D-2 Roadway (Sound Transit)

At last week’s Sound Transit Board Meeting (video here), one of the more interesting reports was the staff analysis of the D2 roadway, which runs between I-90’s Rainier Freeway Station and the Downtown Seattle Transit Tunnel. There are also PowerPoint slides.

No matter what, these lanes will be closed to all traffic during track construction later in the decade. But afterwards, Sound Transit has always assumed that the 554 would terminate at either Mercer Island or South Bellevue Station. Metro has five peak-only routes that serve the I-90 corridor: the 212, 214, 215, 216, and 218, which together amount to about 18 trips per hour in the peak, that it might prefer to keep running into downtown. In 2010, these added up to about 1 million rides a year, or about 4,000 per day*. For these routes, there are four main options:

  • Run joint operations on the roadway, as is currently done in the DSTT. This doesn’t necessarily mean buses would still run in the DSTT itself, but will create similar reliability and schedule impacts to both buses and trains. This is the baseline assumption in the ST budget.
  • Run trains only on the roadway, forcing buses to access downtown via Rainier Avenue and S. Dearborn St. This speeds up the trains a bit but makes the buses slower and much less reliable. It saves on capital costs but Metro will pay more to operate buses.
  • Terminate Metro buses at Mercer Island or South Bellevue. This creates at least some transfer penalty for bus riders, but keeps trains fast (carrying the bulk of the riders) and saves Metro about 15,000 service hours annually, or around $1.5m, that could be invested in more service on these or other routes.
  • Squeeze the tracks on one side of the roadway, allowing a one-lane busway for peak-direction trips. ST staffer Ric Ilgenfritz testified that this is likely cheaper in capital expense than the joint operations option. See the illustration, along with some discussion, below the jump.

Busway Schematic (Sound Transit)

FTA requires an (apparently not very binding) “Record of Decision” next month, and Metro and Sound Transit have converged on specifying a rail-only roadway for now, with strong emphasis on further development of the busway option. At the moment, ST needs FHWA and WSDOT approval, as well as more engineering, to be sure that the busway is feasible.  The busway is described as Metro’s “strong preference.”

The chart below captures the time impacts to both bus and rail riders for each alternative.

Sound Transit

There are two important considerations missing from this chart. First, the number of riders on East Link is likely to be an order of magnitude greater**. Secondly, there’s no consideration of reliability. The joint use options introduce error bars for both modes; the surface option is great for rail but horrible for buses.

The “bus intercept” case is an interesting one. As Mayor McGinn noted in the meeting, the quality of the transfer at Mercer Island is very important to this calculation. The current 2030 operational plan is trains coming every 4 minutes from Northgate and splitting into two 8-minute headway lines. (In the peak, the 4 minute headways will go all the way to Lynnwood.) Even with a high quality transfer, the penalty would therefore be from zero to 8 minutes for buses coming in at random from points east. In the outbound direction, however, it’s practical to schedule buses to depart shortly after trains arrive, and to hold departure until they do.

Personally, I think the busway and bus intercept options are the most acceptable. Joint use will make Link interlining operations a nightmare, and the surface option is so horrific that smart riders will probably transfer to rail anyway, particularly in the afternoon. Between the two good options, it really comes down to the feasibility of the busway and as to whether the saved bus hours from intercept could be better deployed elsewhere. Personally, to encourage good land use and car-free lifestyles, I’m all for de-emphasizing peak expresses in favor of other service, so I’d favor the intercept. But reasonable people can disagree.

* Although some of these are contra-peak on the 212, and would not benefit from the busway option. I’m also not counting the 217, which is all contra-peak.

** East Link projects 50,000 a day in 2030, though I don’t know what share of that is on this segment. It’s likely to go down a bit as economic models adjust to current reality.  The Metro routes added up to about 4,000 a day in 2010. So it’s hard to say what the precise ratio is.

99 Replies to “The Future of the D2 Roadway”

  1. Note that the busway, if built, would either extend all the way to Mercer Island, or require a grade-separated ramp at Rainier to separate bus and train traffic. Such a ramp would be difficult to accommodate in the current design, so I am assuming the planners are envisioning a bus-dedicated lane across Lake Washington. This could have a significant impact on any potential HOV lanes on that bridge.

    1. I did not see any indication of building a busway. Just using the existing D2 roadway.

      If you look at the sound transit plans in the EIS for east link it shows the Joint operation option having buses acessing the roadway from surface level (not the DSTT)using the existing ramp and than merging with trains. The joint operations would continue till just before the tunnel, when the buses would merge with the new HOV lanes. Buses would then just travel in the new HOV lanes as normal.

      I also noticed in the plans that the express lane exit/entrance ramps on Mercer Island will be still in use. Instead of serving the express lanes they are simply re routed to the new HOV lanes and become HOV exit/entrance ramps.

      Reference: http://projects.soundtransit.org/Documents/pdf/projects/eastlink/EIS_2011/15_AppG_ConceptualDesign_Segment_A.pdf

      1. I’ll concur with that. 1 ramp on mercer island will be closed/removed.

        I worked on East Link Seg A

      2. So the buses and trains would not be sharing a 55 mph freeway lane, just the slower on-ramps? I’m not sure I want buses and trains running together at full speed.

      3. Speeds on D2 will be adjusted to accomodate trains and busses. Technically speaking, the road is designed to accomodate busses at certain range of speeds due to curvature. The introduction of trains will require regrading the road to accomodate more precise train speed which will cause the bus operating speeds to be adjusted.

