North Link Groundbreaking by Atomic Taco
North Link Groundbreaking by Atomic Taco

Sound Transit’s second quarter numbers are out, and the agency posted an 11% year-over-year gain in ridership.  ST Express Bus was up 11%, Central Link is up 10% to 26,268 weekly boardings, and Sounder is up 15% on weekdays.  Paratransit decreased 29% due to an accounting change.

The Sounder numbers are interesting because they come despite a decrease in event drips offset by an even larger increase in commuter trips. No doubt higher gas prices and falling unemployment are playing a role here and in the Central Link numbers.  Seattle unemployment is now at 7.8%, down 1% from a year ago and 2% from the 2009 peak.  Gas prices in Seattle peaked at $4.33/gallon in May, up $0.27 from May 2011.

Also, Sound Transit seems to be getting more efficient with its service hours. ST Express ridership increased 11% even with a 3% drop in overall revenue vehicle hours operated.  This lead to a 15% increase in boardings per revenue hour, bringing the cost per boarding down to $6.04 from $7.49, a whopping 21% drop. This efficiency comes at a price, of course, and Routes 540 and 560 were the ones paying it.  Both had their hours reduced and in turn showed double-digit decreases in ridership.

97 Replies to “Sound Transit 2Q 2012 Ridership Report”

  1. So back in the day when there were monthly Link Ridership posts there were some pretty graphs that charted ridership.

    Any way to get those back? I find it much easier to understand where ridership has been and where it is going with those things.

    Thanks for all the work you guys do!

      1. Thanks, but I liked the graphs STB used to put up better. Real easy to see year over year patterns and increases.

  2. It’s great to see the continued progress of ST, particularly with Link and Sounder. Ridership continues to grow, along with performance/timeliness, efficiency, fare-recovery, and a reduction in complains.

    I’m curious about the increase in Sounder though … I was under the impression that all of the P&R rides are regularly exceeding capacity. In other words, it seems that growth would have to come from people carpooling to the P&R or riding transit. Does anyone have insight into what is driving this increase? I’m hoping that we’re starting to see a trend away from single-occupant P&R riders more towards connecting via other transit methods.

    1. Good question. I think you’d need station-by-station breakdowns and train-by-train breakdowns to figure this out, and those weren’t provided in the summary Q2 report.

      There are some hints, though:
      – special event ridership was flat in the South corridor with 15% increases in commuter ridership;
      – but special event ridership in the *North* corridor was down and there were *19%* increases in commuter ridership.

      So part of this is that commuters are starting to use the North corridor more, which everyone agreed had excess capacity.

      I don’t know how the extra riders are getting to the trains on the South corridor, though. There are several possibilities:
      1 – as you suggested, carpooling;
      2 – as you suggested, they are taking buses;
      3 – perhaps they are walking or biking to the station;
      4 – reverse commuters from Seattle ‘out’ could be up (only the detailed numbers on which trains are being taken would tell);
      5 – the increase could be coming from Tacoma, which certainly hasn’t maxed out the parking lots in the vicinity of the station (only the detailed numbers on boardings per station would tell)

      Only more detailed numbers can answer how South Line riders are getting to the train station.

      1. Thought the numbers are old, the 2012 SIP had stop-by-stop boardings/alightings for all ST Express routes and similar station-by-station numbers for the trains.

      2. Unfortunately, there are no trendlines on those numbers (no 2011 numbers, etc.) — so although we have per-station numbers, we can’t figure out which stations have increasing ridership.

  3. Paratransit decreased 29% due to an accounting change.
    What accounting change would affect ridership figures? Are trips formerly considered under Sound Transit paratransit now counted under some other agency or not under paratransit?

    1. “Paratransit showed a major ridership decline during the quarter, dropping by 29 percent from 249 to 176 average daily boardings. The drop is largely due to a change in the way paratransit boardings are allocated to the Link corridor compared with other King County Metro-operated paratransit trips.”

      This is from the summary page for the quarterly report.

      Per this page, ST also appears to have responsibility for paratransit trips due to ST Express. The email address at the bottom of the page can probably provide you the detailed answer you seek.

      1. Ah. So it’s a question of whether they were “alternative-to-ST-Express” paratransit, “alternative-to-Link” paratransit, or “alternative-to-Metro-bus” paratransit. ST is responsible for the first two and not the last one.

        I guess previous accounting was not distinguishing them properly and was overcharging ST.

        It makes sense that Link would require less paratransit service than Metro buses, given that it’s a modern level-boarding train system.

      2. I think less paratransit required for Link is because of stop spacing. You have to be I think it’s within a 1/2 mile of regularly scheduled service. Fewer stations means fewer people eligible.

      3. Isn’t only 1/4 mile mandated, but Metro goes above and beyond to 1/2 mile? I seem to remember this coming up in the auditors report.

      4. Thanks.

        ST does have responsibility for ST Express paratransit? I thought they got a free pass on that because of an exception for commuter expresses – wasn’t there an STB post on that some time ago?

      5. You could be right about the 1/4 mile vs 1/2 mile. I know Metro uses the 1/2 mile and from what I’ve seen in my neighborhood is very generous about stretching that. The point is though that paratransit ridership didn’t decline by 29%. ST just declined to apportion as much money to Metro for provided the service.

