Rainer Valley where the only thing that is there is nothing?

    STB has made its case for light rail over BRT, but BRT can actually stop AND go. They can get to 60 miles per hour, even on a wide turn, and they don’t need 20 miles of distance in order to completely stop. Making special “aquaduct” style roads for BRT would be of comparable cost. What exactly makes link so much better again?

    1. BRT vehicles (hybrid buses) could not come close to hitting 60 on a hill as steep as the one Link climbs from Allentown to TIBS. Down on the surface, similar buses operating the 194 were doing well if they could reach 40 mph.

    2. BRT’s only advantage is if it costs less than rail. if you build new dedicated streets for it, the costs approach rail, so why not go with rail.

      1. This may sound ridiculous and silly, but what about building an elevated BRT roadway directly above WA-99 (almost exactly the same as the currently-planned light rail) from FWTC to 200th St, provide a timed cross-platform connection with Link at Angle Lake, and then from there have buses running nonstop on I-5 transit-only lanes to Downtown (which cost nothing–just restripe)? It would cost basically the same, but it would provide a much faster ride for South Link>>>Downtown than Link and would still be the same for trips to the Airport and Rainier Valley. BRT isn’t always about saving money over rail: it can also improve mobility in certain cases where ROW is already there (like in freeways).

      2. Only if they take away a general traffic lane from the freeway and dedicate it for BRT will it avoid traffic congestion adversely affecting the route. Good luck with that in a place like Seattle.

    3. Link trains can do it, currently, it’s just against the rules. Back when test trains were still running, before Link entered revenue service, a couple operators reported they had “accidentally” held speeds in the low 60’s through that segment.

    4. It is a little painful how some Link operators do not trust the physics of the tilted curves and take those curves at a slower, less safe, speed. It probably doesn’t significantly increase the chance of a derailment, but that extra time having passengers leaning without the counterbalance of “centrifugal force” (or whatever the proper phrase is) can increase passenger incidents.

  13. While admittedly I don’t know too much about this corridor, I’m still kind of confused as to why this extension is being considered. It seems that these would be the primary goals for extending Link (please correct me if I’m wrong on this):
    1. Improve mobility on public transportation, so that it is more competitive with the automobile and provides more “freedom” for users (see Jarrett Walker)
    2. Increase ridership on public transportation, to improve environmental sustainability

    From my view, it seems that extending Link to Federal Way is not the best way to achieve these goals using South King money. While I don’t have the numbers to be sure of this, it seems that the best way to meet these goals, using the money that we have, is to build BRT along I-5, with dedicated transit lanes, nice freeway stations (with direct access from transit-only lanes), and frequent service (<10 min headways all day). Since HOV lanes already exist on I-5 and could easily be turned into transit-only lanes with the addition of a few signs, this alternative seems like it would be much cheaper than extending Link, and would produce significantly more mobility because travel times to Downtown Seattle and SeaTac Airport would be much faster (as shown by this post). The money saved could then be invested in substantially improving the local bus system by adding more frequency and amenities, which would serve primarily as feeders connecting neighborhoods to BRT stations.

    Overall, it seems that this alternative would improve mobility and transit ridership much more, although if anyone can convince me otherwise I'm definitely open to it!

  14. Based on these alternatives, ST hasn’t yet determined whether South Link is intended for medium-speed local travel or express regional travel.

    FWTC is 23 miles from downtown Seattle. I did a comparison of the travel times from locations 23 miles out at a variety of cities in Seattle’s size range: Berlin, Lisbon, Montreal, Boston, Washington DC.

    Answer: most city’s rapid transit lines only extend about 10 miles from the city center. Commuter rail is used to serve locations 23 miles out, with typical travel times of 40 to 55 minutes (these lines operate all day long at least hourly). Driving the same distance (with traffic) typically took 30 to 45 minutes.

    Exception: The Shady Grove Red line in DC extends 23 miles from DC Union Station. Travel time is 40 minutes.

    Application to South Link: (A) The linear shape of the metro area and is the only justification to build rapid transit south of Angle Lake Station. (B) Long term, this corridor should be served by a new commuter rail/HSR alignment directly serving Sea-Tac airport but generally in the I-5 right-of-way. (C) In the interim, South Link could function as an express commuter rail by building only the stations called out in ST2 along the 99 alignment. 50 minute travel time to downtown is actually reasonable per international standards. (D) If and when the new commuter rail service is built, Link could be converted to medium-distance service by adding all the in-fill stations proposed, and replacing parking garages with mixed-use TOD.

    1. Your analysis of comparative distance, travel times, and modal form is fundamentally correct.

      But what your prescription appears to miss is that the areas served by closer-spaced, high-frequency rail for medium-distance journeys must have the kind of land usage and variety of demand-inducing destinations to warrant such service.

      (Read: They must be cities.)

      That kind of high-volume service is not going to be justifiable in the sleepy sprawl 23 miles from town… ever.

