20131113-060625.jpgToday is the deadline for comments on study topics for the update of Sound Transit’s Long Range Plan, which will serve as the list of candidates for an ST3 (and probably ST4) package. Although there are lots of interesting ideas, there is by definition no data. When I start thinking of neighborhoods to serve, I inevitably turn to the ones I know and the ones I’d like to go to, which isn’t a particularly insightful guide to where the need is.

So instead, I’ll say this: Sound Transit, go where there’s going to be density. We should have no repeats of the density fights around Roosevelt, Beacon Hill, and elsewhere. When it’s time to pick corridors in a few years, light rail ought to go where there are already lots of units per acre, or where cities have already zoned for lots of units per acre. For various reasons, Link will have to go some places where the political will exceeds the development potential. In those cases, look for commitments to improve access: bus priority, municipal expenditures on parking, and so on.

Lastly, the plan should separate long corridors into discrete segments that can be evaluated separately. Ballard to West Seattle via downtown is an attractive concept with a certain geometric logic to it; however, if it turns out that the Ballard segment has much higher payoff, ST ought not to cripple it in the selection process by tying it to a lower-performing segment.

I can’t wait to see the data.

50 Replies to “Sound Transit: Go to the Density”

  1. It is doubtful that ST3 can afford both Ballard and West Seattle. However the city could tax themselves to build both. Ballard should be the first priority, but I will point out that as many new units are coming online this year in West Seattle as there are in Ballard. Plus, WS has only one way in and out so it lends itself very well to a rail corridor fed by local buses.

    1. the city could tax themselves to build both.

      Under what authority? In this lovely state, citizens can’t be trusted to tax themselves to buy things they want.

      Talk about a nanny state.

      1. Both the City and the County already have the power to raise excess property taxes, assuming they can get their 60% at the ballot box.

    2. I think it’s premature to declare it “unaffordable” without seeing the enabling legislation. Certainly as a matter or regional GDP it’s very much affordable.

      I agree West Seattle is an attractive target. However, its densification is from a low base, with the additional problem that there is very little on the way to downtown.

      1. If you assume that an ST3 package is about the size of ST2, than North King won’t have enough money to do both. Ballard is much further along in design, thanks to McGinn accelerating it with city dollars. It should be built to accommodate current demand.

        But West Seattle is where we want growth to go. Already it is 17% of the city population. This year 2,000 new units will come online in WS, the same as in Ballard.

        You mention the fact that there is not much to stop for on the way to West Seattle. That is both an advantage and disadvantage. You can already get to West Seattle fairly quickly by bus thanks to the viaduct and bus lanes. Although, that will end soon which is why Metro is currently insisting on reliable transit pathways post viaduct.

        You have an enormous opportunity to build market share. West Seattle is laid out well for rail with supporting bus service on 35th and up to Admiral. There are already four fairly large business and residential centers in WS–Alaska, Admiral, Morgan, and White Center. All are seeing considerable growth in multi-family. Metro has created a new transit hub at Westwood Village for several key routes and now 2,500 board there daily.

        If we want Seattle to accept more growth to forestall sprawl and keep housing affordable we must look forward and build to where growth will happen.

      2. A West Seattle line would be a commuter line. A downtown-Ballard-UW line would be an all-day line and an all-night line. Go to Queen Anne Ave. and Mercer Street any weekend night at midnight and you’ll see total gridlock. Imagine that gridlock can just jump into a station and head home by light rail. I don’t see that happening with West Seattle.

        Also, comparing West Seattle to Ballard is like a grapefruit to a grape. There are plenty of housing starts in the neighborhoods on north and east of Ballard, too. West Seattle is a peninsula.

      3. That’s now in West Seattle. The long-term plan is about what ST is allowed to propose in the future. In five or ten years, your concerns about density and ridership may be superceded. Also, beyond West Seattle is White Center and Burien. Their density is like Broadview with a bit of Greenwood, and the core part has high ridership a la Kent.