      4. One direction or the other the operation you describe as “when the buses would merge with the new HOV lanes” will require the buses to cross the tracks. The HOV lanes will be the innermost lanes in the new road configuration; that works great for inbound if the tracks are south of the single lane or for outbound if the tracks are to the north. But the opposite direction will require the buses to level cross the tracks.

        There will be collisions.

  2. I dont see how the Peak only direction bus lane can work without either an expensive bus flyover ramp, or a delay inducing bus/train grade crossing. Did they indicate how this would be delt with?

    Personaly i like the Bus Intercept option. Yes it adds a few min to travel time but it will reduce traffic (buses) on the bridge and save bus hours. The trains are going to be running. Why not make sure they are full. If the trains get overcrowded (not likely to happen before 2030) then we can think about changing the configuration. But as i see it a 4 car link train can carry 800 people, at 8 min headways that is 6000 people/hour capacity (each direction). Im guesing that is well more than the current buses.

  3. Tell us please ST and KC Metro, in what city in the world does Diseasel Express Bus service run directly parallel to Electric Rail Transit?

    Especially since the Express Bus will increasingly need to stop at the same stops that the rail line will use so as to offer connections to destinations outside the central city.

    And where will you be finding the fossil fuels to power and lubricate all this parallel service?

    1. Sorry, what?

      This happens in *every* city with express buses. It happens in Boston and NYC, two cities from which Seattle could learn a lot.

      The reason express buses exist is that, during peak hours, there are a lot of commuters who want to get to their downtown jobs. And they’re willing to pay extra for the privilege of not having to transfer. (The fact that Metro doesn’t charge a premium for express buses is a travesty.)

      I’m in 100% agreement that running regular, all-day buses in parallel with rail is a total waste. (That’s why I think all the Eastside buses should terminate at Husky Stadium, and why I’m very happy that the 554 will terminate at Mercer Island or South Bellevue P&R.) But if commuters want to pay a double fare so they ca avoid a transfer, why not let them?

      1. The 554 is an express. I think it should keep running downtown. That preserves the existing 3-seat ride for people using an express to get between transit hubs, and taking a local to/from the hubs at each end.

        Bottom line is that I think you should be able to get from anywhere to anywhere in-county with a 3 seat ride.

      2. But if the 554 runs all the way downtown, it will, by necessity, run less frequent than it would if it stopped at Mercer Island. I’d much rather pay the transfer penalty and have a bus every 15 minutes, than have a direct bus every 30 minutes. There are also easy ways to mitigate the transfer penalty. For example, on eastbound trips, you can just have a timed bus connection with every other arriving train. The bus sits there at Mercer Island until the train arrives and when everyone coming off the train has boarded the bus, the bus moves. The time cost of that transfer penalty is a matter of a minute or two at most.

        Also, remember, that not everybody is going downtown and the 554 becomes more useful for Issaquah->Eastgate trips if it runs more frequently. 554->Link can also be used for Issaquah->Bellevue trips, which would be far faster than any existing option outside of the peak-only 556.

      3. I don’t doubt that the transfer penalty can be easily mitigated, and that there’d be minimal to no time loss for current riders. Buses waiting at the station for every 2nd or 4th train totally takes care of it.

        I’m strictly concerned with the psychological impact of an additional transfer.

    2. Diseasel

      Please. That’s so 20 years ago. After an eternity of sitting on their hands, the EPA has woken up and ratcheted up regulations on diesel engines quite a bit since the 90’s. A road-going diesel that meets EPA requirements for 2010 is actually cleaner than a gas engine from previous years.

      The best thing we can do for air quality is get our 90’s era Cummins M11 powered buses (the Gilligs and high floor artics) replaced with modern vehicles, or at least moved off stop-and-go routes to freeway routes.

  4. 50,000 on E-Link and 4,000 current bus riders. That’s a big difference, but that’s the beauty of projections – there’s no right answer.
    When I worked in direct sales, the key was to get your foot in the door by saying whatever it took. Once inside, you can walk things back to make the pitch, and by the time the product arrives, nobody remembers what was said.
    It really works.

    1. These numbers aren’t made up; they’re based on vetted models. If you have a problem with the model, say what it is; don’t imply they’re intellectually dishonest.

  5. Be careful what you tell yourself you can’t do. Repeat it long enough, you’ll start to believe it ’til it’s hopelessly true.

    Current reliability problems with joint rail and bus operations in the Downtown Seattle Tunnel are nowhere near proof that the idea can never work anywhere in the system. Just proof that the communications, control, and training measures designed into the system and left to gather dust for 21 years now absolutely have to be made to work.

    For starters, all the effort and energy developing the other plans, none of which are very good, can be redirected toward the joint operations we already have, and planning positively for their extension. Let’s do something the easy way for once.

    Mark Dublin Charter member, Employee Advisory Committee on the Downtown Seattle Transit Project 1983-1990

    1. Agreed, however the D2 would be slightly more reliable than the tunnel is presently, mainly because their are no stops. However, if a bus breaks down….

      Regardless, from a security point of view, even though there would be gates on both ends to prevent unintentional (or intentional) private vehicle access to the D2, the best deterrent is direct fixation track (as opposed to embedded to allow busses to use the facility).

      1. Would it be horrible if Eastside express buses skipped the stop at Rainier Ave station? I know there are often folks that board and alight there on the 554 currently, but Link could be an option for them.

