The Rapid Transit Plan for the Metropolitan Seattle Area - Metro Area Perspective - 1970

Fifty years ago this week, February 13, 1968, 50.8% of Greater Seattle voters voted yes to the Forward Thrust rapid transit proposition. The construction of a 49-mile rapid transit system modeled after BART and the Washington Metro needed voters to approve $385 million in general obligation bonds. The remainder of the $1.15 billion cost would be picked up by the federal government. Unfortunately, the state constitution requires a 60% majority to approve such a bond so the measure failed and the rest is history.

But for the sake of this post, let’s assume it did pass and the system got built as planned. What would a Metro map look like? We’ve seen the scans of maps from the plans. What we’ve not seen is how the service would have operated. Here is a diagram I made that presents the Seattle Metro rapid transit system as if it were in operation in 1990, five years after completion of the initial system plan after several phased openings than began in 1976. I wanted to create a map with a 1970s design aesthetic but not clone the style of those iconic transit maps of the era.

diagram of Seattle rapid transit had it been built
click to enlarge

The plan outlines the physical route segments, named after the area they served but did not develop operating lines. There is enough information in the plan and its technical appendix that I can infer three primary rail lines. Note that the lines and their names are my creation.

  1. East–Northwest Line (Red) connecting Bellevue, Mercer Island, downtown Seattle, Belltown, Uptown, Interbay, and Ballard. It is much like East Link joined with the Ballard Green Line.
  2. South–Northeast Line (Blue) connecting the Boeing facilities in Renton and the Duwamish with downtown Seattle, South Lake Union, Capitol Hill, the northern part of the Central District, Madison Valley, the University District, Roosevelt, and Lake City. It is the north-south line that serves part of today’s Link Red Line.
  3. Eastgate Line (Green) serving the Eastgate spur with trains coming from the Northwest and Northeast legs of the system during daytime hours.

The plan, for cost estimating purposes, assumed trains would operate 20 hours a day, 5 am to 1 am, every day, with closing time extended by an hour on Friday and Saturday nights. Hourly owl buses would connect all stations when rail service is closed.

The base daytime frequency on each line would be every 5-7.5 minutes. Rush hour frequency increases to every 4-5 minutes, meaning the core downtown segment between Jackson and Pine would see trains consisting of 6 or 8 cars arriving every 90 seconds. In the early morning and late evening periods frequency drops to every 10 minutes and the Eastgate Line does not operate.

Each rail car would be about 70 feet long and 10 feet wide with 3 doors on each side and wide aisles and seats for 74 passengers. Cars would be coupled in pairs to form trains of 2, 4, 6, or 8 cars. The trains would be electrically powered through a third rail and be capable of reaching a top speed of 75 mph. This is comparable to BART and Washington DC Metro rolling stock and train lengths.

The bus lines included on the map are:

  • West Seattle busway (Orange) providing an exclusive 2-lane pathway for buses between 35th & Avalon in West Seattle and a cross-platform bus-rail transfer at the Spokane station in SoDo. Buses from California Av, Delridge Wy, Admiral, Fauntleroy, Gatewood, 35th SW/Arbor Heights would use the busway to connect with rail.
  • Rainier Beach express bus line 21 (Silver) operating all-day every 10 minutes from Rainier Beach bus station to Duwamish station with less frequent service continuing north of Rainier Beach to Jackson Station in the International District.
  • SR 520 and I-405 express bus lines 60 and 66 (Silver) combined operates every 7.5 minutes during peak and every 15 minutes all-day from Northup Way station to Pacific station via SR 520. Line 60 continues to Bothell via I-405. Line 66 continues to North Redmond via SR 520.
  • Aurora express bus line 64 (Silver) feeds Crown Hill station with service from Richmond Highlands bus station every 4 minutes during peak and half-hourly until 7 pm on weekdays. The 47-Aurora bus provides local service half-hourly seven days a week.
  • Airport bus line (Gold), although not proposed to be frequent outside peak, provides a connection from Duwamish station to Sea-Tac Airport. The plan studied a southwest rail route branching from Duwamish station through either Burien or Riverton Heights but did not recommend it be constructed in the initial phase due to PSGC (now PSRC) plans then to construct a second regional airport and cap Sea-Tac airport to 20 million annual air passengers. It did make provisions for Duwamish station to handle such an extension in the future. One wonders how including an airport line from the beginning might have influenced voters.

