CentralEast_KBIBUDUKR_Lev2_060514_Final_Exec.9Sound Transit’s Executive Committee met earlier this month and discussed preliminary analysis of the entire corridor from Ballard to the Eastside via SR 520. Unlike the Ballard/UW segment I discussed on Saturday, continuing to the Eastside involves extraordinary technical challenges, fundamental geometric dilemmas, and relatively mediocre ridership. In a few days we’ll report on the possibilities to connect Kirkland, Bellevue, and Issaquah.

The geometric problem is that no destination obviously dominates. Bellevue is the largest city but is already well served by East Link. Redmond is probably the cheapest to serve, but is also a peculiar place for two light rail lines to converge. Kirkland is the city with no planned high-capacity transit, but it is difficult to serve its dense areas.

There are three light rail options. All of them tunnel on 45th St. after U-District station, run elevated down Montlake Blvd, and elevated for a short run east of the bridge. All other segments are at-grade.

B2b turns north to serve Kirkland to Totem Lake, via the Eastside Rail Corridor.  At $2.1-2.9 billion, this most expensive option would generate only 7000-9000 riders and take 18-23 minutes end-to-end.  Although it makes sense to use existing right-of-way between stations, it seems this corridor will only make sense if the line deviates into dense neighborhoods near station areas.

C2 and C2a both turn South on the Eastside corridor. C2 then turns east to interline with East Link to Redmond, and is the most cost-effective of these options: $1.9-2.5 billion, 18,000-22,000 riders, and a 25-31 minute travel time. C2a interlines in the other direction and stops at Hospital Station. It’s about the same in cost ($1.9-$2.6 billion) but much worse in terms of ridership (9000-11,000).

There are also 3 BRT options: C1 is basically an enhanced 542 to Redmond ($55m, 10,000-13,000) with “potential reliability issues.” B1a ($210m, 7000-9000) is a bus version of B2b. A1 ($400m, 7000-9000) bypasses downtown Kirkland altogether and runs straight up I-405 to Totem Lake.

What makes these largely at-grade options relatively expensive is the enormous complexity of getting from U-District Station onto the 520 bridge deck without using the overtasked U-Link tunnel. Traversing the UW campus reopens old agreements, various concerns about labs, tribal burial grounds, wetlands, utility lines, and views. A new Montlake cut crossing is expensive and introduces possible environmental and view impacts in an area with a very litigious track record. And of course, there’s the expanded bridge deck on SR520 itself.

Of the three communities, I’m most sympathetic to the need to serve Kirkland. However, wedding it to a highly problematic 520 crossing may not be feasible. The money spent on this crossing might better be used to deviate from the Eastside Rail Corridor where necessary to serve the true ridership generators in Kirkland.

CentralEast_KBIBUDUKR_Lev2_060514_Final_Exec.8

98 Replies to “Sound Transit’s 520/Eastside Options”

  1. At this time and probably for at least another 20 years, it makes no sense to cross SR-520 with light rail. Buses should be able to meet demand for some time. If and when LRT is built from Kirkland to Issaquah, LRT on the bridge might make more sense if it could connect into an already built B2b or C2a segment.

    1. A couple of years back, didn’t we discuss forgetting about light rail over 520, and discuss a line across a new rail bridge to Kirkland from Sand Point?

      And didn’t we agree that the 520 corridor would always be too lightly populated to make be worth the expense?

      Most natural seems to be:

      1. Surface central reservation from Downtown Ballard to Phinney Hill,

      2. Subway under Wallingford through the U-District,

      3. Surface central reservation through Laurelhurst to Sand Point,

      4. Bridge from Sand Point to Kirkland,

      5. Best route and structure to Redmond.

      Careful though- usual prejudice here: for me, best thing about this routing is that it will be both really great to drive and impossible to automate.

      Mark Dublin

      1. Could a mixed at grade/separated route be done in a “partially automated” way?
        e.g., for the sake of argument, let’s say that Central Link was grade-separated from the Airport to Rainier Beach station. You could have it run automated from the Airport, then at Rainier Beach a driver steps on for the at-grade segment along MLK, and then she steps off at the ID station, lets the train continue automatedly to Westlake. Meanwhile, the driver crosses over to the other side, and takes control of the next train to return from Westlake.

        This would mean that for each 39-minute scheduled run, a driver would only need to be on-board for 20 minutes of it, so you could get away with significantly fewer drivers per line. When U-link opens, this would have even greater savings.

    2. Maybe by the time it makes sense to re-examine the cross-lake corridor, Ben’s idea of crossing from Sand Point to Moss Bay will be be part of the picture. It could make sense if it connects into B2b.

      1. I don’t think we’ll see another Lake Washington crossing unless it somehow gets the kind of political support the rebuild of 520 has gotten.

        I suppose ST could dig a tunnel too but the cost and complexity of that project would be huge.

      1. I can’t find this online anywhere (the “Trans Lake Washington Study”). The closest thing I can find are some materials on rebuilding the bridges, and a pdf that mentions a potential ferry between Sand Point and Kirkland.

        It seems there is some offline print document… somewhere.

        Is KyleK hunting down that document?

  2. One caution: U-District station will probably have to be mined under the present station, with vertical transfer. But might still be cheaper than dealing with Montlake, the Arboretum,and 520.

    MD

    1. Tower 18 in Chicago has a 5 lines converge at grade, and at peak period sees trains every 30 to 45 seconds. Would the U District station really see so many trains that it needs separating?

      1. Glad somebody remembers Chicago. But also remember that like entire rest of the world aside from streetcar-served hill towns in Portugal, every place else is flatter and wider than Seattle. Chicago is where the great plains meet Lake Michigan.

        So multiline junction at U-District would either require an underground chamber like Grimalkin’s Kingdom under the whole American west on Gene Autry show, or leveling whole U-District.

        If that would recreate Chicago, though- transitwise it could be worth it.