      6. There’s crazy-complicated rules about what qualifies as “commuter” in the FTA regs for implementing the ADA.

        Frankly, the Tacoma-Seattle ST Express, which runs all day, does not really qualify as commuter under those rules. I don’t know whether Sound Transit is avoiding providing paratransit for that route, but I suspect if sued they would be forced to provide it.

  4. Cross-lake routes are all going way up, with most of them (including the 522) 19% up. Is capacity (i.e. number of trips) keeping up with ridership?

    As a side effect, will some tunnel routes have to get bumped to make room for more 550 runs?

    1. I wouldn’t think the 550 would get surfaced anytime soon. Im sure they were planning for this when they bumped the 212 and 217 instead.

      1. I was thinking other peak-hour routes would get surfaced, which, btw, could get done administratively, but not easily in the middle of a pick.

        This assumes, of course, that the 550 gets several more runs. ST has been slow to give SRO routes more than one new run at a time.

    1. Yup. Pictured is all 19 people who will be within walking distance of the post-upzone-fight Roosevelt station.

  5. Does “revenue hours” include time the buses spend deadheading to and from the base? And does “cost per boarding” include the cost of the deadheading, in addition to the cost of operating the bus while it is actually carrying passengers?

    Including deadhead hours in the service hours would have its effects on the productivity statistics. For instance, a full bus that operates one direction only, requiring N miles of deadhead for every N miles of passenger service is actually no more efficient than a bus that runs consistently half full in both directions of travel.

    I’m asking this because we need to be careful that our accounting metrics don’t discourage us from looking for relatively cheap ways to put deadhead buses into service. In particular here, I am thinking about routes 555 and 556. At present, the last 555 bus leaves Northgate TC just after 8:00, and if you don’t want to get up that early, you have to go downtown and transfer to the 550, which takes about twice as long. However, there are 556 buses arriving at Northgate at 8:30, 9:00, and 10:00, which currently just deadhead back to east base. Instead, these buses could be running a truncated 555 route, just to Bellevue TC, then doing a 1-mile deadhead, rather than a 15-mile deadhead. This additional service would cost almost nothing, since the buses are already going that way anyway, and would likely be quite attractive to the large number of Microsoft employees who, when the 8:00 555 is leaving Northgate, are just rolling out of bed.

    However, if we have accounting metrics that say that the cost of the Northgate->East base deadheading doesn’t “count”, the additional trips suddenly look much more expensive than it actually are.

    1. Related question – does anyone actually ride the 555/556 between Eastgate and Issaquah? If not, would it make sense to just delete that segment (which duplicates the 214/215/216/218/554) and reinvest the service hours on additional trips between Northgate and Eastgate?

      1. Yes, the route is quite popular with Issaquah peeps going to the U-District and downtown Bellevue. It’s significantly faster than the 271, which is the only other option.

      2. Post 2016, I doubt the 556 will get you from Issaquah to the U-district any faster than a 554->Link connection would – too much congestion and stoplights around Bellevue Transit Center. Perhaps that might make a good time to truncate the 555 to Eastgate.

        Of course, in 10 years, 555 and 556 will probably be replaced with Link entirely, as Northgate->Bellevue is expected to be just as fast on East Link as the 555 is today.

        In the meantime, I wonder what the actual ridership of that segment is. The report shows the overall number of boardings for each route, but they don’t break it down by section.

      3. Projected travel times for North Link and East Link show a tie between the 555 and the train during rush hour. Add in the transfer penalty to hop a bus from BTC to Issaquah and Link loses. Off peak the bus is 15 minutes faster and since the HOV lanes on 520 will be completed before East Link opens the bus will be more like 15-20 minutes faster at all times. Maybe even 30 minutes since the bus will no longer have any reason to detour through South Bellevue. That’s if of course WSDOT follows their own policy and makes the HOV lanes on I-405 useable during the peak commute.

      4. I’ve counted the riders on early 556 runs in the morning, the bus typically boards 15-20 at the Issaquah Highlands, another 15-20 at Issaquah Transit Center, and 10 or so at Eastgate, about 2/3rds passengers alight at BTC. The are more students on the bus on the way home in the afternoon, sometimes standing room only on articulated busses. Within a few years this route should run 20min headways during rush hour as it has shown stable growth over time. I would hope that it would be truncated when ULink opens and separate the Issaquah to Bellevue Segment,

      5. I’ve heard that one possibility for post-East Link I-90 bus service is to consolidate the 550, 554 and 556 in to one all-day express route from Issaquah to Bellevue via Eastgate and South Bellevue. Passengers headed for downtown Seattle, the U-District or Northgate would transfer at South Bellevue. Metro would probably continue to run some peak-hour express buses to downtown from Eastgate and Issaquah. But who knows what the situation will be 10 or 11 years from now.

      6. That would make sense. Would like to see an ST route that connects Issaquah to Redmond in the meantime. Maybe between U Link and East Link, tweak 556 to run Issaquah – Bellevue – Downtown Redmond; the connections between Bellevue and Downtown Redmond are paltry during the day; even with B line which can take 50 mins to go between downtown transit centers. The only other freeway alternative is 232 which runs a few times during rush hour. I know ST plans to extend 566 to RTC when funds allow but from the ridership between BTC and OTC, i doubt that will be soon.

