Near the Ash Way P&R on I-5 (image: SounderBruce)

The draft ST3 plan in March 2016 extended rail beyond Lynnwood in two steps. The first, in 2036, would bring service to North Lynnwood, serving stations at West Alderwood Mall, Ash Way, and Mariner. The second, in 2041, extended around the SW Everett Industrial Center (Paine Field) and north to Everett Station.

When the plan was finalized two months later, the extensions were combined so the Paine Field and Everett stations would open five years earlier. It was a telling decision that all the extra financial resources of the final plan were put into the northern segment. This looks like an error. While all parts of Everett Link have their value, the immediate rider needs are mostly between Lynnwood and Mariner.

Rearranging the Snohomish subarea resources could still open those stations by about 2030. The trade-off is that accelerating some capital spending generally means delays elsewhere. This may mean a later opening of service to Paine Field and Everett where the need for light rail is less urgent.

Famously, Snohomish County has bad traffic, the worst in the nation by some measures. A significant part of this stems from the booming bedroom communities from which thousands commute daily to Seattle. Almost as many Snohomish residents work in King County (145,000) as in Snohomish (158,000). For those who use transit, Lynnwood Link will deliver faster and more reliable travel times. It could serve these riders even more efficiently with more stations a little further north to intercept buses from across the County.

The Ash Way and Mariner areas are fast growing communities with healthy concentrations of local riders commuting to King County. Link would also connect to Swift BRT service at Mariner.

Snohomish County leaders do not, however, wax lyrical about one of the nation’s large bedroom communities. They envision Paine Field as the urbanized hub of future employment growth in the county with commuters arriving by rail from Everett and Seattle. Such a long-term vision will develop over decades. It is less pressing than responding to the needs of the commuters already stuck on I-5.

An earlier Link extension to Mariner wouldn’t affect the capital cost of the ST3 program much; it’s the same line constructed in two steps rather than one. It may affect the delicate debt management problem Sound Transit faces after 2030. And it would add some operations cost if the trains started service earlier to North Lynnwood. Let’s take these in turn.

Sound Transit debt is limited to 1.5% of the assessed value of property in the RTA (unless 60% of voters approve more debt). The agency adds debt as the ST3 program is built out, and then the debt burden declines as projects are completed. The ‘pinch year’, when the debt load is closest to the maximum allowed, is currently projected as 2032.

Projections that far into the future are uncertain and sensitive to small changes in assumptions. Sound Transit may comfortably stay under the limit if interest rates remain low and costs are well-managed. With a recession, Sound Transit could hit the limit and needing to delay planned projects. It’s important to acknowledge how debt dynamics figure large in Snohomish County leaders’ thinking about ST3. The debt limit is projected to constrain most just as construction along the Everett alignment is scheduled to ramp up.

Prudently, an early opening of the line to Mariner means a balanced slowing of outlays on Mariner-Everett rail so cash flows are balanced until the “pinch year” is passed. That may mean construction beyond Mariner would not commence until after 2032 when the spare debt capacity expands again. But it would be worth it to have earlier service to the busiest stops north of Lynnwood.

An accelerated opening means operating costs start earlier. But they’re not so large. The Sound Transit Financial plan projects that Link to Everett will add $77 million of operations expense beginning in 2036. It is 6.8 miles from Lynnwood TC to Mariner P&R. That’s 42% of the 16.3 miles to Everett. We can ballpark the operations costs of the shorter segment by multiplying these numbers. The added operating cost of an early opening to North Lynnwood comes to about $32 million per year (or $21 million per year in current dollars). That’s not very large, and it would be even less with an offsetting delay to the opening of the Mariner-Everett segment.

Money is not the only challenge to an early opening, of course. But it’s probably technically feasible to open about the same time as the more complex alignment to West Seattle, scheduled for 2030. That would be six years earlier than the current plan for Everett Link. Snohomish County and Sound Transit should accelerate planning to make this early opening possible.