        MD

      2. It’s a lot easier to do that on an above-ground structure rather than 75′ underground. There might be difficulties building diamonds and taking apart the bored tunnels to build a cavern for them. A “+” shaped station could make more sense where N-S trains are on one level, and E-W are on another.

      3. There are some pretty complex junctions underground in New York (NYCTA as well as the main lines leading to Grand Central and NYP). I’m not sure being underground makes that big a difference.

      4. Came back from Lisbon, Portugal 2 months ago. Walking the streets especially down near the shore it felt a lot hillier than Seattle. And yes, those streetcars were navigating those steep streets. However, the Lisbon underground subway is extensive with four lines and at least five stations that intersect two lines. One of those intersections, Baixa Chiado, came very close to the shore.

        If the Portuguese and the EU can do it, why not us?

      5. The question is if it would be any more or less expensive. Split level stations can be fairly expensive because one line has to drop well below the other. In this case, both lines then have to climb back to the surface – one to go to Northgate on an elevated structure and the other to get over the lake. Top of rail is what at current station? 50 feet below the lake surface? Drop below that, and you are looking at maybe climbing 70 feet to get it out from under there to get across the bridge.

      6. snip:
        “top of rail is what at current station? 50 feet below the lake surface?”

        answer:
        using sea level as a baseline, Lake Washington’s altitude is about 30 feet; street level at the Brooklyn (U-District) Station is at about 185 feet altitude. The ST website lists that station’s platforms as 80 feet below the surface — so let’s just say that the rails are 5 feet lower than that. 185 minus 85 equals rails at an altitude of 100 feet; aprox 70 feet “higher” than the surface of Lake Washington.

        A West–>East line could still pass UNDER the North–>South line and continue at a level grade to daylight; roughly coming out of the bluff at/under/below/beside the NE 45th Street Trestle Bridge just west of 25th Ave. NE..

        so no, I don’t think there’s much challenge for this Ballard–> U-District line coming to the surface for a lake crossing — but to me it’d make more sense to turn north at that point and head for Lake City and around the top of the lake through Kenmore if you want to get to Kirkland / “Eastside North”.

        on a related note, the design and construction of the Ballard/U-District line would seem a lot easier to build/bore –and result in a more useful transit segment– if the ends were designed to be 25th Ave NE and Shilshole (with a station connecting to Sounder/Amtrak there. It could be a nearly dead-flat grade along the entire length, coming out level into daylight at both ends.

    2. Eh? I see no reason a station couldn’t be built along 45th. You’d only be dealing with the tunnel bores at that point and crossing under or over them isn’t rocket science. For that matter the station box could start East of the bores so as to eliminate that complexity.

      1. There are also building foundations on three sides of U-District station, including UW Tower, starting five feet from the station box.

  3. Those ridership numbers: are they after completion or are they 20 years after completion? It takes some time for development patterns to settle around these lines once they are built. The double digit increases seen in the current Link segment shows how long it will be before development patterns will be settled along it.

  4. As with the other Eastside study, I’m suspicious of this one. Considering the 3 LRT options, we essentially have rail versions of these 3 routes:
    271 (truncated to Bellevue)
    255 (truncated to Totem Lake)
    542

    I would assume, based on # of trips per day, that the 255/271 both have much higher ridership than the 542. Why then does the 542 route have the highest potential ridership? I’m wondering if they’re counting cannibalization from the existing East Link as new riders or not — but even so, it sounds strange.

    The thing that’s most disappointing is we’re only looking at one segment. In an ideal world we’d be looking at a Ballard -> Fremont -> Wallingford -> Udistrict -> 520 line. This would open up a huge number of trip pairs that are not easy right now. As a resident of the eastside I can say I hate have to go to Ballard because all options take at least an hour if not more. However — if there was a route that connected these destinations directly, I think there would be a lot more new riders.

    Finally, their routing in Kirkland makes no sense. If we went to the expense of building LRT in Kirkland, it would have to serve downtown Kirkland. It sounds like they’re intentionally sabotaging that route, because the 255 has comparable performance compared to the 271 (and I doubt it’s the part of the route post-totem lake that’s making the difference).

    1. They are assuming that a significant portion of 545 ridership will shift to the 520 rail line, as well as all 271, 540, and 542 riders. The 545 is the Eastside bus route with the second-highest ridership after the 550 (which is being replaced by East Link).

    2. I agree the numbers seem a bit suspicious here, especially as they handicap the options serving Downtown Kirkland by not putting stations near where the people actually are. Even the BRT option on the ERC doesn’t deviate into downtown Kirkland. That should be relatively straightforward as you would only need access ramps on either side of downtown.

      I wonder what the ‘real’ numbers would look like if activity centers North of 520 were actually properly served.

      Still I think it is useful in showing there aren’t a lot of really good winners for future transit projects on the East Side other than extending to downtown Redmond.

  5. If this were viewed as a component to a Ballard-Wallingford-UDist-520 Corridor-Bellevue-Issaquah line, that would make a lot more sense. (When is the Issaquah study being released?) Take advantage of the center platform at South Bellevue for transfers, get good interlacing and frequencies though Bellevue, and adds several destinations to the regional system. Note a couple of the O&M options have a wye facing north towards Kirkland. There might even be (there better be) accommodations in the aerial guideway for a wye in the vicinity of South Bellevue.

    Remember, Issaquah trains can’t go to Seattle via I-90; the DSTT will be full.

    1. Then their passengers can’t either, whether you’re talking about transferring at South Bellevue or Brooklyn. People forget that “trains” are proxies for groups of people. If the DSTT will be full from trains necessary to accommodate riders from the existing lines, then those same trains can’t accommodate new transfer passengers.

      Presumably some riders from Issaquah will be transferring from buses to Link at South Bellevue when East Link opens. But will all Issaquah-Seattle service be truncated? If not, then it probably would be when an Issaquah-Bellevue line is opened, and the folks from those buses presumably would not be able to be accommodated.