        Without that stop, would it be easier to get both buses and HOVs to join the busway further down the road and use the 5th and Dearborn ramp in both directions?

        BTW, I couldn’t open the .ppt that was linked to, perhaps because my PowerPoint is too old.

      2. 1 stop won’t create the kind of delays we see in the DSTT.

        You can’t draw a legitimate comparison between Westlake or University Street station and the Rainier Ave station. The conditions are just so different – less traffic and fewer boardings. So long as everything passing through this corridor has to funnel into the downtown bottleneck, I doubt we’ll ever bump up against capacity issues like we do in the DSTT, so any delays that do occur would be much less severe.

        In the DSTT, we’re running so close to capacity that any delays at all tend to result in instant gridlock.

      3. Lack Thereof, I’m not sure if you were replying to my post or Adam’s, but the reason for my suggestion was to simplify the engineering (and thus reduce the cost) of keeping some bus service between MI and downtown. To serve Rainier Ave. station with LRT and buses, how wide would it have to be? Maybe it could be reduced with side platforms for LRT that are shared by buses with left side doors, but why bother?

        I don’t expect Rainier Ave. will have direct access HOV ramps, so why not take advantage of the already existing HOV ramps at Dearborn?

      4. I was replying to Adam. In reply to you, though, skipping Rainier wouldn’t be too bad of a deal to keep a couple of frequent Eastside buses connected to downtown, especially if it’s just the 554. It would add a bit of travel time to people headed down the 7’s route, but they could still transfer downtown.

        I haven’t seen any plan for the bus stop at Rainier, though, and the diagram shows the buses splitting off into their own ROW a good quarter mile west of the station.

        I can open the .ppt with the current version of Openoffice, but the graph on the last slide is broken for me (lol @ “bus lain option”)

    2. Mark,

      I’m willing to entertain the idea that delay-free joint operations are possible, although safety regs dictate that the trains would have to run slower regardless.

      However, I see no reason to believe that the agencies are capable of pulling this off. Until I see it happen, I’m going to assume that joint ops are going to result in reduced reliability.

      1. safety regs dictate that the trains would have to run slower

        How much slower are we talking about? It’s only a couple miles, so it wouldn’t have a huge impact on travel times, but still I’d like to find out.

        The western portion is pretty twisty, so they might be slowing down for the corners anyway, but the stretch across the bridge could be full-speed.

      2. Martin,

        Agencies, and regulations, are among the tools with which very large numbers of people accomplish their common purposes. Like any other tools, if they’re not working, the people who designed and built them can adjust or replace them.

        Also, as with any other tool, users have to be sure they understand how to make them work. They also have to remember that for any tool to work at all, someone has to actually pick it up and use it- which I think is the major problem with our transit system.

        Just as the Downtown Seattle Transit Tunnel is technically capable of smooth and efficient joint operations, the agencies involved with it are also capable of running the system accordingly. The fact that for a very long time they have not been working up to capacity is up to organized and technically informed voters to fix.

        You personally, and the rest of STB staff, and your general contributorship are already achieving a great deal every day, and in so doing are constantly improving your own ability to start getting things straightened out. Once again: Be careful what you tell yourself you can’t do. Because on this matter, you’re wrong.

        Mark Dublin

    3. I’m willing to believe this.

      As a matter of design, there will be around 50% unused capacity on this segment, due to the converging rail lines at IDS. It makes sense to utilize that extra capacity by sharing the ROW with buses.

      It also seems to me that a disabled bus could be cleared more quickly from this corridor than from the congested DSTT.

  6. I’d be even more aggressive to say that there should be no more buses across I-90 (except perhaps to Mercer Island). Bus service should connect to East Link at points closest to their eastern most connection point. Correspondingly, the number of bus trips into the downtown Seattle core should also be curtailed and instead replaced by local (especially east/west) service that coordinates and augments Link service. The public needs to have their expectations of one seat rides reset.

    As for time penalties, given that bus routes are ROUTINELY late in the I-90 corridor (I know from direct experience), I’m going to bet that a transfer to Link would actually save people time.

    1. While I agree with the need to reset the public’s perception of transferring, note that much of metro’s success and relatively high ridership in the past has been based on 1-seat, express, “flyer type” “blue streak” rides (or whatever they choose to call it based on the decade) Provided link is frequent enough, it would probably work. I agree, take the busses off between MI and downtown, no need to duplicate service. Perhaps metro should raise fares on the express services and see what happens, if ridership drops off significantly, then kill the routes.

      I ride the 90 corridor daily and have only noticed infrequent delays. Then again, I ride in the peak direction with the use of express lanes, the other direction typically does look backed up but most of the time I think it’s because of the HOV lane drop at Island Crest Way.

      1. When there was nothing else, the “Flyer type” was a significant value proposition. With Link Light Rail, bus service needs to take on a different role – that of coordinating and augmenting rail routes. Hundreds of thousands of bus service hours can be saved and redeployed to create better local connections and express services to areas presently underserved. Lots of fossil fuel can be saved in so doing. I don’t think we need to have a competition between rail and buses using fares to see who wins. We need to make the point that the significant investment in rail REQUIRES that our bus services be aligned to feed and augment them.

        The sooner Metro and other systems get it through their organizational noggins the better.