      4. West Seattle may be a peninsula, but we’ve got a lot of growth going on and insufficient infrastructure to get people in or out to accommodate the growing population. If we do get some sort of light rail (ST4 in 2040?), hopefully it will effectively connect to the broader light rail system so that a West Seattlite can travel to the eastside and the north end. Not all of us go to work downtown, yet otherwise remain in the peninsula cocoon and go to street fairs, high school soccer, and bake sales. Some of us have a life and would like to potentially get around easily around the sound without a car like everybody else.

      5. As I mentioned earlier, I was struck by the size and vibrancy of West Seattle’s urban village north of Alaska Junction as it has recently emerged, but I was concerned that moving to West Seattle would make me isolated from the places I often go: the east side of Seattle and Bellevue/Kirkland. RapidRide C is great but I fear it will slow down in later phases of 99’s evolution.

        So, here is a typical would-be resident like thousands of others, looking for a walkable neighborhood with reasonable connections to the rest of the region. Seattle is trying to encourage this with its urban villages, and Alaska Junction and Westwood Village are two such villages. That gives a reason for better transit to/from them. If you just refuse to improve West Seattle’s transit, you’re writing off 1/5 of the city’s area and some % of its urban villages that can’t be part of Seattle’s walkable/transit/convenient/affordable future — and that fact that it’s excluded would raise prices in the other urban villages as the demand squeezes into them and refuses to live in West Seattle.

        So good transit to/from West Seattle is important for Seattle’s overall urban policy, and can’t just be ignored. But at the same time, West Seattle has to go in its proper place in the priority list, as I discuss below.

      6. Two points in favor of rail transit to West Seattle:

        1. The population density (and land area, and total population) is the same as in SE Seattle. Both are around 8,000/sq.mi. So if you believe light rail was justified in the Rainier Valley and Beacon Hill, then you should support light rail to West Seattle. If you think rail in the Rainier Valley is a politically-inspired boondoogle, then you’re justified to hate on West Seattle.

        2. But, West Seattle is an even better transit market than SE Seattle due to Jarrett Walker’s rule of transit chokepoints. All commuters have to exit the “peninsula” via the West Seattle Bridge or the 1st Avenue Bridge. A grade-separated transit pathway out of West Seattle will be very competitive with auto travel due to the unavoidable congestion on these bridges.

      7. I would love to see a study of grade separated rail as well as grade separated BRT to West Seattle. In other words, how much would first class BRT to Ballard cost, versus light rail? There is already some transit only lanes, so I’m guessing additional lanes and ramps might not be that expensive. Connect it to the Sodo station from either 99 or West Seattle. If the ramp is from 99, it could be used by other buses as well. Another bonus is that something like that can be built fairly quickly. West Seattle wants a solution NOW! Tell them that they can have top class BRT a couple years from now, or wait ten years for rail, and just about everyone in the area will vote for BRT.

        On the other hand, with Ballard you would be essentially starting from scratch. There is no freeway to Ballard, and turning 15th into a grade separated line complete with a new high bridge (for buses only) would be expensive. The situation is even more striking for a UW to Ballard line. There is nothing close to a freeway there. In both cases, building first class BRT would cost almost as much as light rail.

  2. Do you think it would be a good idea to run the Sounder trains from Lakewood (the one in Pierce County) to Everett (and back) 7 days a week, Every 30 minutes, then discontinue the 590, 594, the Lakewood stop on the 592, etc., and replace as much bus service as we can with the Sounder train?
    (assuming, of course, Burlington Northern Santa Fe is OK with this)

    1. Purchasing more easements would be quite expensive, but the ability to use DMUs on Sounder should make all-day Sounder service a more attractive and feasible idea.

    2. I certainly think that some trips from Everett and Lakewood should be through-routed. BNSF will certainly be okay with running more trains during off-peak, provided that ST ponies up the cash that they will ask for.

    3. Yes, this was part of my comment to ST on the Long Range Plan. Thinking long-term, the region needs to shift from highways and highway-buses to rail for regional travel. The economics of all-day Sounder might not make sense today, but ST needs to be thinking long-term about how to move large quantities of people from Tacoma to Seattle and back on transit, and make investments consistent with that plan.

    4. I dislike the idea of creating a regional spine with such low frequency. Imagine I told you that Link is only going to provide 30 minute service past the Airport — it would be a terrible idea.