      If a new CBD tunnel for the Ballard-West Seattle line is built perhaps some South Link trains could be re-routed to use it, freeing some capacity for more East Link trains, but I haven’t seen any discussion of a cross-link between the lines.

      1. Not really, the problem for the DSTT capacity is the service to/from the North not the South or the East. Past that the headways on the I-90 bridge are limited because only one train can be on the bridge at a time (I believe this works out to 1 train every 9-10 min in each direction). The headways in Rainer Valley are limited by the surface alignment (one train every 5-6 minutes).

        If you run Issaquah trains into Seattle then you limit the headways on each branch to every 20 minutes at best which really sucks. It is far better to force a rail-rail transfer and maintain reasonable service frequencies on both branches.

        If either (or both) UW/Ballard and UW/520/Eastside are built then there could be an issue with N/S trains between downtown and the U-District being too full for all of the riders, especially at peak.

      2. So you’re saying the ST is going to run less than full trains on East and South Link during the peak hours rather than running turnbacks off North Link? Why would they do that, and why do you believe that to be the case?

        If in fact the projected ridership from North Link is significantly larger than the combined ridership from South and East Link then running turnbacks is almost required.

        If it isn’t significantly larger and a three minute headway is necessary from North Link, then whatever fraction of trains cross the lake on East Link would be full at the peaks.

      3. The difference between empty and full trains is 500-800 people (comfortable SRO – manufacturer’s maximum). So there will always be unfull trains because people don’t travel in groups of 500. And they’d better not be full or we’ll have failed to provide adequate capacity through 2060. Non-full trains take the same space as full trains. They could decouple cars but that’s labor-intensive and requires a trip to the base. So there will be at least some room for transferees. Enough for everyone? I don’t know.

    2. For the cost of a full wye at the entrance to ID station, Issaquah-Seattle trains could go south on Central Link toward the airport and wouldn’t need to go into the DSTT.

      1. The issue is constructability of the wye. There’s a huge triple-deck roadway and parking garage in the way. They could try to build something that went down Plummer St, but then the Link ramps to the DSTT would be in the way and those already have a steep grade on them. Then, would it be worth building a Issaquah-Airport line that skips both downtown Seattle and Bellevue as well as the UDistrict?

      2. Mike it also depends on how far east you want to place that Y on Eastlink.

        A Y at the I-90/Rainier station would let you take Issiquah trains south, and join the main line at Mt. Baker

        A Y located just west of I-5 would allow you to either go over or under I-90, across the Bus parking Just south of I-90 and join the mainline (probably south of the Stadium station)

        Really it is imposable to judge until we see the 100% designs for Eastlink between Rainier and the ID station

      3. The question is why would you connect the two lowest ridership branches of the network and limit headways toward the largest destinations (Downtown, UW) just to save a handful of riders a transfer?

  6. All of the options seem to go across the 520 bridge. Do all options involve running at grade in mixed traffic? Or would the light rail take lanes away from the bridge? Why wouldn’t BRT do the same thing?

    The light rail is way more reliable than BRT because they get their own lane… Why doesn’t BRT? Is the assumption that politically it’s a lot easier to force people to stay off light rail tracks than to force them to stay off the BRT lane?

    And, if the LRT is taking lanes away, isn’t that going to slow down all the non-LRT transit options on 520? It just seems like we’re not getting apples-to-apples comparisons.

    I guess it could be that the reliability issues for the BRT is in getting from Husky station to 520, and from 520 to its destinations. I’m still confused though.

    1. BRT does get its own lane shared with HOV on the 520 bridge. Light rail would displace both. For at least the next 20-30 years, rail makes no sense on 520. Other investments are more urgent.

    2. It mentioned snap-on pontoons for light rail, which I guess means widening the bridge. The new bridge is adding two HOV lanes, so BRT would go in those. Widening the bridge for bus lanes would raise the cost significantly, which contradicts the goal of BRT.

    3. I agree. BRT proposals from Sound Transit always seem to be the cheapest way possible, while light rail proposals (more recently) seem to be fully grade separated. As mentioned, you already have grade separation (or very close to it) as HOV lanes much of the way. So why not propose (and study the cost) of getting grade separated bus lanes from 520 to Husky Stadium? It might not be worth it, but my guess is that is would be a lot cheaper than rail, while providing most of the advantages (a very fast ride to the east side from Husky Stadium). As you suggested, it may be that BRT lanes would be unpopular, while light rail would be an easier pill to swallow for Montlake residents.

      Have we forgotten that the best thing this city every built was a bus tunnel under downtown? Long before it ever had a train, it carried thousands of riders each day through downtown. It has saved way more time for transit users as a bus tunnel than anything Sound Transit has ever built. Light rail is great, and certainly appropriate for much of the city, but better bus service would make a huge difference to a lot of people.

      1. The existing Montlake exit ramp has wide enough shoulders so one could probably carve out a bus lane with a minimal amount of additional pavement.

        Simply getting the bus past the traffic jams on the exit ramps would insulate them from most of the delays. There’s still the occasional bridge opening, but one can’t have everythi.

      2. Good idea. That is exactly what I’m talking about.

        Montlake boulevard is sometimes clogged, but unless you build another bridge, there isn’t much we can do about that either. That would be the ideal, actually. Build another bridge, to the east of the Montlake bridge and connect 520 right to the back end of the stadium parking lot, and thus right to the station. That is all public property, so Montlake takes another hit, so I’m sure there would be local opposition. Not that a bridge is cheap, either; but for a lot less than light rail you could have a seamless connection between light rail and buses that doesn’t get bothered by traffic for much of their trip.

      3. A second bridge for transit would be nice, but it would be quite expensive. Given that we could get the bulk of the benefit for a fraction of the cost by simply adding a bus lane to the existing exit ramp, an additional ship canal crossing does not look very cost-effective. I am also not thrilled about the impact a second ship canal bridge would have on the landscape.

  7. Perhaps it’s time for me to get educated on how ST arrives at its ridership predictions. I try and defer to the experts and their data, even if I don’t like what they tell me, but in this instance I have no idea how the numbers could be so low.