    2. Exactly my view. Just like i think that Sr-520 express buses should terminate at the Husky Stadium link station when U-Link is completed. Saves bus hours, improves peak reliablity/travel times, decongests downtown. I realize there will be a few min a trasfer penalty but it will be much easier to plan for that, then it would be to guess traffic conditions. Personaly im not a fan of 1 seat rides. It seems to lead to too many redundant/underutilized routes. Trasfers work, especialy if the service is frequent. That way if you are delayed and miss a trasfer, the next one will be along shortly (<10 min).

      1. I don’t think 520 and 90 can be compared apples to apples when it comes to truncating express bus service. With 520, its a good 8-10 minute freeway drive to get from montlake to stewart offramp, without much traffic, even longer in the afternoon to go opposite. The trip from Husky Stadium to downtown will take 6 mins (if i recall) on link, making it a natural time savings even if the bus takes a bit extra time getting to husky stadium.

        As for D2 and 90 with LRT and Bus service, the routing doesnt change and save time, its identical so the time penalty for transferring is more significant in the mind of most riders. Also, peak direction bus riders currently have pretty reliable travel times since D2 and express lanes feature very little interacting with private vehicles.

      2. actually, terminating at husky stadium misses an opportunity to serve new markets. Instead of running to downtown, routes could hit SLU, Wallingford, Ballard. Just take a look at the MS connector routes, and the gaps in service become pretty apparent – it would be great to plug them.

      3. Chazz: Yes and no. I’m definitely sympathetic to the idea of linking cross-lake routes with east-west Seattle routes, for exactly the reason you propose. But the converse is that, now, your bus from the U-District to Fremont might be late because of commute traffic from Redmond. And that just sucks.

        Also, no matter what you do, you’ll only be able to avoid a transfer for some people. Obviously, you can’t run a line from Ballard to Kirkland, *and* another line from Ballard to Redmond, *and* another line from Ballard to Bellevue, each with 15-minute frequency. And there are multiple corridors in North Seattle as well. So either you have a complex network of branching/interleaving routes, or you arbitrarily pair up Seattle corridors with Eastside destinations (e.g. 45th to Redmond, 85th to Bellevue, Leary/Pacific to Kirkland), and force everyone else to transfer. In the end, you haven’t really gained much.

        Running buses (or even trains, in some cases) from the suburbs to the first available metro/subway station is a common practice in cities around the world. That seems like a very sensible model for us to follow.

      4. runner,

        You’re right that the savings isn’t as obvious for I-90 as for 520. But when you consider the alternatives, like joint operations on D2, the savings could be significant.

        There’s also the frequency benefit: all the bus hours not spent on redundant service over I-90 can be spent somewhere else, such as on extra runs of the peak buses, extra peak destinations, or better service on trunk lines.

        And there’s also the fact that, while most peak bus riders are heading downtown, not all of them are. For someone whose final destination is somewhere on U-Link or North Link, the only difference is that they’re transferring at a different point.

      5. A “few minutes'” transfer penalty, a “few minutes” Montlake Bridge penalty and a “few minutes” in the off-ramp penalty.

        “A few minutes here, a few minutes there and pretty soon we’re talking real time!” — Everett Dirksen’s ghost.

    3. Bus trips on the I-90 corridor will be a completely different game once the dedicated HOV lanes open in each direction. That will probably put an end to the reliability issues.

      1. Travel into downtown along I-90 is only as reliable as the exit ramp. The 550 has a bus-only exit ramp that goes directly into the tunnel. But the 554 has to use an exit ramp that is also used by thousands of private cars whenever there’s anything going on at either of the two sports stadiums. Delays at the exit ramp during events can be easily as much as a half hour or more.

        Because of the uncertainty of the exit ramp, a transfer to Link will always be more reliable, even with the HOV lanes. Furthermore, when traffic is really bad a 2+ HOV requirement is no even really sufficient to keep traffic moving, especially during non-commute periods where most of the cars are friends and family members traveling together anyway. You typically need at least a 3+ requirement for it to mean much in terms of reliability.

      2. Travel into downtown along I-90 is only as reliable as the exit ramp

        Assuming the joint-use option for the busway, we could still have the Dearborn HOV ramp as a transit-only ramp.

        Also, 2+ HOV seems to be working OK on I-90 where it exists now from Issaquah to Mercer Island. It’s not as bad as the situation on I-5 (yet). And we’re supposed to be able to up that to 3+ if traffic gets bad, if someone somewhere grows a pair of balls.

  7. Running parrallell service misses a great opportunity to serve multiple markets:

    Terminate the routes, take the service hour savings and double up the frequencies on the eastside portions of these routes, or lay a few new pathways to improve connecting coverage. If you double the frequency, you’ve more or less overcome the transfer penalty – sure you have the wait time to transfer to link, but you’ve reduced the wait time to catch your ‘first mile’ bus at the start of your trip. At the same time, you’ve also improved connections for everyone who wasn’t headed to downtown Seattle (a big market – Downtown Bellevue and Overlake are both very signficant job centers, for example).

    1. The 212 and 218 duplicate the 554. Cut them tomorrow. Literally (the service change is this weekend, including Eastside restructure). The 214 is slightly less duplicative, but not much.

      Change the 216 to run between Issaquah TC and Redmond TC. Or double peak frequencies on whatever route already does that.

      The 215 is the only route that gives me pause. I’d keep routing it to downtown via Rainier and Dearborn.

      I can’t believe we’re making Link less reliable, overengineering the roadway, and delaying East Link’s completion for these routes.

      Although it’s worth noting the 216 is the only one of these routes that currently stops at Mercer Island.