      7. consolidate the 550, 554 and 556 in to one all-day express route from Issaquah to Bellevue via Eastgate and South Bellevue.

        Why go to South Bellevue instead of directly to BTC and let people transfer there that aren’t riding the bus to Bellevue to Bellevue. Of course if you want to cut costs and bump Link ridership just terminate at M.I. P&R. One thing about all these suggestions. Except for 550 riders from BTC, who will save between 1 and 7 minutes, people using the bus today will have longer commutes post Link.

      8. Both off-peak and peak, in a post-2016 world, the 556 has very little marginal value for Issaquah->U-district trips over Link.

        The 554 takes about 20 minutes to get downtown, add 5 minutes to walk down the stairs to the ID station can catch a train (ST could shave a minute off this if they wanted to by moving the inbound 554 stop south by one block). Then add 5 minutes for the train to get to Westlake, 6 minutes to get from Westlake to UW, and 5 minutes to walk up the escalator and across the bridge to campus. Total time is about 35-40 minutes, depending on how far you have to walk on the UW campus from the station, which is faster than the scheduled running time of the 556 today.

        The completion of the HOV lanes on 520 will help some, but the real problem with the 556 is that the detour into Bellevue TC on the way to the U-district eats up so much time – about 15 minutes of sitting at stoplights and traffic PLUS however long it takes to actually load and unload the passengers. And I don’t see a way to fix this without either causing an unacceptable inconvenience to car drivers or costing tens to hundreds of millions of dollars in new ramps.

        I’m not saying the 555/556 should skip this stop because it is a very important stop and is the destination of a large number of riders. But I am saying that given that the 555/556 has to serve this stop negates most of the time advantage of getting between the Eastside and the U-district.

        The real reason we need the 555 and 556 today is to work around the fact that making connections through downtown is so abysmally slow – at least 30 minutes between exiting I-90 on one bus and entering I-5 on another bus. This is the ultimate problem here, and is one that Link will go a long way towards addressing.

        Once Link goes to the U-district, the only bus routes we should need going from the Eastside to the U-district should the 271, 540, and 542 (maybe add 265/277 also). Everybody coming to the U-district from I-90 or points further south should be going downtown and riding Link instead.

        As for travel between downtown Bellevue and points further east, after the eastside service restructure of 2011, the 240 was rerouted to provide the first ever all-day bus that takes the most direct route possible between downtown Bellevue and Eastgate, without any detours. (It does have some stops along the route, but the bus blows past almost all of them as almost no one gets on or off there). For people going between Issaquah and Bellevue, you’ve got the choice of a transfer to an every-30-minute 240 or a very slow 271, so I think a peak-period express alternative is warranted. But after 2016, I don’t think this bus needs to continue on to the U-district – just turn around at Bellevue and use the saved service hours to increase frequency and/or span, or run additional 554 trips.

      9. Depending on the time of day the current running time for the 555/556 is 4-12 minutes faster than projected running time for Link BTC-UW. But it is true the final destination whether somewhere on campus or the U-Dist. can change that a great deal. Then you add the transfer penalty; and remember the feeder bus from Issaquah will be lucky to have today’s frequency. You’ll be adding a 1/2 hour to a lot of peoples daily commute which is already pushing two hours. But with East Link operational S. Bellevue can be skipped. Couple that with HOV lanes on 520 and the delta is even bigger.

      10. The comparison I was making wasn’t Bellevue TC to UW – it’s Issaquah TC to UW. I was basing my calculations on the assumption that you would have a frequent bus connecting Issaquah to downtown, without detouring to Bellevue TC, which would add lots of time to the trip.

        Once EastLink is up and running, this bus could potentially truncate at Mercer Island That opens up a whole can of worms, but I was mostly thinking about the 2016 scenario, where Link goes to the UW, but EastLink is not running yet. In this case, the transfer point would be the international district station downtown. If buses get kicked out of the tunnel, allowing the trains to cruise through the stops, the time penalty of going through downtown Seattle, even with the transfer, wouldn’t be any more than the time penalty of going through downtown Bellevue, without the transfer.

        The train->bus transfer for the outbound direction could cause issues if not done right, since it is unlikely the bus will be able to maintain a frequency as good as the train. The solution here is a true timed connection. What this means is upon the arrival of every other Link train at some tunnel station, there would be a 554 bus sitting next to the station waiting for it. The bus driver would be monitoring the real-time status of the trains and when the train arrives that has the timed connection to the bus, the driver would wait two minutes for anyone making the connection to get off the train and hop on the bus, then take off. Of course, the schedule would clearly indicate which Link trips would have the timed connection.

        Ideally, the transfer point and the western terminus of the 554 would be at the ID station, not Belltown. This avoids the unreliability inherent in traveling through downtown on surface streets and people whose ultimate destinations are in the north part of downtown would be able to take advantage of this same timed connection, making the train->bus transfer quick and painless. On the inbound trip, you might have to wait a few minutes for the train, but we’d be talking about an average wait of 3 minutes 15 seconds at current peak headways. And when East Link comes online, the average wait will be more like 2 minutes. A wait time that short should not be a dealbreaker.