    The 520 Link not only serves A) the UW to Eastside commute and the B) Kirkland/Redmond to Bellevue commute, it C) makes the UW to Ballard line worthwhile and especially D) opens up the entire north end to commuting to the eastside by rail.

    This line would open well after Link had been piped up past Northgate and into Lynwood, and roughly concurrently with ST rolling into downtown Everett. That should relieve considerable pressure off of 405, and put Snohomish real estate on the map for people currently working at the eastside’s larger employment centers. And that’s on Day One, before development patterns re-adjust.

    Since the plan all along was to convert 520’s “existing” HOV lanes to rail, I would think the cost-to-benefit ratio would be substantially favorable.

    1. I’m curious too about Sound Transit’s methodology. Some of the recent studies have me scratching my head a bit, especially the low numbers shown for Ballard/Downtown (and the high numbers shown for some of the suburban proposals).

      I’d love to know what the segment numbers for each of downtown tunnel, Downtown/Ballard LRT options, Ballard/UW A3, and UW/Eastside LRT options look like if they are connected as a single line.

      1. I wouldn’t say that I find ST’s first-round numerical algorithms universally persuasive (either ridership, or trip times on contorted ROWs), but nothing about this terrible segment surprises me.

        Mostly what any version of 520 rail does is to A) cannibalize East Link. And East Link hasn’t exactly proven the gold standard of cost-benefit ratio to begin with.

        The fact that anyone could opine, with a straight face, that extra miles of floating rail to the sprawl would be required to “make worthwhile” the three most desperate-for-transit urban miles in this mobility-kneecapped city, exemplifies how broken the conversation is, and should help to illustrate why this blog leaves me so frequently incensed.

        That “makes worthwhile” claim is not a real thing. That anyone still thinks it’s a real thing reveals gross failures of both education and advocacy.

        And Chris, while I’d have liked Ballard-downtown’s numbers to look better, their weakness doesn’t surprise me either. All of the north-south options have hobbled walksheds, troublesome transfer potential, and a depressingly small catchment north of the Ship Canal. That line just isn’t as useful as it should be, as the comparative strength-per-dollar-spent of the Ballard-UW study helps to illuminate.

      2. I’m fine sticking with buses along 520, given that the post-construction HOV corridor should be pretty reliable and the demand is within the limits of what can be reasonably handled with frequent buses.

        However, it is not reasonable to expect everyone traveling between Redmond and the U-district (or anywhere in north Seattle) to go all the way around through downtown and take East Link. In contrast to the 194 debate a few years ago, Link would take nearly 30 minutes longer to get from the U-district to Overlake than a bus along 520 would.

      3. Very good point, asdf. This is why Sound Transit should focus on the “last mile” problem in the U-District. Had we done things right, there would be a station at 520, and riders could walk (at 520 level) from the train to the bus stop. But there isn’t, and my guess is there never will be (because we — sorry — Sound Transit — never though about it) so the line in there is at an angle, and you can’t put a station in there, unless you spend a lot of money retrofitting U-Link. Be that as it may, I would love to study the whole problem (from both ends).

      4. I would tend to agree with that, ASDF.

        But the heavy lifting that East Link will be doing — the thing that requires bi-directional rail capacity and infrastructure — is Seattle-Bellevue, and to a lesser extent Seattle-Overlake. The first is taken care of completely (albeit with flaws). The second is taken care of adequately (albeit indirectly). 520 rail cannibalizes both (with a transfer to downtown or Capitol Hill), while newly enabling faster point-to-point trips for a relative few.

        The other great flaw in Anon’s comment — and in all the “regionalist” logic — is his justification “D)”. No, 520 rail doesn’t “open up the entire north end to commuting to the eastside”. Not by a long shot. Because getting from and around the middle-density parts of the “north end” is set to remain a slog, and because so few workplaces on the eastside are remotely designed for transit access.

        People will use a bus of reasonable quality to feed into a train system of reasonable speed, on their way to another part of the urban area, or to a major destination like the airport. They absolutely will not bus feeder to a really, really long train ride to another bus (or unpleasant walk) to a suburban endpoint. When things get that complicated, you get back in your car. And that’s why suburban rapid-transit schemes largely fail.

        As Ross says, Sound Transit needs to creatively and economically solve suburban corridors and last-mile access where it can. But expecting Bitter Lakers to start taking transit to Crossroads because the middle 1/3 of the journey becomes a train is no more than “faith-based planning”. Even ST’s roughest research shows that to be the case.

      5. The ideal solution to the “last mile” gap between the UW station and 520 would be to combine the 542 and 545 into one route that would operate between Redmond and the UW station, using the 520 HOV lanes, plus a dedicated bus lane down the Montlake exit ramp.

        Ideally, the stadium parking lot right next to the station would have consisted of a sheltered bus stop right next to the station, plus ample room for all those buses to turn around and layover between runs. (With Link going on to the U-district station much more quickly than a bus, there is little point in paying the service hours to make the 542 continue that slog).

        Instead, we have valuable real estate right next to the station being used for car parking, not even by commuters riding the train, but by people who want to drive all the way to avoid the train.

      6. Except that if you drive from Bitter Lake to the Eastside, bridge traffic, tolls, and I-5 traffic are in between. Driving may be a less annoying experience (although I would take the train), but it’s not a pleasant experience.

      7. So you would spend a couple of billion, knowing that you’d probably still fail to be the least-worst of the travel options?

    2. The one thing that I find unexplained here is whether or not the LRT trains actually go to the Downtown Bellevue station or stop at Hospital Station in Bellevue. If the additional one-station transfer is forced in the forecasts, that’s a travel time loss of at least five minutes to Downtown Bellevue jobs. That single assumption would create lower ridership.

      1. We’re on the same path: I just noticed that myself.

        None of these appear go to the Downtown station. Moreover, the reason why C2=C2a in cost is because the two lines are exactly alike, except for the stub connection in C2a to Hospital Station.