I have changed some station names from their original plan names. The Boeing associated name of Plant II is changed to South Park for the bridge near it. University Hospital is changed to Pacific. Rainier changed to Judkins Park. Wedgewood is corrected to Wedgwood. Northrup Way is corrected to Northup Way. A provisional Wilburton station in the recommended East route and future Tukwila station on the South route was included.

For reference, I have produced a Google map of all the routes and stations in the plan with their projected opening dates.

81 Replies to “Seattle Metro Rapid Transit Map “circa” 1990”

  1. That’s a pretty good network. I especially like the line to Lake City, near where I live. It’s now 28 years after these lines were originally supposed to be in operation. Today there is just one line, from “Pacific” to the airport. Even in another 25 years, after ST3 is finished, there still won’t be a line to Lake City or Crown Hill, though the rest of the network will be in existence. Over 50 years late.

  2. It’s interesting to imagine what an expansion of the system would have looked like.

    The bus connection to the airport would have likely been upgraded to rail and then possibly extended further south.

    The east side lines could have been extended to Issaquah, Redmond, and possibly Kirkland.

    The Renton line could have been extended around the bottom of the lake and up to Bellevue.

    The Ballard line could have curved back to either 99 or I-5 and continued up to Lynnwood.

    The Lake City line could have headed off to Bothell or turned back to I-5 and continued up to Lynnwood.

    The Ballard-UW line would have been even more necessary given the two north seattle lines are father apart in this plan than our modern day.

    1. Agreed. That was my first reaction that an airport line would have been built. That’s a pretty good guess for future expansion. Likely a lot on the Eastside.

      Gasp… Three lines in one downtown tunnel!!!

      Where would Arboretum station have been? 520? By Montlake library? or further south by even Madison Valley?

    2. A long range plan was presented on page 19 of the plan I linked. It’s described it as “a logical ultimate development of the rail rapid transit network”. They didn’t specify a timeline on when these extensions would be needed.

      Ballard would extend north along reserved ROW to the Interurban to the Lynnwood area and possibly beyond.

      Lake City extends along the Burke-Gilman to Bothell and possibly into Woodinville.

      Bellevue extends from Northup along reserved ROW to Redmond along SR 520 and possibly beyond. A branch diverges from downtown Bellevue up along the I-405 corridor through Kirkland to meet the Bothell extension.

      After Judkins Park a line branching off to serve the Rainier Valley, intersecting near Boeing Access Rd, splitting into two branches, one going north to join the West Seattle line and one south to Burien and the airport.

      South of downtown, a connection between the east and south legs of the system.

      Eastgate extends to Issaquah.

      At Eastgate, a line branching south through Coal Creek area and Renton, rejoining the Renton line and then branching south at Tukwila along the old Interurban route towards Tacoma.

      1. Oran and everybody else, been waiting for this question, so I guess it’s time. Second vote passed, but not by enough.

        Especially after 2016, time for somebody needs to point out that “Supermajority Requirement”means “Minority Rule.”

        Why wasn’t there a third vote? 1964, maybe? Or 1968- lot of energized demand for change. Or 1972- might’ve left a few more lights on.

        Contrarian, maybe, but necessary inquiry after defeat that close: Any chance that backers themselves decided that considering the terrain, and even more, the pathetic inheritance of right-of-way, the project could not yet deliver the system promised?

        Also, I wonder if the truth massive tunneled and elevated urban transit is never led into existence as an anticipated future need. But rather, blasted into existence from behind by a large enough tax base fed up for years about being stuck.

        Might be worth a trip to Downtown Library 10th floor to see if any there’s any post-game analysis.



      2. Andrew, the federal funds earmarked for Seattle went to help Atlanta build its rapid transit system. In 1972 a bus-only vote that created Metro Transit was successful. They would get some federal funding for buses but not nearly as much.

      3. I think it’s important to note that how Seattle funds went to Atlanta had nothing to do with some formal criteria-driven decision-making process in DC. It appears to have been mainly political. Here’s a bit of what I’ve learned about the national history on Federal transit capital funding (admittedly not familiar with the Seattle context of that period)…

        The whole eligibility of Federal funding for major transit investments has several chapters to it. It strongly emerged as a response to the Freeway Revolt in the 1960’s, where the Federal government would give money for highway projects but if a city cancelled a project because it was removing neighborhoods, they didn’t have a system in place to exchange those funds for transit.