      1. Yeah I think that is a really good point to keep in mind. We’re talking about slowing Link and complicating the design for peak only routes. If they were frequent all day routes it might be different but they aren’t.

      2. +1. The reversible express lanes were a mistake the day they were built. Are we really going to make more capital investments for the sake of improving peak-only routes?

      3. Cutting 212 and 218 will hurt ridership more than help it. I believe those two routes are the highest use routes on the eastside and frequently run at capacity. While adding more frequency to 554 to accomodate 212, 225, 229 would probably work as there is only 1 extra stop (MI) between eastgate and downtown, this is not the case for 218 which runs on average 10 min headways between 7 and 830 running full. Forcing all these riders onto 554, even with more service will end up costing MORE since 554 takes an extra 20 minutes longer each run. This is the primary reason why people at my P&R (Iss Highlands) don’t take 554. For instance, I rode my mile down the hill to the P&R today as usual and i got there a bit earlier than normal(6:22). 554 was waiting and departed around a few minutes later. Myself, and the 40 other people lined up waited till 6:38 to catch the more reliable, quicker, and tunnel route 218. Its a 25 minute ride compared to a 45 minute ride typically and the ridership in downtown issaquah simply isnt there to justify frequent service.

        Is there enough ridership between Issaquah TC and Redmond TC? Doubtful.

        Once again this conversation will dive into the debate between what metro’s goals SHOULD be, efficiency or service for as many as possible…ideally something in between. But to be honest, 212 and 218 do that during peak time. Cutting them would be the urban equivalent of cutting 71/72/73/74 with more runs of 70. Sure it would economize the trip, but it would slow it down during the peak hour.

      4. If the 554 won’t do for a 218 transfer, make it an East Link transfer instead. Between East Link and the 554, we should be able to do this. East Link will support the riders and have the headways.

        And why the heck DOES the 554 take that goofy routing through Issaquah? Is it really faster to do that than to just get back on the highway?

      5. Why would you have the 215 take Ranier/Dearborn into downtown when a transfer at Mercer Island would get people downtown faster?

        The only justification of it that I can see is, if the bus is going to be deadheading into the downtown base anyway, you may as well have it in service. But even then, you still don’t need to pay the service hours to have the bus go from one end of downtown to the other and back.

    2. Totally agree with Chaz. According to Metro timetables the Mercer Island portion of downtown routes is only 1/3 of the total service hours. Therefore you could even triple frequencies in Mercer Island (or alternatively provide more late night service) without increasing service hours.

      Although not the main advantage of building link a residual benefit of link is the ability shorten up low density commuter or local routes. These saved bus hours can then be used to bolster high density routes in denser parts of the city. Thus, while allowing low density transit users (especially those dependent on it) to still have reasonable service, it becomes much easier to be car free in dense neighborhoods. Once link is built there is little reason to provide one seat rides from suburban neighborhoods to downtown.

      1. Correction: The Mercer island portion is less then half of the total trip, but the rest of the points stands.

      2. Just keep in mind that peak only routes almost always have a dead head, so you have to add that time to the number of service hours.

  8. The reversible lane looks like the best way to continue offering good service to areas without light rail. Commuters do not want to transfer. Look at the rebellion that arose among Community Transit commuters when they were presented with an option that would have broke thier commute into two pieces with a transfer at Lynnwood. I would expect Eastsiders will be just as vocal in favor of thier one-seat rides. It doesn’t look like the reversible lane has much impact on rail, so that would seem to make the most sense for maximizing ridership until the LRT network branches out to reach more eastside origins and destinations.

    1. The reversible lane isn’t necessary. There’s plenty of spare capacity for joint use, because downtown will continue to be the capacity limiter for the foreseeable future.

    2. Commuters have had their way for far too long, and it robs service hours from suburban local bus routes that could make it possible to get to the supermarket on a bus.

      The first proposal for the DSTT was to have trolleybuses running in it (and later rail), with large depots at both ends for long-distance routes to terminate — like Denver. But suburbanites compained that they really didn’t want to transfer, and so Metro had to order dual-mode buses that could run on the freeway without polluting the tunnel.

      Community Transit has much less local service than Metro, and there are fewer walkable destinations to go to up there, so it’s not surprising that they are even more insistent on commuter routes. I actually have sympathy with them, and think that their commuter routes should remain (but they should be consolidated into fewer routes with more transfers at Lynnwood TC and Mtlk Terrace TC). But King County has more local buses and more job centers outside Seattle, so it’s easier to find a job in your own suburb and take the bus to it. So the demand for commuter routes is not as loud or universal as it is in Snohomish County, which is why Metro will be able to eliminate the 225/227/229 after tomorrow without more than a handful of neighbors calling for its head.

      I saw one such neighbor at Metro’s Kirkland hearing this summer; he said he had bought his house decades ago partly on the basis of the 22x stopping there. I really wondered whether he actually rode it regularly, or whether he rode it once a year or less, and just wanted it there in case he worked downtown someday. I also wondered why the lack of an all-day bus didn’t bother him. What good is a peak-express if you don’t work downtown, but you do go to the supermarket, and your kids (this being a single-family area) want to get out of the house.

    3. I suspect a big reason for Snohonomish County residents screaming so loud about the prospect of having to transfer is that the headways of their local buses are awful and if a 511 arrives in Lynnwood 10 minutes late (entirely possible with traffic), you are left with an hour long late for your connection! Thus, in order to ensure you make your connection, you have to leave downtown earlier and aim to arrive at Lynnwood a good 15-20 minutes before your connecting bus. In short, the worst-case travel time to get home from downtown becomes the everyday travel time to home from downtown.