        My thoughts are to truncate C2 to where it cuts out the east of 405 section (several miles of track), and simply connect it to either A) the downtown station or B) the hospital station. I would think this brings down the costs dramatically, but still serves the immense numbers brought in from the expanding Link network that is already planned to run out of Seattle and into North King and Central Snohomish.

      2. Really all I was expecting was a light rail line connecting a station in U-Dist to the downtown Bellevue station, with at least one new station (which would serve south Kirkland). Are there logistical reasons why it can’t be as simple as that?

      3. Because Kirkland needs something and Bellevue will already have East Link. Do you expect Kirklandites to take a bus to South Kirkland P&R and transfer, in a place that’s only slightly more pleasant than the Montlake freeway stop? Or will they still take an express bus from Kirkland to Seattle, and wonder why they should pay for two Seattle-Bellevue lines when they second-largest city on the Eastside doesn’t have any lines?

      4. What Sound Transit can reasonably give Kirkland without overspending on boondoggles:
        – replace Metro route 255 with a Sound Transit route and run it more frequently than existing route 255 (especially during the evening). (Or a truncated version that would connect to Link at UW station to go downtown).
        – Nonstop shuttle between downtown Kirkland and downtown Bellevue via the I-405 HOV lane.

        (Note: Microsoft already provides a nonstop shuttle between downtown Kirkland and the Redmond campus, so there is no need for Sound Transit to duplicate this).

      5. “Microsoft already provides a nonstop shuttle between downtown Kirkland and the Redmond campus, so there is no need for Sound Transit to duplicate this.”

        That works for you, but not for the general public.

  8. Now we see another example of how the absence of a SR 520 station option on U-Link has come back to bite us. It does make me wonder how the cost of adding this would affect things, though. Right now, the corridors are all pushed through the UW campus.

    1. A station by 520 would have made a lot of sense. You could even connect a station at the bus stop level. This would mean a U-Link rider could transfer to a bus on 520 without going up to Montlake Boulevard and back down again.

      My guess is that it is too late for that now. Sound Transit has really done a poor job of considering future improvements. It is building this piece meal, and at each turn seems to be oblivious as to what should be built next. It doesn’t cost much to make the section flat as a possible future stop; but if you don’t do it initially, it costs a lot more to do it later.

    2. Would you add a 520 station or move UW station? Adding a station would put two stations practically next to each other. Moving the station would put the entrance in a freeway no-mans-land, in a single-family area, a longer walk from UW. That’s a recipe for substandard ridership, and is what we avoided by not putting University Link on Eastlake/I-5.

      1. The 520 station important to many people today, especially those going to/from the south on Routes 48 or 43 and connecting to Translake buses.

      2. Add a station. Good God, add a station. The time penalty is minimal. Think about it. Imagine a bullet train going from Tokyo to Kyota. Very fast, right? Now add a stop half way between. You’ve slowed the line considerably. It has to go from maximum speed (275 MPH) to zero and then accelerate back again.

        Now add a stop a half mile from Tokyo. The train at that point is not going top speed (unless things have gone terribly wrong). The stop does very little to its average speed. The only time penalty is for the time spent loading and unloading passengers. This is a really easy thing to manage. If people don’t use the station, keep going; if they do, then spend a little time at the stop. My guess is they would, but I agree with you, more would use the station next to the hospital (because its next to a hospital and the south end of a major university).

        By the way, University Street station and Westlake are much closer than these stations are (as are I. D. and Pioneer Square) but I have yet to hear anyone propose that we skip downtown stations because they are “practically next to each other”. Often times, close stations just make sense. This is one of those times.

      3. I don’t think I would add a station, at least not without more analysis.

        I’ve walked from the boardwalks along 520 to the UW campus a few times. The distance really isn’t that far, if you can make it a more direct walking route than exists today. It seems to me that it may be cheaper and just as effective to get a 520 access point to the UW station. Even if you have to do a few blocks of small bore pedestrian tunnel and a moving sidewalk, it may be cheaper than a full new station.

      4. I agree with that strategy as a cost effective one, Glenn. Further, with the last 520 segment unfunded, accommodating something like this into the design could be included in a final funding package as a pedestrian improvement. The big challenge is the Montlake Cut bridge crossing, but I’m sure some creative minds could envision different options to solve that.

      5. Transferring between modes via a nearly half-mile out-of-network walk that involves a drawbridge is a nonstarter. Full stop.

        There’s no easy solution here, but real solutions should have been explored. This isn’t one.

      6. And by “nonstarter”, I mean nofuckingbody’s going to do it. This is how everyone’s faith-based “network effects” collapse.

        “Who planned this shit? Tomorrow I go back to driving.”

      7. It’s a half mile walk using the current tangle. Like so much of the rest of the Seattle transportation network, it has a real bad case of “you can’t get there from here”.

        However, as a direct straight shot, it is more along the lines of 5 blocks, which while not terribly pleasant to walk when having to make a transfer is along the lines of a number of other walks I have had to do to get between transit routes in Seattle. A small bore tunnel, large enough to allow pedestrian traffic to get from a 520 freeway stop directly to the UW station as I suggested, could probably be negotiated just as fast as the tangle of staircases from 4th down to the downtown transit tunnel. If you add moving sidewalks, as I suggested, such as found in pedestrian tunnels at certain airports, it would be shorter time wise than walking it only.

        If you look at some of the places that have deep level metro stations, such as Moscow, that depth is used very effectively to use the escalators, mezzanines, and pedestrian tunnels to extend the effective reach of each station. Seattle has the deep stations already. What I am suggesting is only the additions of all the equipment that has been used very effectively in other parts of the world to make those stations effective reach much farther. This particular case seems warranted. For that matter, it seems like it would be warranted for a few other areas scattered through the UW campus, but then that wanders off topic for a 520 transit route.