        Of course, I’m not aware of any transit agency around here actually implementing a timed connection the right way ( a promise that if you’re on xxx train, yyy bus will be waiting for you at zzz station). Rather, the way everybody does it today is the bus leaves at a fixed arbitrary time on the clock no matter what and if you just miss it, too bad. While this scheme might seem to make it slightly more complicated for people boarding the bus directly at ID station (no train transfer), the experience would be no worse than today in which a bus that is scheduled to arrive at the ID station at 10:03 might actually arrive at 10:10 because of delays going through downtown. In fact, the 554’s punctuality would be considerably better than today, as the delta between the scheduled departure time and the actual departure time of the bus would now be based on the reliability of the train going through downtown, rather than that of a bus going through downtown.

      11. “Why go to South Bellevue instead of directly to BTC and let people transfer there that aren’t riding the bus to Bellevue to Bellevue.”

        Two reasons. One, transferring to East Link at Bellevue Transit Center instead of South Bellevue would add about 5 miles and 10 minutes to people’s trips. Two, there’s currently no way to get between I-405 and the Eastgate freeway stop. The bus would have to meander through Eastgate and then deal with the I-90/I-405 merge. That would negate any time saved by skipping South Bellevue.

      1. I see. So putting a deadhead bus into service is bad because it decreases the “riders per revenue hour” statistic.

      2. It decreases that particular metric, but it probably increases others, like “passenger miles per platform mile” (seen often in KCM reports); since I believe deadhead time is charged to the route that just finished, a good deadhead-to-revenue conversion would reduce cost per boarding (since a route’s cost includes its deadhead time). ST surely considers more than just one metric.

        Looking at a metric like “riders per revenue hour” that doesn’t consider the deadhead is a nice way to compare the performance of routes that terminate different distances from the base. And looking at the differences between efficiency metrics (using cost or platform hours) and popularity metrics (using revenue hours) can identify potential places to focus on efficiency improvements.

      3. platform hours is very important from a planning perspective on sound transit and their partners. You want to keep platform hours as close to revenue hours as possible.

  6. Hmm. Things which pop out at me, running from ‘most passengers’ to ‘fewest’ in the buses:
    (1) East Link can’t come soon enough, and the extension to Redmond should be accelerated. Look at those 545 and 550 passenger counts.
    (2) If only Sounder could be sped up enough to replace those 590/595 trips. Not seeing how to do it, though.
    (3) Lynnwood Link may actually be popular, based on Lynnwood bus ridership.
    (4)…but Woodinville is getting nearly as many passengers, with ridership growing faster. I bet the 522 will get redirected to Northgate when Northgate Link opens.

    People have been trying to account for the predicted ridership at Northgate. I expect that the 510/511/512/522 will be truncated at Northgate upon opening. (The 513 might stay.)

    That’s roughly 3.6 million passengers per year, which is nearly enough to account for the projected 15,000 passengers per day by itself. Now, the 510/511/512 will probably be truncated again when Lynnwood Link opens, so that would take away two thirds of that ridership, but there’s still the 1.2 million passengers a year from the 522.

    Things I noticed about the trains:
    (1) Central Link is by far the most popular single route ST has.
    (2) South Sounder is the second most popular, edging out the 550 and soundly beating the Tacoma-Seattle buses. Imagine what it could do with more speed and more frequency.
    (3) Tacoma Link gets a lot of riders for such a tiny line; perhaps this is because it distributes people from the train station into the active parts of the city. Tacoma Link expansion further into downtown Tacoma really should go forward.

    1. Regarding your 4th point, and the truncation of 522, I would argue that Roosevelt makes even more sense for truncation… avoid the congestion on Northgate Way, take some of the transfer pressure off of Northgate station, and provide better connections to Link for East Maple Leaf/western edge of Wedgwood.

      Regarding Tacoma Link, I would also imagine that some of this is because Tacoma Station also serves as a free “park and ride” for downtown Tacoma commuters. Parking in downtown Tacoma was a major hassle a few years ago, although this may have changed with the departure of Frank Russell.

      East Link shows a lot of potential when you consider existing 550 ridership, plus all of the “i won’t ride a bus but will ride a train” crowd.

      1. +1000 re transfer from the 522 to Link at Roosevelt…providing, of course, a layover area is found (perhaps continuing the few blocks on 65th from the station to the area under I-5?). This is a much better solution when the time often spent from Lake City to Northgate is factored in–and not even considering the zoo that the Northgate area is in the month+ around the holidays. The 75 will still provide LC-N’gate service at least until N 130th station is opened and cross-town service is instituted on 125th/130th.

        (the route Roosevelt–LC–Bothell–> was in the 1968 rail plan, which is still superior to what we’re getting at least until Seattle Subway comes into being)

    2. Re (4) — add in the 306/309/312 and that doubles the number of buses on that corridor. So ridership is probably a lot higher.

      Personally, I wish the north link extension was routed through Lake City proper and up along 522 towards Woodinville instead of along I-5 with a bunch of park and rides. Given the bus ridership, the route would obviously perform alright. But oh well.

      1. It would be nice to have a LINK route going from downtown Seattle to Woodinville. Then, you’d also have a way to have each route go through downtown to a southern or eastern destination. Woodinville to SeaTac for one route and Lynnwood to Redmond for the other route.