      We can do better than this without operating redundant routes to avoid transferring. A train with a dedicated right-of-way will be more reliable than a bus that has to contend with freeway traffic. We can offer timed connections so if you are in the downtown tunnel at, say 5:07, you are guaranteed to have a connecting bus at Mercer Island waiting for you when you arrive. Finally, we can use the service hours saved to run connecting buses every 15 minutes or so, rather than once an hour.

  9. Martin,

    In the first paragraphs, you may have neglected to include routes 225 and 229; this week, their trips will be replaced by added trips on Route 212. the combined average 2011 ridership on routes 212, 214, 215, 216, 218, 225, and 229 was 5,200 per day and 1.3 million annually (larger than 4K and 1 million annually of your text).

    Today, other routes use the D2 that were assumed to be restructured before 2020; they include routes 111, 114, 210, and 211, in addition to the 554.

    for comparison, consider the 2011 weekday load and annual ridership on Route 550, that will be elevated to Link: the sum of loads was 5,900 and annual rides were 1.7 million. so, the current loads between the Link market and the east I-90 market is much closer.

    as other posters have commented, the first bullet re joint use is over stated, as there would be no stops; it would be more like the Pittsburgh busway than the DSTT.

    truncation at Mercer Island could work well but will require investment by ST and cooperation from MI.

    1. The trasfer at Mercer island will be super easy if you look at the station design. Westbound buses will exit at 80th ave HOV exit. Turn left and drop passengers at a stop on the west side of the overpass. From there it will be a single escalator ride down back to freeway level and you are on the link platform. The same bus could then just turn left again and enter the freeway eastbound from the 80th ave HOV onramp. It would be extreemly quick.

  10. To me, the only options that make sense are
    a) shared operations
    b) truncate the I-90 routes

    I’ve got no problem truncating the various Metro local buses, but I instinctively don’t like truncating the I-90 expresses. You have to assume that an express rider is already making a 3-seat trip – a local to their home transit hub, an express to a distant transit hub, and finally a local to their destination. Adding a transfer in the middle of their express leg bumps it up to a 4 seat ride, which starts to seem ridiculous. It’s not “just one transfer”, it’s “one more transfer” to a rider who’s probably already making 2 of them. And with the transfer point at Mercer Island being so painfully close to downtown, it adds insult to injury.

    On the other hand, we have street capacity issues on the downtown bus routes right now, which are only going to get worse once the RFA is ended and the DSTT goes rail-only (to say nothing of the SR99 tunnel). We might find ourselves forced to get quite a few buses out of downtown, in which case truncating I-90 routes would be low hanging fruit.

    Truncate them at South Bellevue station rather than Mercer Island to get rid of the psychological “but I’m almost there!” aspect, maybe. Yes, I know it takes just as long to get to S. Bellevue as it does to Mercer Island, but it’d be better for Eastside mobility (quicker connection to DT Bellevue), and it wouldn’t be such a slap in the face to Seattle commuters.

    1. “You have to assume that an express rider is already making a 3-seat trip – a local to their home transit hub, an express to a distant transit hub, and finally a local to their destination.”

      I doubt it. If they were taking a local bus from home, they could just as easily transfer to one of the main trunk routes (550 or 554) instead — or they could if the local route stopped at a transfer point (114 vs 240). Some streets with peak-expresses have no local routes at all, so they don’t have that option. They — and others without any bus near their house — drive to a P&R. You can call that a “third seat” if you want to, but it’s a shorter travel time than a “true” 3-seat ride.

      1. Oh, I’m with you on that. I only mean true hub-to-hub expresses should be preserved, like the 554 or 550. I’m totally in support of truncating Metro’s peak-only “express” buses, that do a local suburban route with an express tail to downtown. I didn’t mean to lump routes like the 114 in with ST Express service.

        And I thought about the P&R users, but they’re a minority of riders. We can consider that a 3rd seat if you want, but in the end a full P&R lot doesn’t fill that many buses compared to walkers or local transfers.

      2. Not including west of the lake, 554 stops at only 1 non P&R route.

        And are P&R riders really a minority? At last check about 27,000 on average used a P&R and if we are talking suburban routes (ie areas that have park and rides) thats probably pretty significant.

        P&Rs supply the majority of riders on 212, 214, 215, 216, 218 and a multitude of other 100, 200, and 300 series metro routes as well as nearly all the other ST routes.

      3. Your link actually says 18,000. I’m not saying that P&Riders are insignificant, just not the majority. In my mind, Park and Rides should be only on the edges of the system, where there really isn’t practical local service to get to the hub. In the end, they serve the same purpose as local residential routes feeding to the trunk lines, and I’m willing to consider those riders the same way as I would transfers from a local bus.

        Regarding the 554 stopping at lots of P&Rs, that’s true, but those aren’t one-route stops – there’s multiple local transfers at every one. Every local bus in Issaquah stops at the Issaquah P&R. Too many buses to count swing through Eastgate. Every bus on Mercer Island serves that P&R. P&Rs are, as a rule, situated at transit hubs, which is terrible land use, but the only way to make it convenient for users.

        Eastgate P&R worries me, because the area is growing, and those single-family neighborhoods to the southeast are going to densify some day. Eventually, it’s going to be dense enough to get all-day local bus service, and then the P&R should be removed and redeveloped, but it will be damn near impossible to do it without residents raising a fuss.