      15 minute all-day service should be an absolute minimum, with more trips at peak times. If this is impossible w/ the current infrastrucutue (I have no idea what the max frequency for heavy rail is), then we need to stop investing in Sounder and devote effort to a system that can scale to the future. E.g., I would much rather see us get HOV 3+ lanes from Everett to Lakewood w/ busses every 5 minutes on peak, 10 mins off peak, than a train every 30 minutes. Assuming travel times are competitive (if not, make it HOV 4+, or whatever), I’m certain a system with 10 min headways will get far, far more usage than a system with 30 min headways.

      1. Trips of that length don’t need the kind of frequency shorter trips do. People mostly meet their spontaneous daily needs with shorter trips and are willing to plan around longer trips or build a routine around routine ones like work commutes. Off-peak performance for long trips like this is so poor that off-peak frequency is rightfully gutted even when it was initially planned.

        Examples include BART and Chicago’s Metra (commuter rail). Despite their significant differences, both are primarily long-haul systems, both are much more mature than any Seattle rail systems, and both need to provide lots of capacity during peak and dial down the frequency off-peak.

      2. We discussed this recently. Half-hourly Sounder South would be a game-changer in transit accessibility in the Tukwila-Kent-Auburn-Puyallup corridor. Sounder is so much faster than express buses, and so much less expensive than building Link in the Kent Valley, that it’s a promising idea. ST’s long-term plan already calls for almost-hourly frequency, and it has already acquired the vehicles; it just needs money for track leases and operations. If WSDOT and ST can be prodded to build separate passenger tracks in the mainline ROW, it would allow Sounder to run as frequently as it wants alongside Cascades and the Coast Starlight and future HSR.

        Sounder’s speed advantage disappears for Seattle-Tacoma due to the wide detour to Auburn and Puyallup, but on the other hand if a separate track allows it to reach 90 or 110 mph, it would gain that back.

        Sounder North is a completely different case. It’s single-track on a coastal hillside prone to mudslides, bypassing most population centers, and with a one-sided driveshed. I can’t support expanding Sounder North, although through-routing its existing runs with the Sounder South would be beneficial. In the north end we really should cancel Sounder North and put the money into replacement buses and accelerating Link to Ash Way or Everett.

      3. Mike Orr: “Sounder North is a completely different case. It’s single-track on a coastal hillside prone to mudslides, bypassing most population centers, and with a one-sided driveshed.”

        Isn’t the route between KSS and Everett largely double-tracked now? Part of those giant payments ST made to BNSF were for the train easemnets were for capacity expansion, weren’t they?

      4. The demand simply isn’t there to fill up such large trains every 30 minutes all day, 7 days a week. To run Sounder with a schedule like that would mean running nearly every off-peak train mostly empty. Off-peak, buses are more “right-sized” to provide capacity that means actual demand.

        If operating costs for Sounder decline, due to better trains and our own tracks, I can see hourly headways midday, plus 4’ish daily trips on non-event weekends, but not much more than that.

        That is south Sounder. North Sounder already doesn’t make a lot of sense, and in 10 more years, it will make even less since. A combination of Link extensions towards Everett and non-stop shuttle buses to Lynnwood from Edmonds and Mulilteo (schedule coordinated with the ferries, of course) would be a much better allocation of resources.

      5. “I don’t think 15 minute headways are necessary for such long journeys.”

        Let’s stop assuming how long people’s journeys are, and what time constraints they’re under. Most people may be going from Kent to downtown, but some people are going from Kent to Auburn, Tukwila, Kirkland, Lake City, UW, Greenlake, or Everett. They may be staying several hours at their destination, or they may be just dropping something off or picking something up, and they may have time constraints after this trip. So, the vehicle’s speed is whatever it is, and adding 15 or 30 minutes of wait time may or may not be critical for different trips, and may be the most we can do with limited resources. But the way passengers look at it is, “Driving doesn’t have any of these constraints. I go when I’m ready, and it takes the same time as Sounder, or maybe even half as long, but much faster than most buses.” That’s one of the largest reasons people drive: because transit doesn’t go when they’re ready to go. So if we’re trying to incentivize people to take transit, it has to at least approach the convenience of driving. Constraining transit at 30-minute intervals does nobody any good; it just wastes people’s time. It may be that we can’t afford to run every 15 minutes, but that’s a limitation of resources, not a statement of value. And it should be done in a fair comparison of all the costs and benefits and alternative lines, not just a premature assumption that Sounder South can’t have or doesn’t deserve 15-minute service.