    2. In large cities, such as New York, they serve both types of transit riders.

      They have Express and Local service on the same line. Very expensive, since you now have double the infrastructure in the same corridor, although concentrated in a small footprint.

      Some lines even supplemented local service with only one express track, to save money.

      The other variation is skip-stop service, or have close spaced stations served by ‘locals’, and express trains not making selected stops. Much like CT does with Swift.

      I asked an ST staffer about doing that with Link through the Rainier Valley.
      His comment referred to the ‘Kill Zone’. That is, many people would behave as though the train were going to stop, and at those at-grade pedestrian crossings put themselves in front of what would turn out to be an express.

      1. Skipping stops at the Ranier Valley would not make sense. First of all, it would save, at most, 2-3 minutes of end-to-end travel time – not enough to justify cutting back Ranier Valley’s peak service from every 7.5 minutes to every 15 minutes. Even if you ran peak trains that served no stops whatsoever between SeaTac airport and International District Station, you would still not be able to save more than 7.5 minutes of travel time, as, at that point, the express train would merely get stuck behind the local train in front of it.

        It would also make the system a lot more confusing. In New York, when trains run every 5 minutes, if you accidentally take an express train when you should have gotten on a local, you can turn around at the next stop, and it’s no big deal. When trains run only every 10 minutes, the time penalty to turn around to correct a mistake increases accordingly.

      2. Re: New York:

        As I’ve pointed out before, the minute that headways on either the express or local track drop to longer than 7 minutes, the expresses stop running. Because at that point your divided service and inability to easily transfer between the two reduces utility to the point where having both does more harm than good.

        Oh, and South King County is not New York City.

      3. The Puget Sound Region is not New York City

        We’re Wet Los Angeles.

        And our automated flight path of transportation spending is exactly that.

      4. @asdf

        Yes, the metrics you state are spot on, it’s only a few minutes difference.

        That’s one of the issues of whether Link is being built as a commuter service, or providing a close spacing local service.

        The Hwy-99 alignment, (being the better choice,) would do better with some more in-fill stations, but even though, as you stated, the travel times aren’t dramatically different. For those people using it from the farther reaches as a commuter service, they perceive the multiple stops as an impediment to a useful commute (as incorrect as that perception might be), and choose to drive.

        There are places that you could have passing tracks for an express service, it would just be a matter of pulling that off operationally. The system isn’t mature enough for that, yet.

      5. The system will never be “mature enough for that”, because there’s practically nobody and nothing along the South King alignment where you hypothesize express trains.

        Again, even New York knows there’s no reason for expresses unless each service can run every 7 minutes or better. Otherwise, your service divide makes everything less useful for everyone.

        Amazing how the rail obsessives fail to understand even the basics of demand service and operations.

  15. ST is hiding the best (most cost-effective, providing the most transit mobility for available funds and rights of way) option from themselves: improvements to the regional express network; their studies forecast little ridership for the rail options; they spend too much on unneeded capacity; they invest too much in reliability; a regional express option could greatly reduce headway on routes 574, 577, and have a more frequent Route 594 stop at the Federal Way TC via the Sound Move funded center access ramps at South 317th Street; regional express routes could be elevated to BRT with faster fare collection; I-5 could be tolled; regional express options could be implemented within a few years rather than decades.

    1. Sorry, but a fleet of buses stuck in traffic on I-5 is not BRT, unless you are looking at the playbook of BRT-is-better-than-rail proponents who only support BRT when rail is on the table. Real BRT involves dedicated ROW, which then begs the question, why not just put it on tracks and electrify it? What ST is proposing for northeast King County is not BRT, BTW. Yes, we should have much better frequency between Seattle and Federal Way, but please don’t call it BRT (and sorry if I occasionally use that acronym for something that is merely a frequency improvement).

      I suggested a proposal recently that has a beefed-up 594 serving the I-5 stops of the 574, scavenging the service hours of the 574, and meeting with crossover traffic at Kent-Des-Moines P&R with a beefed-up 578 going via Kent Station rather than Federal Way. The only rider group going to the airport this would inconvenience is Pierce County riders, who would have a short transfer wait at KDM P&R, and only until Link reaches Highline Station. Riders from the Auburn-Puyallup crescent would have excellent airport connection, but would have a short transfer wait at KDM P&R for getting downtown, or could just stay on the bus to Angle Lake Station and ride Link from there. Kent riders would have a faster trip downtown than what they have now even with the KDM P&R transfer, but would also have vastly improved connectivity to the rest of south King County and Pierce County.

      Continuing to build Link would improve upon all this connectivity.

    2. +1E6 for EddieWiz
      The further LInk gets from Downtown Seattle the more absurd and useless it get. I can’t wait for the proposals to the 2032 legislature for extending Link to the fucking capitol bldg steps. That should make 3 or 4 legislators all gaga to fund the next pointless extension.

  16. No bloggers want to stick their neck out and tell which alignment they prefer? Interesting.

    1. Do the astute comments of the international award-winning peanut gallery mean nothing to you?

Comments are closed.