Ridership Map
Even with optimistic growth targets for Everett, ridership on Link is projected to tail off precipitously beyond Mariner (image original: Sound Transit)

37 Replies to “Could Link to Mariner open early?”

  1. As a reverse commuter from Seattle to Paine Field who takes transit, I would be ok with this. The link to Swift transfer to get up to Paine Field would still be an improvement over the current condition (512 to Community Transit’s limited 107 route). If I could get this sooner, I would gladly take it over waiting longer to get a one seat ride.

    1. I’m guessing that once Link gets to Lynnwood there will be good connection service from Lynnwood to Paine Field. Otherwise you are basically saying you want to connect two areas via billion dollar rail, but aren’t willing to spend money connecting them via a bus. In general I expect Lynnwood to be a major transit center for all of Snohomish County (or at least everything to the north of it).

      1. I shouldn’t have said “you”, I meant “Snohomish County” (is basically saying …).

      2. Good point! If they improved the frequency of the 107 or had some other bus connection to Lynnwood that would work too. I was a little to focused on the Mariner proposal.

        In a perfect world (for me at least) Sound Transit would truncate their Everett to Seattle routes at Northgate and use some of that money to make the 513 run in both directions and not just in the peak direction.

      3. I agree Curtis. Truncating at Northgate is controversial, but I would definitely do it. I really don’t see the point in running buses on I-5 past Northgate once Northgate Link opens. The time savings are huge with a Northgate truncation, which not only opens up the possibility of far more frequent service, but new runs at no extra cost.

      4. One of my big complaints about the Paine Field diversion is that that area had no existing all-day bus service until Swift opened. Why jump ahead to rail before even testing its viability with a bus? I’m also a reverse commuter who works in the Paine Field area, so I stand to benefit if I’m still working there in (checks notes) 2036.

        The Swift Green line already works well for the segment between Mariner and Paine Field. It’s very reliable and runs frequently all day, so there’s not much of a transfer penalty for using it. The problem is that it doesn’t share a stop with the 512, so my only good option today (there are many bad options) is to take the 512 to 112th, take my bike on the Interurban south to 128th (not a bad ride, but not really walking distance for a transfer), and catch Swift Green that way. If Link is the first thing that adds that reverse commute stop at Mariner, then adding it faster makes my commute much easier.

        The 107 just doesn’t have enough runs to be a useful route. If CT decides to make it all day after Lynnwood Link opens, or even to add a few extra runs, it becomes much more useful, and maybe everything else i said is moot.

      5. I’m also a Paine Field reverse commuter, and although I ride a vanpool most of the time, I occasionally take CT107/ST512 home. I’d consider switching from vanpool to transit if there were an easy Link-Swift connection at Mariner, although I’ll probably give Link/107 a try once Lynnwood opens. I’d bet CT could use the saved service hours to make the 107 an all-day route.

        What I found really weird about the 107 is that a pretty decent proportion of SB ridership doesn’t originate at Seaway TC, at least for the last (5:20) trip of the day. I’m not certain whether or not these people are treating it as a local route (and taking something else on the way back) or are just workers at the strip malls along SR 525.

        @ Jeremy
        Have you considered transferring from the 512 to the 201/202 at Ash Way, riding to Mariner, and then taking Swift from there? Peak frequency on the 201/202 isn’t awful.

      6. “take the 512 to 112th, take my bike on the Interurban south to 128th”

        There’s a trail between them? I may walk that rather than wait for an hourly bus. I’ve never seen Swift Green yet because the connecting buses are so infrequent on weekends. It’s take the 512, wait 25 minutes for a connecting bus, just to get to a Swift stop. Or take the E and Swift Blue to Swift Green, which takes longer but doesn’t have the long wait in automobile hell.

      7. @Pat: That’s a 4 bus commute for me (45 -> 512 -> 201/202 -> Swift). How many people are really going to do that unless they have to? I’ve never had a commute longer than 3 buses before, and even that’s pushing the limits of what I would do. I also usually vanpool, but bike -> 512 -> bike -> Swift is my backup.