      2. Me too, regarding north Link–that’s why that route (or similar) was chosen for the better 1968 plan. But, as you, say–oh well. We’ll just need to keep pushing for the next better alternative.

    3. East Link can’t come soon enough, and the extension to Redmond should be accelerated. Look at those 545 and 550 passenger counts.

      While it would certainly be “nice” to be able to take the train to Redmond proper, rather than have it dead-end betwixt soulless office parks, there is absolutely in those numbers to suggest high ridership demand beyond Overlake.

      The cumulative quarterly boarding figures are there because they look impressive. Weekday averages (8,000 on the 545, 9,000 on the 550) are omitted because they are less impressive.

      North of Overlake is a tiny fraction of that. (Though I’ll be the first to admit that a straight ROW, rather than the current 10-minute loop through the TC and back to the highway, couldn’t hurt demand.)

      1. I’ve been on a lot of 545 trips that are crush loaded up to Overlake, at which point almost everybody on the bus gets off. North of Overlake, the bus is usually almost empty.

      2. (Rechecks)… yes, East Link as currently planned will get to Overlake. OK then!

        For some reason I momentarily thought it was segmented to stop in Bellevue. Perhaps because there’s been so much sound and fury in Bellevue and none from Bellevue to Overlake.

    4. I like the idea of truncating the 522. The question is will people give it a chance? When ST tried to create the 599, everybody just ignored it and drove directly to the Sounder station. At least on weekdays, limited parking at Northgate might makes things somewhat different here. However, on weekends, I fully expect everyone in Bothell and Kenmore to just drive directly to the train, leaving the truncated 522 almost completely empty, regardless of how frequently it runs.

      1. Suppose the 522 gets truncated, but instead of reinvesting the service hours saved by the truncation in increased frequency, we invest it in extending the route outward to accommodate the additional sprawl that will inevitably happen in the next 10-20 years. For example, imagine a 522 route that goes from Roosevelt to Monroe (with a peak extension to Snohomish, replacing CT 424), rather than downtown to Woodinville.

    5. “I bet the 522 will get redirected to Northgate when Northgate Link opens.”

      Doubtful because there’s no way to get from Lake City Way to Northgate TC without degrading Woodinville-downtown travel time significantly. The 75 takes 10 minutes from Lake City to Northgate TC and typically makes only 2-3 stops along the way, so the 522 can’t be much faster than that. The 522 was created specifically to bypass Northgate [1], so there may be pushback at making it go to Northgate again if it degrades travel time. Maybe it could be done in one of my patented Off-Peak Truncations (TM), but ST should really give something in return like more frequency.

      When 130th station opens, it would make a better meeting point. Downtown travel time would be about the same as now (not including transfer time). Northgate travel time would be faster than the bus going to Northgate TC (not including transfer time).

      [1] The 522 was previously an extension of the 41, then called the 307, running hourly.

      1. Northgate Way is most of the problem, yes. The 75 has additional problems of too many turns, but that’s mainly on the other side of the TC. Seattle has planned a street improvement on Northgate Way which will supposedly help buses get through traffic, but it’s hard to put a lot of hope in that.

      2. Huh? The 522 was the 307 pre-ST, yes, but it ran more often than hourly (particularly at peak) and wasn’t an extension of the 41. It got off I-5 at Lake City Way and ran north to Woodinville. The local version went through Northgate and then via 125th to Lake City Way and north.

    6. The 15,000 was for 2030, so Lynnwood will be getting all those riders, which at best is only 2,000 boardings per day. (1.2M / 2 (to get origin boarding) / 305 to convert annual to daily. So Lynnwood is still short 13,000 per day.
      I doubt CT/ST will take the heat for jumping out of the express lanes to slug it out to the Northgate platform, then have everyone wait for the next train at 8 min. headways, per the SIP yr 2017. Add the 10 min transfer penalty, and the uncertainty of giving up your warm seat for a packed train, standing, and I think riders will be vocal enough to squash the notion that transferring is a good idea – If you think the 42 riders were vocal, just wait and see.

      1. Pete, what about all the weekend riders? You have to deduct those if you want to divide by 260. It’s easier to just use 305, or in the case of Link which started out at 305, changed to 325 because weekend riders are much higher than predicted in relation to weekday riders.
        It’s just an estimate.

      2. Yes, your very right Bernie. I overlooked all those MFSTees wanting to live in Lynnwood, and take Link on the world tour to get to work – easily 13,000 or maybe even double that when you factor in the superior ride, more educated riders, and …. did I mention cup holders?

      3. …I don’t know. Depends on traffic on I-5, doesn’t it? Depends on how crowded the trains are, doesn’t it? In general, people *do* prefer to get off their bus and onto a train.

    7. i doubt the snohomish county st express buses will terminate at northgate. The 577 provides service from federal way to Seattle skipping light rail at seatac. Once st link goes to lynnwood expect it to replace the 511.

  7. I’m not one who buys that we need obsess over monthly Link fluctuations, but I am pleasantly surprised the year-on-year growth has remained so robust. If only the economy could do more than limp toward recovery and we could get that Graham St. infill station.