        I said above that 30-minute service would be good, but that’s only because of the the context of this corridor and the other potentials within it; it’s not a frequency that can be automatically applied to other locations. In general we should start by assuming 15-minute minimum frequency because that’s what makes the network better for humans. Then we can go down to 30 or 60 minutes if the context of that corridor or the limitations of resources require it, but at least give greater frequency a fair consideration. Because the purpose of transit is to be convenient to humans, and humans like to not wait more than 15 minutes.

      6. “The demand simply isn’t there to fill up such large trains every 30 minutes all day, 7 days a week.”

        ST told me they’re holding back three of the ST2-funded runs until ridership reaches the appropriate levels. So we don’t have to worry about this. All we have to do is decide in principle that 30-minute service is a goal, and take incremental steps toward it. And nobody is seriously proposing weekend service yet. That will automatically come over time if Sounder is well-used weekdays.

      7. Frequency is important, but simple math says that if we want high frequency, yet a reasonable cost per rider, the only way to do this is to use a lower-cost vehicle, which means using a smaller vehicle that you can run every 15 minutes and still fill up the majority of the seats. The only vehicle that can satisfy these constraints is a bus.

        “ST told me they’re holding back three of the ST2-funded runs until ridership reaches the appropriate levels.”

        The big elephant in the room is how all these new riders are going to get to the station. Parking is already at capacity and people out there have shown a general unwillingness to ride connecting buses to the station (although I’ve heard the 596 does get decent ridership), and development in the walkshed of the stations is not particularly impressive. If each new train run requires building a new parking garage that costs as much as the train itself, things start to get vary expensive.

      8. 15 minute bus expresses would be great. It’s just that agencies are more willing to install frequent trains than buses. Link runs every 10 minutes midday/evenings, while even RapidRide doesn’t match that, or the 550 or 512 or 594. And I’m concerned about the long-term cost of investing in a lot of buses that would be wasted if the rail is upgraded, plus the fact that buses are delayed in getting from freeway exits to transit centers, and in everyday congestion. So many times “buses instead of rail” translates to less frequent, less direct, lower quality service, and no willingness to spend the money to bring it up to rail standards.

      9. Al: you’re right regarding frequency, though there is a minimum frequency for service.

        Longer trips can be quite popular even on a one-an-hour basis — if they’re run on a proper “clockface” hourly basis so that people don’t have to remember very much to remember the schedule. In a number of countries train frequency is typically measured in “trains per hour” or tph, with anything less then 1 tph considered to be absurdly low.

  3. When I asked why Central Link deviated to the Rainier Valley instead of serving one of the largest employers in the state (Boeing), I was told because that’s where the density is. When I asked why East Link, then, isn’t going to the densest neighborhood in Bellevue (Crossroads), I was told it’s more important that it serve one of the largest employers in the state, Microsoft.

    You people are much smarter than me, so help me. Is it important Link serves dense neighborhoods, or isn’t it?

      1. Boeing is not and never will be a significant generator of all-day ridership. Whatever ridership it does get would be peak-only because that’s when people go to work.

    1. It doesn’t have to be either or. In the case of Rainier Valley, there was and is potential for transit oriented development and a ready made ridership which has been proven. As for Boeing, we could always infil a Boeing access road station.

      1. On one of the Link construction tours I took before it opened, ST staff told us the Boeing Access Road stop was all but dead and would be removed from all ST maps shortly. You know where else Central Link missed a huge opportunity? Southcenter Mall. I’m not blaming ST, but wow, what huge TOD potential there!