        @Mike Orr: Yes, the Interurban Trail does connect the South Everett P&R (112th) to Mariner P&R. The segment is roughly 1 3/4 miles, so not really walking distance, but if you can get a bike on the 512, it’s not bad, though it’s not really well signed. You need to take a right out of the P&R to find the trail, turn right and cloverleaf under 112th St, and take the left fork to cross over I-5 when the trail splits. It pretty much just dumps you out on 128th right at the Swift stop.

        Loading a bike on Swift isn’t bad, either – you take the bike inside the bus, and they have racks inside near the back. It would be prudent for Everett to make sure their scooter share serves these two stations – this seems like a pretty good use of that.

      8. “One of my big complaints about the Paine Field diversion is that that area had no existing all-day bus service until Swift opened. Why jump ahead to rail before even testing its viability with a bus?”

        Current service isn’t a reliable guide to potential. Current service represents CT’s budget limitation and all the other needs in the county. Kent never had 15-minute service to downtown until ten or so years ago; that doesn’t mean nobody wanted to take a bus in between the 30-minute service; it just meant there was no bus to take. Snohomish is way, way underserved. I’ll defer to the locals on whether the Seaway-Canyon Park corridor was the second-highest priority, and whether it makes sense for full-time BRT to terminate at a weekday-daytime industrial center, but at least it’s something in a county that hasn’t had much transit. To see the real mobility demand, look at the cars on the highway. If there are cars, there should be buses.

  2. For many riders, it won’t make that much difference. If I’m riding a bus from Everett, and the bus stops at Mariner instead of Lynnwood, is it really that much faster? If I’m not mistaken, the station at Mariner will be just south of 128th. This means that a bus coming from Everett will need to move from the HOV lane to the right hand exit lane, then wait for the light to cross 128th. A rider will then take a train which will stop twice before Lynnwood. In contrast, a bus that continues to Lynnwood will use the HOV lane and HOV ramp. It will get to the station without waiting for a traffic light. Even if it encounters heavy traffic, it isn’t likely to be as bad as coming to a complete stop (twice) and waiting several seconds (along with whatever delays are involved with getting to Mariner Station).

    From a bus connection standpoint, the only ones that come out ahead are those those whose buses are likely to run on 128th, 164th or Alderwood Mall Parkway anyway. But even then, you don’t come out that much ahead. The stations will be close to the freeway. In the case of 164th, there is an HOV ramp headed southbound anyway. A train would likely beat a bus from there to Lynnwood, but not by much. If the final destination (or transfer point) is Lynnwood, you will likely come out behind. Since 128th doesn’t have HOV ramps, riders there are likely to save a couple minutes. Alderwood seems like it would have a smaller bus catchment area.

    There are some obvious advantages to stopping at Mariner. It means trips between Mariner and Lynnwood are easier. Riders who actually live or work close to one of the stations will benefit as well. Neither groups seems very large. Community transit would also save some money (by truncating buses) but probably not that much.

    In contrast, the stations north of Mariner may be overkill (well, the entire Everett Link is overkill) but at least they offer a huge improvement over an express bus. Stations at SR 99 or Casino Road are a long ways from the freeway, and thus a long ways from Lynnwood or Everett TC. The train is likely to beat an express bus just about any time of day.

    1. A significant advantage of truncating busses at Mariner and Ash Way is being able to serve the surrounding communities with more frequent busses, instead of sending busses on the freeway. Congestion between Alderwood and the interchange with 405 gets serious, and is there any reason not to expect that to extend up to Mariner by the 2030s? This could be alleviated with shoulder lanes, but you’re still burning those bus hours on the freeway.

      1. That would seem to suggest that the Link project split should be one station further at 99/Evergreen and Airport Way. That’s only about one more mile and it’s at the intersection of Swift’s Blue and Green Lines.