    1. Good questions both. What I can’t recall is whether they rebuilt the roadway in that stretch to accommodate an eventual station, or if all that work would have to be done as well. Other than that I don’t believe the individual stations cost all that much, although in the current climate anything is significant.

      I imagine, as quite the layperson, that coordinating the signal timings all over again would be time-consuming and complicated.

      I guess ST3 would be the best bet, politically.

      1. Whoops… looks like I missed this little set of sub-threads.

        Anyway, ASDF is 100% wrong, and this is one of those circumstances where all opinions are not “created equal”.

        Your words — “there’s a reason no working model of a walkshed extends up to a mile” — should be tattooed on some people’s foreheads.

        Anyway, to answer your immediate question, past STB discussions have revealed that when MLK was rebuilt curb-to-curb, provisions were not to allow for Graham to be easily added. No room for a center platform, no bowed lanes for side platforms.

        So we’re looking at a total rebuild (again) just to do what should have been done in the first place. Lame.

    2. Graham St. is a one-mile flat trip from either Othello or Columbia City station. If walking that mile is too slow, get a bike, skateboard, or kick scooter. Much cheaper than building a whole new station and slowing down everyone’s travel time to the airport.

      1. No, no, no. Two miles is way too far apart in that portion of that corridor.

        There are a lot of people who are never going to walk or bike or skateboard or kick scoot a mile each way, and that’s before you even start talking about the elderly, obese, and otherwise mobility-challenged. There’s a reason no working model of a walkshed extends up to a mile.

        And you’re only even within a mile of Columbia or Othello if you happen to live in that invisible house smack dab in the intersection of Graham and MLK, not if you live or work anywhere else along Graham. And once you leave MLK it is not flat.

        Not to mention that Graham Street is a commercial node that merits a station regardless, which is why ST planned for a station there before a shrinking budget forced its elimination.

        Adding a couple minutes of airport trip time to better space stations and serve a commercial district is not a big deal. On the other hand, shortchanging a bustling area full of transit dependents in order to function as a *minimally* faster airport shuttle is a bad trade.

        But we can agree to disagree. I’m not going to get into a whole thing. Maybe d.p. is in the mood.

      2. Could you imagine BART diverting to McArthur Ave in downtown Oakland to hit 4 or 5 stops, at grade, then jump back on the elevated for a quick trip out to Livermore.
        That’s as silly as designing Link to be fast to the airport, FedWay, and Tacoma, and after this little side trip down MLK.
        Link is trying to serve as a streetcar, heavy rail trunk, and rapid feeder from the exurbs. Jack of all trades, and master of none.
        I have to laugh when Northgate and Lynnwood stations are pegged at 15,000 each, but when ST decided they didn’t want to stop at S.Center Mall they estimate the ridership at only 3,000, so not worth the diversion to add a couple of minutes to the schedule.
        Amazing that S.Center is only 1/5 that of Northgate.
        It’s a much larger mall.
        It’s at the junction of 2 major freeways
        There are just as many possible truncation of routes there, maybe more with PT/ST riders, than at Northgate
        It has just as much, if not more TOD potential.
        It’s funny how where you want your stations generates the ridership numbers they end up with. Seems backwards to me.

      3. asdf-

        I bike and walk (and run) more than most people, and I probably walk and bike faster than at least 99.99% of the general population. If I had to walk a mile to get to a station, I could deal with it most days, but I also realize I’m an outlier in this regard, and I’d be much, much much, much happier with a quarter-mile to half-mile walk (or less).

        Realistically, if someone has to walk a mile (or more) to get to a station, then maybe walk to transfer to a bus, then walk to their destination (and then do the reverse on their return trip), a lot of people are going to say “Fuck this” and drive instead.

        I’m all in favor of improving and expanding Seattle’s bicycle infrastructure- right now cyclists (mostly) get the dregs of our cratered streets- sharrows and bike lanes smack in the middle of parked cars’ door zone. If Seattle can build safer streets and institute a well-thought-out bike share program, we could see a large increase in bike mode share in Seattle. However, there will still be many, many people who can’t or won’t bicycle most days. That’s just reality. And the transit system should stand on its own- we shouldn’t have to tell people living along a rail line that they have to buy a bike if they want to use the train.

      4. A Link stop a Southcenter would have required additional track miles, plus additional freeway crossings – not having it probably has more to do with a desire to minimize construction costs that to minimize travel time.

        As it is, Link to the airport is not that slow. Commercial buses that go between downtown and the airport advertize exactly the same travel time that Link does. The difference is that, while Link loses a little time going down MLK, it makes it back by going through downtown more quickly and not stopping at the front door of every hotel.

        As it is, the slowest part of Link is not MLK – it’s the section between Westlake and Stadium Station, where the train gets stuck behind buses in the tunnel. Today, the time it takes between going down the Stewart St. exit ramp on a bus and going by Stadium Station on a train is 20-30 minutes, depending on traffic downtown and how long you have to wait for the train. In the reverse direction when you’re waiting for a half-hourly bus, it’s longer. By contrast, driving that stretch on I-5 takes about 1-2 minutes, and if it takes 10-20, that means there’s a big traffic jam. The real bottleneck of Link to the airport is downtown, not MLK.