        I can’t easily find a good summary of the early days of Federal transit capital funding, but here’s a link describing the process that UMTA (now FTA) followed in the 1980’s. (

        My abbreviated observation on how this as played out:

        Before 1984 when UMTA started rating projects on cost-effectiveness (note this is the Reagan administration), there wasn’t any sort of a Federal review process to see which projects had the greatest merit. There were some notable transit projects that failed to produce any of the ridership that was originally forecasted (keep in mind that computers in the early 1980’s were in giant air-cooled rooms, run with punch cards, reviewed with computer printouts that were basically automated typewriters, and had very slow processing power). The forecasters in some metro areas (Miami being the most notorious) began to play with things like artificially increasing employment numbers and parking costs in the basic calculations to inflate ridership. Inflated forecasts were common and resulted in a culture of skepticism and derision about forecasting in that time period.

        The cost-effectiveness ratings ‘game’ refined through the eightees. Eventually, there was lots of complaining that there were other factors about rating that should also be considered beyond a single cost-effectiveness number. Forecasting became more transparent, and FTA started doing thorough independent reviews and verification of the forecasts to make sure that local metro areas weren’t padding their numbers, and setting rules on what kinds of forecasts that they would accept.

        The 1991 ISTEA changed things substantially. It renamed UMTA to become FTA. It required metro areas to quit publishing unrealistically expensive long-range transportation plans that they couldn’t afford. It required justifying future land use assumptions. It created several criteria on how to evaluating if a project deserved funding. It put MPO’s more in the role of coordination and reining in grandiose transit proposals.

        I’m adding this here not to be an anointed expert on the topic, to revise history or to defend what I perceived what happened. I just think it’s important to understand how Federal transit capital project selection generally happened for those who were too young to see it happen.

  3. The biggest difference I notice with this plan is that in the absence of sub area equity, the lines are built more purely based on performance/merit. Places that don’t warrant rail are given BRT or express bus service and there are no lines snaking far out into the suburbs.

    I wonder how those bus lines would have fared in given the anti bus land political pressures that forced the nearly exclusive use of light rail in the ST expansion plans.

    1. The situation in 1969 was very different. King County’s population was 1.15 million vs 2.15 million now. Seattle had half of that, so the built-up Eastside, South King County, Shoreline, and Lynnwood as we know them didn’t exist. Bellevue and Tukwila were still transitioning from farmland. Bellevue’s size was somewhere between Snoqualmie and Puyallup. Auburn and Everett were separate job markets: hardly anyone commuted to Seattle from them. The only long-distance commuters were Boeing workers, who were often required to switch between Everett, Renton, and Seattle so they couldn’t live near work. Northgate was not fully enclosed yet, and Southcenter was built a year earlier in 1968. Metro was an independent transit-sewer agency serving approximately King County. So it’s funded entirely by King County taxpayers (with federal grants), covering King County’s then-population (Lake City to Renton) and planned growth area (the Eastside).

      How that would have affected growth by 1990 is unknown of course. In reality, King County’s population in 1990 was was 1.5 million (350K increase). The commuter belt was Lynnwood – Bothell – Redmond – Renton – Kent – Des Moines. Auburn and Tacoma were still not common commuter origins. In the 1990s the Everett – Seattle – Tacoma job markets merged and the commuter belt stretched to Everett – Woodinville – Issaquah – Auburn – Puyallup – Lakewood. That’s because people moved to low-density residential developments that gobbled up land.

      As we see from BART, high-capacity transit in the 1970s did not lead to dense station areas, so we can assume its ability to catalyze urban villages would be modest. That only started in the 1990s with the “back to the city” movement and later climate change, rising gas prices, and the housing-price crunch.

      Is Oran’s plan based on 1990 as assumed from 1969, or 1990 with actual population distribution (which as I said above would not be influenced much by the subway)? Because if it’s actual, I’d expect more extensions and infill, at least with bus routes. Because Metro served all of King County, and Federal Way, Kent, Redmond, and Bothell, were built up by then.