      2. Sure, but my point is it really isn’t that much in terms of service hours. The big savings come from truncating in Lynnwood. Lynnwood is ideal because it has HOV lanes heading north which feed directly into the station. Mariner does not. When there is no traffic, you save 6 minutes on the freeway, but spend an extra minute waiting for the light. That is a significant 5 minutes savings, but not exactly huge. When traffic is heavy, it is more complicated. A bus might spend an extra ten or even fifteen minutes in the HOV lane getting to Lynnwood. But when traffic is that heavy, the general purpose lanes are really, really slow. That means a bus might spend just as much time getting over into the exit lane (and exiting) than it would just slogging its way to Lynnwood.

        Meanwhile, if you truncate at 128th (or 164th) you are truncating in the middle of nowhere. There are lots more people who are headed to Lynnwood (as a final destination, or to transfer to other buses). There will be riders who will appreciate having the buses running every 15 minutes instead of every 20, but miss that one seat ride to Lynnwood.

        I’m not saying there isn’t value in extending the line (from a truncation standpoint or otherwise), but I’m saying it just isn’t that much.

      3. @Al — Yeah, that would make more sense. SR 99 and Airport Way is a logical terminus, since it connects with SR 99.

      4. One good use of those bus hours if Lynnwood opens but Mariner doesn’t, is that all busses from the north, east, or west, that end at Mariner, be extended south along the freeway to get to Lynnwood (including Everett busses, but just CT). This gives all of these passengers a direct connection to rapid transit years earlier. Once Mariner does open, these routes could be shortened again.

  3. Given that Tacoma Dome Link is scheduled to open in 2030 and the design step is well underway, I would think that the first studies would need to be underway this year! Luckily, the first segment appears mostly non-controversial and little land acquisition appears required for the southern segment. So, the decision to split the opening year and start early engineering has to be decided pretty soon or 2030 appears unrealistic.

  4. In terms of traffic, there are several things that could be done to speed up the buses. Many of these would have long term benefits, as they would speed up the express buses that are likely to continue to run between Everett and Lynnwood.* These include the following:

    1) Change HOV 2 to HOV 3. This is politically difficult but very cheap. The biggest problem is a lack of leadership, as it is quite possible this would be favored by most residents. It wouldn’t surprise me if more people (although fewer vehicles) ride in buses and 3 (or more) person cars than ride in two person cars. Even if this isn’t the case, given the slow speeds, it seems like folks would have a compelling argument.

    2) Add passing lanes for buses. This is not that expensive compared to most freeway projects. You would retain the HOV 2 lanes, but to the left of them you would add bus-only lanes, but only where it is cheap to build them. This is similar to the shoulder bus driving, but as a more established lane. For example, between 128th and 164th there is a sufficient median and no overpasses, which means extra lanes could be added for that section. Between 128th and the South Everett Park and Ride ramps there is one pedestrian bridge (it is possible you could avoid the pillars). Likewise, it seems like you could avoid the pillars at 128th ( which means that it might be possible to have a bus only lane from South Everett to just short of I-405 without spending a huge amount of money. Unfortunately there are a lot of overpasses up in Everett. To improve that either requires the first option, or spending big money, like the following:

    3) Extend the HOV lanes into Marysville. A lot of the northbound congestion is caused by the fact that the lanes end (and there are a lot of people headed to Marysville). Not only is it very congested getting from Everett to Marysville in the evening, but all those drivers merging into the general purpose lanes causes backups for those headed to Everett. Unfortunately, extending the HOV lanes would require building another bridge, which would be costly.

    *That section of Everett Link is meant to serve Everett, rather than replace express bus service (in my opinion). That makes it different than Lynnwood Link. I really don’t expect express bus service along I-5 between Lynnwood and Seattle once Lynnwood Link opens. But I do expect I-5 express bus service between Everett and Lynnwood (which would serve South Everett Park and Ride) even after Everett Link opens.