    2. I was surprised too that Link didn’t go to Crossroads, but its routing is more about Bel-Red than Microsoft. It would have served Microsoft in any case, by going east on NE 8th and north on 156th like RapidRide does. Bel-Red was chosen to complement redevelopment of the industrial area and create a walkable neighborhood that was not possible on NE 8th. It’s also a straighter line from Bellevue to Overlake.

      Boeing Plant II was never raised as a destination for Link, and it has the 124 and commuter routes. I think it has fewer employees than the other plants; at least I rarely hear about it. Boeing Renton was too far east of the airport to be a consideration.

    3. Sam,

      We all know this is a troll, not a genuine question. But the genuine answer is that East Marginal Way does not and never will have, ever, ever, ever, any demand at any time other than Boeing shift changes.

      Microsoft doesn’t do shifts. Sure, the big majority of folks arrive between 7:30 and 9:00 AM and depart between 4:30 and 6:00 PM, but their workforce in Overlake is five or six times that of Boeing along Marginal these days and they have people on JoltCola projects who are leaving at 10 in the evening night after night.

      That’s the way I/T is.

      1. Given the traffic congestion on 520 as late as 10:00, I’d guess that many of those Microsoft workers aren’t getting there by 9:00.

  4. It’s funny that you won’t let go of the density mindset of the 00s, but many cities are moving on towards something approaching sparsity…taking large swaths of decayed neighborhoods and turning them into farmland and parks!

    Detroit for example:

    Urban decay to be replaced with farmland in Detroit

    A private company is snapping up 150 acres on the Motor City’s East End — property where more than 1,000 homes once formed a gritty neighborhood — and turning it into what is being billed as the world’s largest urban farm. Hantz Woodlands plans to start by planting trees, but hopes to raise crops and even livestock in the future, right in the midst of the once-proud city.


    How about an alternatie scenario. Instead of building concrete blocks of density, buy up most of the decayed South Seattle. Run all the criminals out of town. Take the polluted Boeing areas (they’re leaving anyway) and restore the Duwamish Valley to a green condition. Then add low density garden apartments and cottage living (similar to the new Yesler Terrace designs).

    1. That is so apples and oranges I can’t believe you find them similar. No, actually I do, because Seattle is depopulating in the Bailoverse. Detroit is a vast size and has lost over half its population, so it really needs to downsize to consolidate services. These areas have zero or one houses per block occupied.

      South Seattle is not “decayed”. It’s not even like south central LA or Oakland. Tens of thousands of law-abiding residents, many of them choice residents, would be displaced by your conversion. Saying Boeing is leaving is way premature; it hasn’t even announced yet where the 777X will be, much less giving moving plans for the other lines. The Renton plant is what’s keeping Boeing afloat now, and as of now it’s running at capacity with no slowdown in sight.

      I do like the idea of restoring agriculture to the Duwamish Valley and Kent Valley, although I had more in mind bulldozing low-density office parks. In the future we will probably need more local agriculture, as it becomes more precarious to depend on high-energy long-distance transport.

      I’m not sure what you mean by “Yesler Terrace cottage design”. My understanding of Yesler Terrace is it’s going to be mostly apartment buildings and offices, with maybe some townhouses like New Holly. People have criticized it for not having enough ground-level retail and too many pocket parks, but it’s a far cry from garden apartments of the 1950s.

      1. In fact, what Detroit needs is greater density.

        The problem is, indeed, that it has zero or one house per block, and at that density Detroit *cannot support city water and sewer*, let along city roads. Electrical service is barely supportable at that level.

        So the plan in Detroit, as in Youngstown, is basically to chase out the resident of that last house per block so that the sewer and water service can be shut off and the roads can be returned to gravel. The former resident will presumably move to somewhere denser.

  5. My initial comments to ST suggested that they use Portland’s model, to the extent possible, of conditioning station location on local government’s commitment to TOD zoning and policies. If you’re willing to be Arlington County, VA on the Orange Line; 1/2 mile spacing can make sense. If a whole suburb insists on parking lots with no height or density, no stop for you!

    1. Good policy. It’s basically just a matter of admitting reality; *mass* transportation does better when there’s a *mass*.