      5. “Everyone” on that train is not traveling to/from the airport. Some of us actually live or work in Southeast Seattle. Speeding up your travel time to the airport made the train significantly less useful for a large amount of SE Seattleites who aren’t really going to be able to ride a skateboard along MLK.

      6. asdf: “A Link stop a Southcenter would have required additional track miles, plus additional freeway crossings ”
        And we currently cross I-5 how many times to Lynnwood? A straight shot Link is Not.
        “As it is, Link to the airport is not that slow.” … unless you count the 30 min trip on the MT194 to Links 38 min. trip, or over 20% slower with about the same on-time performance record.
        “The real bottleneck of Link to the airport is downtown, not MLK”
        You can huff and puff all you want, but the buses are doing most of the heavy lifting under Seattle, not Link, so screwing a ton of riders just to make your trip to MLK isn’t going to happen until Northgate opens 10 years from now.

      7. One-mile spacing is hardly excessive for an urban rail system. I’m not a big fan of the Link routing, which seemed to be chosen to tunnel under every peak in South Seattle. But since that routing was chosen because it presumably serves population centers, it seems appropriate to stop at others.

        If a fast ride to the airport was the only objective, ST could have taken over the 194.

      8. These decisions were made at different times. You can’t say ST used exactly the same checklist to exclude Southcenter and include Northgate. Southcenter was decided before ST1, when we had to keep the price tag low enough or we’d risk not getting any rail at all and waiting another thirty years for our next chance. There was also a dispute between ST and either Tukwila or the mall about the land the track would go on; I’ve heard different things about who refused to deal. In any case, it was also recognized that Southcenter was not well-positioned for the primary north-south line (just as Kent isn’t), but it would be included in a future Burien-Renton line if that’s ever built.

        Northgate has been a major transit center for decades, and the neighborhood is much more willing to embrace an urban village and feeder buses. Those are the main reasons Northgate was a priority.

        The main inquity in Link is that Rainier Valley has a surface route while north Seattle-Lynnwood gets a big long tunnel and a fully grade-separated line, and East Link gets a mostly grade-separated line with maybe a downtown underground station. But again, these decisions were made years apart from each other, and based on the then-current conditions and local contexts, not on one unchanging checklist.

      9. While Northgate and South Center malls might be similar the land use around them is entirely different. A density map of Puget Sound shows South Center in a sea of low density blue. The whole Kent Valley could quite literally become a sea. Before the Locks lowered Lake Washington it used to drain to the sea via the Black River through Renton to the Duamish. The imfamous Southcenter hill and the 405/I-5 interchange make access a real challenge. Add to that the 400′ elevation change from the east side of I-5 to the west side and the engineering challenge would have been very expensive to solve. That’s why the tracks remain elevated for so far north of TIB.

      10. ““As it is, Link to the airport is not that slow.” … unless you count the 30 min trip on the MT194 to Links 38 min. trip, or over 20% slower with about the same on-time performance record.”

        No. Just because a trip is slower doesn’t mean that it is slow. Thirty-eight minutes from a major airport to a city core is still far from slow.

      11. If you’re visiting from out of town the hour or so to a DT hotel isn’t a big portion of your total trip; even if you’re only coming from Portland. But it’s not competitive with the ~20 drive time for a hotel shuttle often included in the room rate. for someone that lives in King County and isn’t close to a Link Station, almost everybody, it’s pretty rugged. From Bellevue it’s a half hour drive to SEA. Transit is an hour and a half if you park at S. Kirkland P&R and take either Metro 235 to BTC and the 560 to the airport or MT 255 to Link. About 1:20 either way. If you have someone drop you off it’s a half hour faster to take ST 560 which is an incredibly slow “express” bus but at least doesn’t involve a transfer. And East Link will be even slower by 7-9 minutes in travel time plus the transfer. At $5 billion dollars that’s a pretty poor ROI.

      12. Sure, it’s a “pretty poor ROI.”

        So long as you pretend Link’s sole purpose is to serve as a tourist and suburbanite airport shuttle, ignore the people and communities it serves along the way, and focus on the, um, “rugged” 55-minute two-seat ride coming to Bellevue instead of the 40-minute one-seat ride coming to Capitol Hill or the 44-minute one-seat ride coming to the UW. (I don’t live near Link, transfer to it for all my airport trips, and somehow never thought to describe my experience as “rugged.”)

        Link is not all things to all people. That is not breaking news. There will always be a place for hotel shuttles. And no airport train in the country is time-competitive with taxis or driving if your destination—as in your examples—is both in the burbs and not near a train station.

        Not saying Link is perfect. I’d prefer it run along Aurora and generally had stop-spacing that encourages TOD corridors instead of TOD nodes. But the downtown-to-airport trip times are the least of its problems.

      13. 44 minutes to the UW shows that it’s just as bad the other direction. Very few people on the eastside will ever live within walking distance of Link. Right now you can get from S. Kirkland to DT on the 255 in 22 minutes. Sure you’ll be able to park at the new lot on 130th assuming it’s not taken by 8am with people avoiding the cost of parking in DT Bellevue but link will take twice as long to Seattle. To the UW a staggering 4X as long. Let’s see, East Link won’t get you to the airport as well, it’s worse to DT Seattle and abysmal for the UW. On the bright side, travel time to the RV is much improved.