      8. It is 2200 feet from the Montlake overpass to the station escalator. That is along an arrow-straight path.

        That means an 8-minute walk. Outside.

        There isn’t going to be any “pedestrian tunnel with moving walkways”, because there’s a 30 vertical freaking feet of water between the two.

        Have any of you people ever visited the real world!? My god!

        If a direct transfer were ever to happen, a direct transfer would have needed to be built into the design of the subway itself.

      9. The “solution” such as it is, would be to truncate 520 routes at UW Station.

        Perhaps run the busses to the U district or Northgate until North Link opens.

        Perhaps continue to run busses between the Eastside and Downtown during peak hour.

        Focus efforts on improving bus access between the 520 HOV lanes and UW station. Maybe even get the UW to allow some layover.space and a decent transit center in the parking lot there.

      10. Given the actual station location on the ground, the obvious solution is, as Chris suggested, to truncate 520 buses and have them drop passengers off right at the station. The savings could be re-invested to boost frequency.

        That said, when considering what I would do for visiting downtown after work, assuming the 542 and 545 retain their existing service patterns exactly beyond 2016, I would unquestionably choose to ride the 542 to Link over taking the 545 all the way downtown, regardless of what conventional wisdom says everyone else is going to do. While the 545 may get through Montlake faster, the inevitable stop-and-go traffic along I-5, plus the 15-20 minutes to get from the Stewart St. exit ramp to 5th and Pine, put the Link transfer ahead – even if the 542 has to sit in traffic for 5-10 minutes to get through the Montlake exit ramp because WSDOT can’t be bothered to put in a bus lane. (I walk relatively fast, so I expect the trip from the Montlake/Shelby bus stop to the train platform to be around 5 minutes). On top of that, the 545 is usually standing room only all the way from Overlake to Seattle, whereas the 542 usually has available seats.

      11. The one problem I have with truncating the 520 routes at UW is that the times that I have walked through this area, I got between the UW area and the arboretum area faster by walking than what traffic was moving, and only one of those times was with a bridge opening. Maybe I am just phenomenally unfortunate when it comes to traffic and visiting Seattle?

        With the US Navy having left Sand Point, what is the practical height limit of new non-draw bridges over the ship canal in that area? Obviously there are sailboats and so on but it doesn’t seem to me that there are any truly huge craft through that part of the canal now. If the Coast Guard could decide on a lower limit for a non-draw bridge there a pedestrian bridge might work.

        A second station could work too, but adding an underground station to this line is not going to be cheap at all. There is only so much money to be had, and the $200 million or so it costs to add an underground station could build a lot of other improvements elsewhere. Furthermore, it seems to me that it is important to get this line up and running and adding a station at this point seems like it would run the risk of delaying the opening.

  9. I’m imagining the tunnel station at the U-District for the light-rail options would be similar to the Bloor/Yongue subway station in Toronto: Upper level-Angle Lake/Northgate, lower level-Bellevue/Kirkland (This obviously does not include the mezzanine if there is one). Who agrees?

  10. “The geometric problem is that no destination obviously dominates.”

    This is essentially what SDOT told Mayor McGinn when he asked it to study light-rail over 520 early in his administration. Sounds great in theory until you realize there’s no place to go on the other side of the lake.

    1. Yep, there is no good reason to put light rail over 520 other than “because 520 exists” or “it worked for I-90”

      What we really need are better connections on the east side to the I-90 bridge, a full east side N/S corridor connecting the urban centers and maybe (when it makes sense) a bridge further north to connect actual urban centers.

      1. If we have a full N/S corridor on the east side, and it connects TIBS in the south and either a Lake City Way branch in Bothell, or north Link in Lynnwood, then a 520 route becomes 100% useless, as opposed to it’s current 75% useless.
        Of course that design is probably ST6 or ST7.

      2. @Lor Scara
        That may well be the case… In the mean time though, there is nothing wrong with running more buses over 520 in the express lanes that are already going to be built. Especially if we can give them a reasonably good connection to existing stations or ones to be built by 2023.

        There may still be an argument for a Sand Point – Kirkland bridge at some point. Its also possible that having a full around the lake route in both directions could make a second bridge unnecessary.

      3. Going from Kirkland to Bothell to UW is supposed to be short and quick and acceptable, while going from Redmond to downtown to UW is not?

      4. I agree, Mike. Either way there is a huge geographic penalty, but going south is a smaller one (as anyone who has driven the roads when 520 is closed knows all too well). 520 buses make a lot of sense, and will make a lot of sense for a long time.

  11. None of these options look good. I think we’ll have to support B1a (Kirkland – UW BRT) and C1 (Redmond – UW BRT) as stopgaps, and focus on Kirkland – Issaquah as a light rail corridor. A1 can go out the window because serving downtown Kirkland is the whole point of this line, and bypassing it on 405 is almost as bad as bypassing it on 520.

    However, not having light rail on 520 runs the risk of subarea imbalance if North King and Tacoma want expensive projects, Snohomish and South King want medium projects, and East King only wants cheap projects. That would require a compromise which may be high (in which case East King builds 520 rail anyway because it has money to burn) or low (in which case the East King refuses to pay more than $1 billion for BRT and the other subareas agree, and Seattle loses its Ballard lines). We should think more about this.

    1. It’s almost as if across-the-board taxation rates for wildly disparate transportation needs doesn’t make any sense!

      1. Yeah, definitely. I fear that ST3 will be a major awaking for the entire region. But I’m an optimist, so I see a few possibilities:

        1) Seattle (and liberal voters in general) simply overwhelm the rest of the voters. This could happen, especially if the election takes place during a presidential year.

        2) Sound transit comes up with relatively cheap, important light rail for Seattle (like Ballard to the UW) while providing very high end bus improvements for the rest of the region. This would take a lot of politicking and way more appropriate engineering than Sound Transit has shown before, but I could see this winning. The problem, of course, is that Sound Transit has spent oodles of time suggesting that rail (even slow as molasses streetcars) are the answer to everything. Right now there are thousands of commuters going from places like West Seattle to Factoria who could care less about whether they ride a train or bus; they just want a ride that doesn’t spend half its time in traffic, or a commute that involves two transfers, each of which takes fifteen minutes.