  6. I’ve always been a huge transit supporter, and Ballard does need light rail, but that doesn’t mean West Seattle needs it lest. This is just like the districted city council format that will pit neighborhood against neighborhood. When even transit advocates like the author are suggesting we abandon some areas of the city (West Seattle) in favor of others (Ballard), then any hope of city wide mass transit is dead, as of now. Now that transit planning is to be framed even by the transit blog as neighborhood vs neighborhood, I have no choice but to play by those f-ed up rules. That means if ST3 does not include light rail to West Seattle, I will not vote for it. I just want to emphasize how destructive the neighborhood vs neighborhood mentality is going to be for Seattle, and I’m disgusted that my fellow transit advocates are apparently buying into that BS.

    1. He said if the Ballard segment has a higher payoff. He didn’t automatically assume it would. That’s what ST is doing these studies to determine. Look at it in terms of Metro’s performance metrics: some routes are in the top 25% and some in the bottom 25%. Metro’s guidelines say to consider shifting some hours from the bottom 25 to the top 25. Of course it also considers the particular route and overall coverage and other factors too. So consider that here. We’re assuming Ballard will be in the top 25% of ridership compared to other Seattle lines, and West Seattle in the bottom 50%. They may be far apart, or they may be close together: that’s what the studies will answer. If West Seattle’s results are around 50%, then there’s a good chance we should include it if we can afford it. But if it’s around 25%, then maybe we shouldn’t. And if it turns out that West Seattle is far below Ballard and that including it could jepordize the entire line (i.e., the Ballard segment), then we have to think about jettisoning it or postponing it.

      Obviously some West Seattlites believe they deserve a train or that they’re getting a lot of apartments. But we have to go by the numbers. Two thousand apartments in West Seattle is not necessarily a big deal compared to the several times that number in Ballard/Greenwood/Uptown/Belltown or their solid tradition of high multi-directional transit ridership in north Seattle and northwest central Seattle.

      To be clear, I favor light rail to West Seattle. But it has to go in at its proper priority; i.e., as its ridership, population size, distance from other lines, and two-way potential (i.e., non-residential destinations at the West Seattle end) compares to other areas that are also not on Central Link. But in any case, since this is a long-term plan, of course it should go in the plan.

  7. Martin, I agree heartily with your general premise that future HCT should follow the density, or more to the point, lead where future density is desirable and probable. But please let’s abandon that urban myth that there was a “density fight” on Beacon Hill. It never happened.

    I was heavily involved in neighborhood planning efforts during the ’90s. Beacon Hill was one of the first neighborhoods to embrace Mayor Norm Rice’s urban village concept. Our neighborhood plans supported upzones around the Beacon Hill Station site. We supported and fought for station construction (ST originally wanted to just mine out the station shell and finish it “later”).

    When the most recent upzone finally got to City Council a couple years ago, the community supported it, allowing El Centro de la Raza to proceed with its major redevelopment project adjacent to BH Station. The major problem remaining is convincing upzoned property owners to take advantage and build something. All that vacant land around BH Station is embarrassing.

    Please stop assuming that Seattle neighborhood activists are reflexively anti-density. I know it’s as convenient meme, handy to throw out there in lieu actual research, but it’s getting tiresome, and offensive. Most neighborhood people I know are fine with density increases in and adjacent to their neighborhood business cores, where density will contribute to a more vibrant and walkable neighborhood.

  8. I agree, Sounder North needs to be trimmed. Since the mudslide-prone area is between Edmonds and Mukilteo, why not truncate the line at Edmonds and add stops at Shoreline/Richmond Beach and Ballard (Edmonds to Seattle train)? Any extra money – including savings from not having to pay for replacement buses (during the mudslides) – could be funneled into adding buses to the Mukilteo-Seattle and Everett-Seattle segments.

    1. If you’re going to build a Sounder stop somewhere in Seattle, I’d prefer one at 20th and Dravus. Magnolia’s transit ridership is primarily made up of commuters, and the stop would be in a location where virtually all Magnolia drivers would have to pass through anyway. In contrast, a Ballard stop would be way out of the way for most Ballardites, and it would do little to address the all-day mobility needs of Ballard.

Comments are closed.