  5. Is there any messy engineering planned for the Lynnwood-Mariner stretch? Tunnels, bridges, stuff like that?

    It seems strange to me that if Lynnwood opens in 2024, Mariner wouldn’t open until 2036 for what appears to be a fairly straight ~7 miles stretch. For comparison, Northgate to Lynnwood is about 10 miles, and is opening just 3 years after Northgate.

    1. I think it is all about funding (which is what Dan is getting at). There are some engineering issues (e. g. going over I-405) but getting up to Lynnwood has issues as well (the big cloverleaf at the county border).

    2. Yes, Mariner has to wait until Snohomish generates enough savings to pay the bills. Northgate to Lynnwood was originally two years, and it’s split between two subareas. (North King pays for Northgate-185th; Snohomish pays for 185th-Lynnwood.) Snohomish may have had a down payment saved up as Pierce did, because it was collecting taxes in ST1 and 2 but couldn’t spend it on light rail until it reached 185th.

      1. “Snohomish may have had a down payment saved up…”

        That is indeed the case. Shortly after the ST3 vote, I did an analysis based on the annual subarea reports from 2009-2016. In that time period, the SnoCo subarea added $336M toward the general reserve, the biggest contribution of all subareas. The North King subarea actually ran in the red during this period and had to draw off reserves by some $89M. All other subareas were contributors to the general reserve except for North King and System-wide Activities.

  6. The more I think about it, the more I think this won’t happen. Everett leadership seems determined to open Everett Link as soon as possible. They see Everett to Tacoma light rail as being essential, and the project that should be prioritized (regardless of ridership or rider time savings). They have history on their side. As dubious a plan as it is, it really was why Sound Transit was created, which makes them even more determined. They think it should have been built years ago, so any delay will be seen as outrageous.

    1. When Snoco leaders are asked about this, one invariably gets some version of “you don’t get it. This isn’t about commuters to Seattle. It’s about the future development of Snohomish County”.

      That’s what drives their prioritization, and why both Mariner Link and the Snohomish portion of I-405 BRT got short-changed. Nobody in the room was focused on the commuter belt.

    2. It’s Snohomish’s decision, and you’re right about the politics. Snohomish is dead-set on Everett and Paine Field ASAP. Snohomish is heavily imbalanced, with the majority of residents working in King County (Seattle and the Eastside) and a large percent of King County residents working in Snohomish (primarily Boeing). Snohomish wants to generate more jobs so that its residents can work in their own county (and shop in their own county too). A German corporation looked at Paine Field a few years ago and said, “What’s your transit plan to get the majority of workers to jobs?” When it found out there wasn’t one and Snohomish’s plan was large parking lots, the company walked away. In Germany an industrial center would not be allowed if it didn’t have a high-capacity transit plan. The county doesn’t want to lose more companies like that.

      Sill, the Paine Field vision is rather asperational. Aerospace companies will come, as long as aerospace is a thing and Boeing can keep the planes safe, but will other companies locate there when it’s twenty miles from the high-income shoppers/workers in Seattle/Bellevue? Eventually they’ll have to because the closer-in places will be filled up, but how long will it take? Lynnwood’s grand downtown is still invisible, as is Federal Way’s.

      As for “urbanized hub”, what plans to increase density does the county have? If it’s not dense it’s not urban. You get off the train and then what, a mile walk on highway-like roads to the companies’ entrances? Everyone will have to use shuttles. Will Snohomish allow multifamily housing nearby so people can live near work? Will people want to live and shop next to a loud airport? Can the industrial companies be denser? Obviously an airplane factory needs a lot of space, but not every support company has full-sized airplanes on their lot. In New York multiple industrial companies live together in multistory buildings in a walkable neighborhood with a subway station within walking distance. Could Everett have some of those?

      1. I mean, if you look at how much of Boeing Everett is parking lots, it’s pretty clear that the site could get a lot more compact.