      4. Actually, the inbound 554 stops at 2 non-P&R stops:

        • 9th Ave N.E. at N.E. Ellis Dr.
        • W. Sunset Wy. at First Ave. N.E.

        And that’s not counting the stops in Sammamish and Redmond for the extended runs.

      5. Both issaquah P&R’s don’t exactly have local service…highlands service is entirely peak only with the exception of 554, which i wouldn’t call local, it only has 1 non-p&r stop on the eastside.

        issaquah transit center provides transfers to 271 (sometimes), 200, and 927 which I guess you could call local though the frequency and destination mix is poor.

        I don’t follow your logic on why eastgate will need to be redeveloped…
        firstly, there already is somewhat local all day service with 246, 245, and 271. secondly, the neighborhoods that will be redeveloped are mostly on the north side, the south side has expensive view property and large homes primarily that are newer than the lake hills neighborhood. also much of the land on the south is county land, whether bellevue annexes it remains to be seen http://en.wikipedia.org/wiki/File:Bellevue-wa-map.gif. thirdly, it will serve as it does now, a gateway for bus, and hopefully light rail service. maybe i misunderstood your comment.

      6. the ellis street stop was to accomodate sweedish hospital patrons/employees/riders. while it may technically be a local stop, you can SEE the P&R from the stop (its about 2 blocks) so its more to shorten the walk for folks going to the hospital. The first street and sunset stop is a true non-p&r stop in downtown issaquah.

      7. blegh, I don’t like the extended runs. The 269 should just get better service instead. Having the extended 554 is just caving to peoples desire for 1 seat from everywhere to everywhere.

      8. runner: all those other areas around Eastgate ripe for redevelopment already have some level of local all-day bus service. What I was trying to say was that the developments to the SE are the only ones still dependent upon the P&R for transit access. Once they get transit, we can (theoretically) shutter the P&R – but to get transit, they’ll have to get denser, which is inevitable given a long enough time frame (at least in this area).

      9. The extended runs of the 554 would be deadheading back to East Base if they weren’t in revenue service. They’re cheep.

      10. “And I thought about the P&R users, but they’re a minority of riders.”

        Not on the 550. The bus leaves downtown with standing room only, and more than half the bus often gets off at Mercer Island or South Bellevue.

    2. Despite the name, not all of the ST routes are peak-only expresses. Most of them — and certainly the ones you’re talking about keeping — are frequent, arterial routes, which simply happen to serve destinations along the freeway.

      The difference between the 550 and the 212 is huge. The difference between the 550 and the 71 is the color of the bus.

      1. I see exactly what you’re saying. I don’t want to divide buses up based on who operates them, or what name the agency has decided to stick on the route. I had a definition in my head that never made it into text.

        If its purpose is to connect transit hubs and transfer points, I don’t want it truncated.

        If its purpose is to reach origins/destinations, go ahead and truncate it at a sensible transfer point.

        The only things I’m concerned about preserving are 3-seat local->express->local rides between different parts of the service area. I don’t want to see those bumped to 4 seat rides. 2 seats just to get from one hub to another (Issaquah TC to Downtown) worries me.

    3. Commuters on the 554 from Issaquah are not likely making “3 seat rides.” They are most often parking at either the Highlands P & R or the Issaquah TC. Unless you’re coming from North Bend or the like, your probably driving to catch this bus.

    4. I won’t object to keeping the 554 running all the way to downtown in order to preserve its hub-to-hub status. The point is to get people to use the trunk routes when they’re travelling across the county, rather than expecting custom point-to-point routes. The 554 is the trunk route between Seattle and far-east-central King, as the 550 is for Seattle-Bellevue, the 545 for Seattle-Redmond, and 255 for Seattle-Kirkland.

      If the 212 is necessary to relieve overcrowding on the 554, that’s fine. I’d rather renumber them to 554 and arrange some compensation money for ST to run them. The only difference in their routing seems to be SE 36th St and Richards Road. The current 240 could take these ppl to SBP&R; I’m not sure if the new 240 will be too far away.

      Another option is to split the 554 into a peak-only route to Seattle and an off-peak shuttle to SBP&R. That would get commuters to their jobs while avoiding redundant service in the slack hours. (I suggested the same for a 101/102 consolidation.)

  11. I’m glad ST is planning to truncate the 554. That makes my day.

    Adding asphalt lanes for a few marginal peak-expresses sounds like a large expense for little benefit. It’d be different if they were all-day routes to suburban downtowns, rather than peak-only routes to single-family neighborhoods.

    The whole concept of peak-residential routes is grossly unfair. Why should one neighborhood have an express bus to downtown when the neighborhood next door has no bus at all? There’s no difference between the neighborhoods, so why is one privileged? I never understood that about Kingsgate or Timberline or parts of Bellevue. Or why the old 340 (which ran on 405) stopped at 70th, 132nd, and 160th, but not 85th or 124th, which half the time I was going to.

    1. 340 probably didnt stop at 124th and 85th because they had much more GP traffic, no close residential areas (comparitively) and no park and ride whereas 70th, 132nd and 160th do.

      Some neighborhoods may get a bus route due to development agreements. I think the last new P&R built was issaquah highlands (or tukwila, but that’s different) and that park and ride was part of the developer agreement.