        If I was a betting man, though, I would bet against ST3. Sound Transit has to come up with a very elegant proposal to satisfy the masses. Propose a streetcar to Ballard and everyone (including me) will tell them to forget about it. Propose a big ticket system, and a lot of suburban voters will just say no. They have what they need, for the most part. Light rail will go to Kent, if not Federal Way — does it make sense to spend billions more to send it further? We are about to build light rail to Redmond — what else does an eastsider need? Another line to Issaquah? Issaquah? There are only 30,000 people in the entire city, and they can all ride buses (very fast buses, traveling in their own lanes) to the nearby rail station. Kirkland is only a bit bigger (50,000 people — roughly the same as the Queen Anne neighborhood) and things will get a lot better for the bus commuter really soon (when the bridge is complete). Can Sound Transit engineer a system that adds real value for the suburban commuter (which means solving the “last mile” problem on both ends) or will they continue to offer up “Yugo” and “Lexus” plans? I hope they do the former, but I wouldn’t bet on it.

      2. ST is not pushing light rail like a Disney movie or cocaine. It’s making it available to those who want it. It’s certainly not pushing streetcars: the SLU streetcar was an anomaly and a concession. The public has two primary factions: those who want high-quality transit and those who don’t want their taxes raised much. Light rail is just superior transit, period: every 10 minutes full time, immune to traffic, better able to absorb spikes, more environmentally friendly, better preparation for the future. The no-more-taxes faction doesn’t understand transit, so they think minimum service is fine. ST’s proposals reflect these two viewpoints, and the result depends on which one wins in each corridor. In some corridors like the Eastside, light rail is not absolutely necessary, so compromising down is more acceptable. But making BRT more expensive diminishes the primary advantage of BRT, so it pleases neither faction. That’s why ST didn’t propose it and why it wouldn’t win.

        The issue with Issaquah is not just “30,000 people in Issaquah”. A Kirkland – Issaquah line would anchor the Eastside and give it a transit-focused spine to grow on. If it strengthens intra-Eastside trips, that will diminish cross-lake trips and lessen the 520 problem. So it’s a good goal, although again not absolutely necessary.

      3. So what does it say to you, Mike, that you could build billions and billions of this magical anywhere rail — “every 10 minutes full time, immune to traffic, better able to absorb spikes…” — and barely a few thousand people will show up to use it!!?

        It’s not because suburbanites are inherently transit-hating gas gluttons. The usage estimates are awful because most of this sprawling region employs land usage that is simply, fundamentally not fit for the concentrated rail investments over which you salivate.

        Nothing is less helpful than “planning” for a million new transplants with $40 billion worth of investments that only will only — that can only — ever see paltry ridership. Your rail bias is off the charts, Mike, and 99% of the time these boondoggles would even be useless to you!

      4. Good transit is the one thing that might make it turn around, or at least sprawl more slowly. Trains are centripetal; highways are centrifugal. The reason cities have decentralized into peanut butter is they lost their moorings when the trains were ripped out. Putting trains back gives them a seed to centralize on. You will say it failed to centralize in Fremont CA. But it’s just a chance, not a guarantee. And Bellevue is much closer than Fremont, and more serious about centralizing. And it may take decades because change happens slowly and gradually until it reaches a tipping point.

        Whether that chance is worth $X billion is up to Eastsiders to decide. I’m sure the ST staff don’t have favorable views of all these rail alternatives. Some corridors are more suited for rail than others. But that’s the purpose of these studies: to quantify the advantages and disadvantages of various possbile choices.

      5. The government push toward decentralization and the demise of development patterns oriented around streetcars or regional rail spines occurred concurrently and reinforced one another.

        Don’t oversimplify the history. It wasn’t as simple as “trains disappeared: the world sprawled” and it isn’t as simple as “trains return; the world is cured”.

        Fixing the suburbs will be a complicated and fraught endeavor; nearly every inch must be reoriented from scratch to enable pedestrian connections, human-scaled commercial infrastructure, and legible transit corridors. And most of those corridors will be buses — that’s the scale at which the demand exists; that’s the logical use of thousands of miles of existing asphalt.

        The last thing that will fix anything is a bunch of arbitrary noderail, terribly interconnected and useless for 99% of trips of any sort. That doesn’t inherently reorient anything, as thousands of examples of terrible ex-urban “TOD” have made clear for decades.

        But don’t believe me. Believe ST’s own universally awful estimates, even for most-major-node-serving thing we’re already building!

        Imaging 19th-century-style distant train spindles to magically reorient the 21st-century world is as flawed a worldview as I’ve ever encountered, on par with the lie that cities are regenerated with streetcars too slow to get you anywhere. It’s monumental bullshit; move past it.

      6. It’s the single biggest missing factor. The lack of it makes people think, “We can’t ever restore our cities/suburbs again”. But if that one thing is done, it makes it easier to do the other things incrementally.

      7. >> Light rail is just superior transit, period

        Sorry, that statement is ridiculous. Want an example. Sure? How about we build light rail to West Seattle. Add a stop at the junction, which is probably the one place where it makes sense. But wait, just adding that costs 3 billion dollars. So now what? For the people who live or work close to there, it is great. But what about everyone else? The surface streets are still pretty busy, so do you really want to reroute the various buses to serve that station? Not only is there a transfer penalty, but a penalty in the overall speed of the buses. It takes a lot longer to get to the station then it does to just get on the freeway and go at least to SoDo (because — I know this is shocking — sometime buses travel on grade separated lines too). On the other hand, spend that money on a handful of improvements, and West Seattle residents could have really fast bus service without ever worrying about traffic. Hell, for half the money it costs to build light rail, you could probably deliver faster, more frequent bus service to the entire peninsula.