      14. If you think a line that provides U-Dub with a short ride to the U-District, a four-minute ride to Capitol Hill, a seven-minute ride to downtown, and 44-minute one-seat ride to the airport is “bad,” we’re never going to agree, and very few transit systems are going to satisfy you.

      15. And Bernie your entire post focuses on the fact that Link won’t do much for people who live in a town four or five miles north of a town where Link actually stops. This is news how? Yes, trains don’t go everywhere, and developing systems generally serve the densest areas and biggest destinations first. (Perhaps one day Link will cross 520. ST studied it. Then SDOT studied it for the mayor and confirmed what has been said by ST and by many here for years: right now it doesn’t make sense for a zillion reasons.)

        As for the town it *does* serve: Link is scheduled at about 22 minutes from Bellevue to downtown Seattle, and it’s another 12 or so from the ID to UW. On dedicated right-of-way with good headways. That’s not “abysmal,” that’s entirely reasonable.

        Again, the point of Link is not to shuttle tourists and Kirklandites (Kirklanders? Kirklandians?) to the airport, downtown, and UW with all those miles of pesky dense city whizzing by in a silent blur.

      16. Bernie,
        I don’t know why you would go downtown and transfer to Link when you could take something to the U. District and transfer.

      17. “unless you count the 30 min trip on the MT194 to Links 38 min. trip, or over 20% slower with about the same on-time performance record.”

        It’s funny how the 194 gets faster every time someone brings it up. One of these days no one will remember anymore and people like Mic and Norman can claim it made it to the airport in 10 minutes.

        The 194, on it’s best day, took 34 minutes from Westlake to the airport. On a bad day, which was every time I rode it, it was closer to an hour. Link takes 4 minutes longer, while making 6 additional stops. Not a bad trade-off in my opinion.

      18. I had 194 trips (mid-day, weekend) that did it in 27 minutes. Honest.

        Nevertheless, as someone who usually lands in the evening Link is about 1-1000x preferable to the thing that used to run half-hourly and then die at 9:30.

      19. I don’t know why you would go downtown and transfer to Link when you could take something to the U. District and transfer.

        It depends on how routes are restructured after 520 is rebuilt and Link is operational. Right now the 255 is a tunnel bus so the transfer is easy. It looks like peak buses across 520 may go to UW instead of DT but off peak will use a reconfigured flyer stop. Either way East Link is useless. Worse than useless if it kicks superior bus service out of the tunnel.

      20. We’ve moved from “abysmal” to “useless”? Classic. Again, you’re talking about a town miles away from where Link actually stops, Bernie. It’s also useless for a West Seattle-Ballard commute. So what? That a line could be extended or a system expanded is not in itself a remotely valid criticism of the existing line or system.

      21. “I had 194 trips (mid-day, weekend) that did it in 27 minutes.”

        And then another 27 minutes at the airport while people fumbled for change and drug their luggage up the aisle. :-) But it was still preferable to risking your life and sanity on the 174.

      22. The capacity of Link is a nice asset, in and of itself.

        Before, I’ve occasionally had crush-loaded 194’s actually pass me up in the tunnel, with a 15 minute wait for the next one, on top of the 10 minutes I’ve already waited.

  8. I’m very much looking forward to the stop-by-stop numbers in the 2013 SIP.

    Looking at the 2012 numbers for the 560 (which will be under serious consideration for revision or elimination), a rather dramatic point of math stuck out: Only 7-8% of the riders boarding east of the airport and heading south end up alighting at the airport. Even more dramatically, fewer than 5% getting on west of the airport and riding south end up alighting at the airport. So, to the dozen or so who ride the bus from West Seattle to the airport per day, I challenge you to defend the continued existence of the 560. The combination of the C Line and 120 provides better service on the corridor between Burien and the Alaska Junction.

    Similarly, most of the ridership on the eastern portion of the 560 is traveling on the same path covered by the 566. It’s time to give the 566 more service, and let the F Line provide (hopefully) fast service between Renton and Burien.

    1. I will also give a clearer picture of how adding ULink will effect the metrics of rail and bus travel, as years 2017 and 2018 hours and costs will be included. This includes the highest ridership stations over the first couple of years of operation, which should put to rest a lot of the speculation over frequency, service, buses in the DSTT, and ridership.
      That document, plus the soon to be released ‘Before and After Study’ on how Central Link has compared to initial projections made 10 years ago are working out, will either confirm current trajectories or require some significant mid-course corrections.

    2. On weekends, the 560 is the only way to get from Bellevue to Renton in anything less than the full hour the 240 is scheduled to take.

      I could see getting rid of the 560 if the service hours could be reinvested to provide more frequent 566 service, as well as operate the 566 on weekends.

      However, Renton to airport trips would suffer, as the F-line, as I’ve seen is circuitous enough that it’s impossible to avoid it being much, much slower than the 560, even if it is more frequent.

      Nevertheless, downtown Renton to the airport is just under 7 miles, which is about $20/15 minutes on a taxi – still a bargain compared to driving and parking, especially if you can take the F-line on the way back, when you’re not under time pressure to catch a plane. Given that so few people are riding the 560 to the airport, maybe just letting the Rentonites choose between the F-line and a taxi and the Bellvuites take 550->Link, that’s good enough.

Comments are closed.