      I find it interesting on this board how many posters lament any eastside route at all. First it was the “empty” milkruns that meandered seemingly aimlessly from one pocket to another which i get, streamline those. But I’m not sure why folks are now bashing the commuter routes that are frequently full and make minimal stops off freeway. These typically are some of the more efficient routes in the system. Ill agree that the 90 routes should be truncated at MI and i agree with cutting many underperforming routes that are being cut this fall (ie my old route 266, sniff)but please leave our eastside express routes out of it! :)

      1. I’m not against Eastside routes. I’d like to see 30 minute frequency full time on the routes in Bellevue and Kirkland. If it’s good enough for 3rd Ave NW, it’s good enough for 148th Ave NE. One of the reasons people cling to their cars and parking spaces and freeway lanes on the Eastside is because the bus service is so spotty. That builds entrenched trip patterns over the years. I’m very happy about this RapidRide B and the Eastside reorganization. It means more frequent service from Bellevue to Kirkland most of the time, which should be a no-brainer. The Eastgate-Overlake bus could do with some more runs, but it’s a step in the right direction. I grew up on the east end of the new 226 (Northup Way — current 230) and new 249 (north Bellevue Way — also current 230), so my high school routes will be turning into even slower, less frequent milk runs. Still, I’m glad about the change overall because it’s a chance to establish future frequent service corridors in the areas that have grown the most.

  12. So the question is: in what way should we have half-empty buses follow a half-empty train over the bridge? The answer is to leave the buses on the east side, of course. No bonus points to those that found the correct answer – this one was too easy.

    1. In fairness, the 212 is hardly a half-empty bus — from what I’ve heard, many peak trips are pretty much full.

      1. But it will be following a train that will have the capacity for everyone in that bus, and more. Those riders belong on the train, and that bus and driver belong driving on roads where the train can’t go.

      2. Riders of the 212 do not belong on the train unless the train had a station at the Eastgate Park and Ride.

      3. Have it drop off at the S. Bellevue Station. It’s right off the freeway. Then it can turn around and run it’s loop again. And again. And probably again, considering it’s saving a trip all the way to Seattle. The frequency you’d gain will more than make up the transfer time.

        If this little jog off the freeway is really unacceptable, then end at Mercer Island.

      4. If the 212 was truncated at South Bellevue or Mercer Island, the riders of the 212 would decide to drive to South Belleuve or Mercer Island instead of parking at Eastgate for a two seat ride. That would cause underutilization of some park and rides and overutlization of others. The transfer penalty is just too great for peak only commuter routes.

      5. “If the 212 was truncated at South Bellevue or Mercer Island, the riders of the 212 would decide to drive to South Belleuve or Mercer Island instead of parking at Eastgate for a two seat ride”

        Only if those parking facilities had unlimited capacity. Mercer Island is already full enough as it is. Limited parking will inevitably necessitate accepting a bus transfer to get to the train. The trick here is to make the bus transfer as painless as possible.

      6. The 212 runs at nearly 5-minute headways at rush hour and is usually standing room only. I can’t see making those passengers take a 5 minute bus ride to Mercer Island and then transfer to the train. I’m sure most would either use a different P&R or abandon transit all together. My guess is that most people would probably prefer the delay of a 212 rerouted via Rainier and Dearborn than a transfer at Mercer Island.

      7. Crazy drunken late-night idea: What if we truncate all these insanely popular peak-only routes at convenient transfer points, but then instead of using the extra service hours to increase frequency during peak, make the new, shorter route all-day at low frequency off-peak?

        Would that be a bigger benefit to the communities?

      8. Ok, crazy idea time: Gondola to the S. Bellevue station. Total trip length: 5.5 minutes. Frequency: every 30 seconds. Number of operators required at any given time: 2. Capacity: far more than you’d need.

        Basically, we need a shuttle to light rail that minimizes inconvenience for the commuter. Something this quick and frequent that can be tied right into the Link station sounds perfect to me.

      9. I suggested above: split the 554 into a peak-only route to Seattle and an all-day shuttle to SBP&R. Consolidate the 212 into the peak 554 if possible.

        That would leave out riders on SE 38th St and Richards Road, it looks like. We’d have to provide compensation appropriate to their ridership. That could mean taking the 240 to a transfer point. Although with the 240’s rerouting I’m not sure if it will run close enough to them; it may require another route.

    1. that and to mitigate the “loss” of the carpool lanes / aka express lanes. however if you are an MI resident you can use the express lanes as a SOV.

      the carpool lanes are also useful to encourage vanpools etc. if 90 is not tolled they may turn into HOT lanes.

      1. politics, if memory serves

        the gist was if MI was going to go from a 4 lane highway to a 8 lane major interstate, they were going to get lids and golden tickets out of it.

      2. One reason I’ve heard (around here, IIRC) is that the express lanes would be underutilized if they depended only on HOVs from the single entrance east of MI plus HOVs from MI. Add SOVs from MI and it gets up to almost decent usage. Of course, back when it was built, the express lanes made a little more sense. Now, due to the growth of the Eastside, the traffic backups occur on the so-called reverse-peak direction.

  13. Just terminate the buses at South Bellevue. It will create some transfer penalty, but it will allow for much more bus service and will allow the same route to provide service to downtown Seattle and downtown Bellevue.

    And besides, with frequencies as high as they will be, transferring is easy.

  14. An interesting point is that East Link won’t open for ten years, and by that time trip patterns may have changed, especially if gas goes above $4/gallon and stays there. So maybe it’ll be easier to truncate those routes then than it is now.

Comments are closed.