        Sorry, light rail is not always superior. Light rail is appropriate for many areas of the city, and inappropriate for many others. Unless we spend upwards of fifty billion dollars on a system, we aren’t going to get it in all of the city (and that number probably goes to 100 billion if you start including the suburbs). The key is to spend the money wisely. The key is to figure out where light rail makes sense, and where bus improvements make sense. It is also in figuring out the little things, so that they can work together well.

      8. >> ST is .. certainly not pushing streetcars

        Except that 2 out of the 5 final proposals for Ballard to downtown service include streetcars. By “streetcars” I mean that they function as “streetcars”. They spend a substantial amount of their time in traffic. In the case of Corridor B, it would be no different than RapidRide (except it gets a new bridge). It would be on the surface through the most congested part of the entire thing (Denny and downtown). Corridor E is similar (going along Westlake instead). The initial 8 proposals were even more slanted towards streetcars (over half had light rail lines mixing with traffic).

        But you are right, Sound Transit hasn’t pushed light rail over BRT. It is only that every time they propose BRT, it is really crappy BRT. Sometimes the light rail proposals (like those in the Ballard study) are bad as well, but I have yet to see a BRT proposal that spends sufficient money to really excite people. It is no wonder that knowledgeable people like you conflate grade separation with light rail and BRT with buses stuck in traffic. It doesn’t half to be this way, and we have numerous examples (even in our own city!) to show otherwise.

      9. “Except that 2 out of the 5 final proposals for Ballard to downtown service include streetcars.”

        Seattle paid Sound Transit extra for those streetcar studies. The city was going to do its own streetcar study but then piggybacked on ST’s study because it’s cheaper to have one set of consultants study both simultaneously, and they could look at how two dissimilar lines could complement each other rather than duplicating. Streetcars don’t meet ST’s qualifications for its own lines (“mostly exclusive-lane or grade-separated”).

  12. We currently see more bus ridership in Bellevue and Redmond than Kirkland, and the rail projections are going in a similar direction. The 550 is full all day and I hear the 545 is similar, and RR B is humming along. But the 540 ahd 234/235 have been less than expected, and the 255 is desolate evenings. Why is this? Is it an artefact of imbalanced bus service (less in Kirkland?) or defective projections? Is it the cruel fate of geography (Kirkland in relationship to 520 and 405)? Or is it really true that Kirkland and the northwest Eastside are less interested in transit than the central and northeast Eastside? Are Kirklanders are proportionially richer (all those waterfront condos) and thus less interested in transit? Or is there another factor? Does it mean we should focus more on Bellevue and Redmond (e.g., C2 and C2A) and not worry about Kirkland?

    1. The 234/235 are slow and indirect, wasting a full 5-10 minutes to deviate into South Kirkland P&R alone. No wonder not many people ride them.

    2. The question shouldn’t be “why is transit ridership in Kirkland low?” But “why is transit ridership in parts of Bellevue and Redmond so high?”

      Kirkland has the same factors working against transit use as in most affluent suburbs. The only real ridership base is Seattle commuters.

      Bellevue and Redmond have areas of relatively high residential and employment density, regional destinations, and a larger transit dependent population.

      While you may be able to attract a few more transit riders in Kirkland and Issaquah with better transit service the ridership isn’t going to be spectacular without major changes in land use.

      1. At least in Redmond, transit ridership is largely driven by Microsoft commuters, which is helped by the fact that, unlike Boeing, Microsoft actually seems to care about making transit a viable option to get to work. This includes giving free buses passes to employees, designing the campus to be walkable, and providing an excellent shuttle service for those that don’t want to walk. At least during the summer, there are also a substantial number of Microsoft interns who walk to work and don’t have cars, which helps drive transit ridership in the area, even during evening and weekend hours.

        Kirkland, by contrast, has no employer with remotely the Microsoft level of scale.

  13. okay, so its not like anyone needs/wants Sound Transit to start investing in a different form of transportation, but if one of the goals here is to get folks from Kirkland to the UW stadium station then why not start looking at the lake as an opportunity rather than an obstacle?

    Kirkland’s on the water; the stadium station is right next to the ship canal. even considering the 7kt speed limit west of Webster point a commuter ferry could easily make Kirkland–>Stadium Station runs in 20 minutes dock-to-dock.

    1. IIRC, the King County Ferry District studied several Lake Washington and Lake Union routes a few years ago, with the general conclusion that none of them were practical. I’m not sure I would agree in cases where there was a neighborhood center and/or connection to very frequent transportation immediately adjacent (Kirkland and S Lake Union–maybe–would meet the criteria, perhaps Fremont), but Husky Stadium station is too far from any potential landing place for there to be any sort of reasonable connection to rail. The dock would have to be on the lake side of Husky Stadium–you can’t tie up in the canal itself–and that’s about a 1300′ walk, cutting across the parking lot. That’s uncovered and uphill. (If you could tie up in the canal, it’s quite short, but would require an elevator to get up to the level of Montlake–and it’s doubtful that the Coast Guard would ever approve of a ferry dock in the Cut, even immediately parallel to the north embankment.)

      I’d love to see a Sydney-style network of small boats crisscrossing the lakes if it were practicable, but the Sydney ferry terminal is directly on the edge of the CBD and even closer to a major rail station, on a level path.

      1. I never have seen that county study (do you have a link?) but have heard that they were looking at routes all the way to downtown / south lake union — and the fact that you’d be slowed to 7kts or less from Webster Point was a killer (Webster Pt S Lake Union = about 4 miles / 35 minutes)

        I think there might have been some talk about landing at UW as a destination (not certain exactly where); but I don’t think any consideration was made to making a multi-modal connection.

        as for a dock, I don’t think its nearly as impossible as you think, considering the size of boats which could / would be used; with a bit of dredging & bulwark construction for a dock… — especially at the east end in the area of the old “canoe house”. I work through there fairly regularly and though snug its wide enough for commercial traffic to pass.

Comments are closed.