        I kind of wish they’d move all the non-production employees to downtown Seattle, shrink the Everett footprint, and lease the space to other companies. Plenty of people at the site have jobs that don’t require them to be near the factory.

      2. @Mike Orr. Re car dealerships along SR 99, you aren’t kidding. I share your sentiments. In the 15 plus years I’ve lived up here in SW SnoCo, I’ve seen these dealerships expand and new ones move in and construct these mega stores/lots on prime transit corridor real estate. (One caveat. Some of these properties were already operating as used car lots and were simply gobbled up by these larger dealerships.) It truly is disheartening to see, but I guess the land, for now, is still cheaper than building a stacked garage along with the dealership showroom/offices.

      3. As I’ve said before, the criticisms of Paine Field seem way off base to me.

        Boeing employs 30,000 people at Paine Field. Pretty much the same number that are employed at Microsoft’s Redmond offices.

        Furthermore, the station being planned drops employees off directly in front of Boeing headquarters, the largest building in the world. No one has to walk a mile down the road to get to the companies entrance… It’s literally 300 feet from station to entering Boeing.

        Of course, there are offices further from the station that would probably require a shuttle. However, the main building won’t require one. Overall, this seems at least as good as the Microsoft station, aka the Redmond Technology station.

    3. @RossB I agree with your assessment. SnoCo leaders to a large extent feel that the subarea has already been shortchanged when Sound Transit reneged on its commitment to build light rail to Everett in phase two of its expansion. That part of the history is also on SnoCo’s side as that commitment was documented with a passed board resolution (the number of which escapes me at the moment). The next slight, as it were, was when ST shortened the northward extension in the second version of ST2, after the failure to secure passage of the 2007 ballot measure (Roads and Transit). In the 2008 ST2 plan, stations at Alderwood and Ash Way were (unwisely imo) shaved off of the SnoCo extension.

      So, yes, SnoCo leaders are determined to see the completion of the light rail spine from Everett to Tacoma, as dumb as that idea has always been imho, in as short of a time span as possible. With that perspective in mind, I think they will see a two-step proposal such as the one suggested by the OP in this piece as a distraction, regardless of its merits.

      Fwiw. This is all going to change anyway. It is inevitable that we see an economic recession long before we get to the expensive part of these ST3 projects, the construction phase. The financial plan on which all of this is based is going to need to be altered as a result of that economic reality.

    4. The case for Mariner Link is even stronger for reverse commuters. Reverse commuters can’t take the 510, 511, 513, or 4xx routes, and the reverse 512 can’t use the Seattle express lanes.

  7. A variation on this proposal could be to pass by the Mariner Way Station site to a station with a new, large bus transit transfer center at 99/Evergreen and Airport Road as the early terminus. Mariner Way isn’t very well positioned for massive TOD and its park-ride function is duplicated at Ash Way anyway.

  8. YES! People are underestimating the potential of frequent feeders to Mariner. The feeders wouldn’t be like the current ST Express that get caught in traffic around the Ship Canal and Northgate and drop off to 15-30 minute frequency off-peak. Even if you’re caught in traffic in Everet it’s only a short distance to the station, and then you have a predictable, reliable, smooth, quiet ride to downtown or the U-District or Northgate or Lynnwood or SeaTac. Buses from Marysville will want to go there.

  9. If Mariner opens early it would raise the issue of its impact on line scheduling I mentioned yesterday. If DSTT2 is not open yet, the original plan of Everett-DSTT2-West Seattle won’t be able to go into effect. ST doesn’t want to run Everett-Everett trains because of driver breaks, so would Mariner-Tacoma be short enough? Or would Mariner-Redmond be extended instead? Or could Mariner-West Seattle operate? Probably not if the stub doesn’t have a turning, and the extra trains of a third line may overload DSTT1, and ST may not have enough trains for a third line then anyway.

  10. Kind of funny to look back at the comments on that old Paul Roberts article and see things like “Paine Field is unlikely to get commercial service” :-)

Comments are closed.