      There was one other growth decision that affected things. In the 1980s King County deliberated between channeling growth to “metro towns” (satellite cities), “concentrated” (one mass in Seattle), or in parallel notes (north-south oblongs in the Eastside. It chose metro trowns, This is supposedly what we have now, although the concepts showed natural green land between the islands rather than subdivision sprawl everywhere. I’m not sure if that was an inability to anticipate the consequences of sprawl, or a bait-and-switch by subdivision advocates. But this raises the question, if the rapid transit had been approved, would the subsequent decision how to channel growth have been any different? For instance, one could postulate a single mass over the rrapid-transit area. Or the metro towns might have been chosen differently, or growth outside the towns managed differently. Are any of these likely?

      We should not overlook the rapid change in the 1970s and 80s. Let that sink in: Bellevue and Kent in 1969 had the remnants of rural houses, Kent had a streetcar-suburb center (e.g., James/240th Street), and Bellevue had a one-story center. Most of Kent and Timberlane’s houses date to the 1970s. Mountlake Terrace was built up just after WWII. Lynnwood’s residential mass was built in the 1970s, and the Mukilteo subdivisions in the 1990s and 2000s. All this was based on decisions then, in the 1970s through the 2000s. If the cities and region had decided differently, it would be much different.

      1. The photos in the linked planning document show that the 1970 versions of Bellevue, Redmond and Kirkland were very much less developed than they are in 2018. Bellevue Square is completely unrecognizable in its photo. The drawings in the plans also show that there wasn’t much attention paid to the concept of TOD around the stations. Most of the drawings show trees and pastoral vistas surrounding the stations outside of Seattle.

      2. @Mike Orr. Great post. Good questions too. Obviously we can only specupate about how things would’ve worked out had this proposal passed and these plans come to fruition.

        “But this raises the question, if the rapid transit had been approved, would the subsequent decision how to channel growth have been any different?”

        This question brings up the subject of WA’s GMA which I believe was enacted back in 1990. Would such a plan had been pushed for earlier than it was and would there have been a consensus around said plan in the communities most directly impacted by it at that time?

        One minor correction. There were many single family homes (many of them being small ramblers) built in the Lynnwood and Edmonds and Snohomish SW MUGA in the 1960s. These were built on some of the original plat subdivisions, such as the one my home sits on. The split-level building boom came in the 1970s and early 1980s.

  4. Interesting… a busway to serve West Seattle. Definitely what it should be now given the geographic layout of WS. And I bet the cross platform bus – rail transfer at Spokane would have been really good, way better than anything we would get today. MARTA has a few stations that I’m aware of (Arts Center & Buckhead) with mezzanine level bus transfers that are within the fare paid zone (just as Toronto). We need more easy seamless fare paid zone cross platform or mezzanine transfers between bus and rail.

    1. Another example is the Harvard Square MBTA station. They have the trolley buses go into a tunnel the few blocks around the station. The routes that go through there disperse in a number of directions.

      It could be a solution to some of the bus-rail Subway transfer challenges coming up in ST3. It would probably require Metro providing some funding and support for it though.

  5. Westlake and a Pike-Pine station??? Perhaps Westlake was a few blocks further east to squeeze in Pike Pine station by the market. The other downtown stations are all pretty much the stations we have today.

    And look at those Capitol Hill and CD stations!!!

    1. Westlake Station location in the linked plan would have been on Westlake Ave — but further north, like around Lenora. It’s ironic how that Station would have been a great place today given how it would be between Amazon and SLU.

    2. SLU’s development was anticipated but stalled for decades. Construction halted from 1929 to 1945 because of the Depression and WWII. Afterward came the freeway plans, including the Bay Freeway at Mercer, the 99 freeway through SLU, to form a ring road (I-5, Mercer, 99, Spokane St). Only I-5 was built, but the city punted on articulating the zoning vision for SLU. So it stagnated through the 1950s to the 1990s with decaying warehouses amidst the slash woulds, of I-5, the Mercer Mess, and the 1930s Aurora expressway. In the 1990s Paul Allen bought up land and proposed a series of visions that failed (the Commons, a biotech hub), until finally the tech+streetcar hub succeeded, the city finally zoned for it, and Amazon’s HQ move and its successful marketing of “cloud computing” catalyzed the highrises.

    3. I suppose the Bay Freeway would have been a block further north at Valley Street, right next to Lake Union. There was also going to be a short N 50th Street freeway between I-5 and 99.

  6. The map lines were likely influenced by the stadium proposal. Some of the leading contenders for stadium siting were Seattle Center, South Park and Northup Way. Fifth and Yesler was also studied as a location, but the actual location where the Kingdome was built wasn’t considered at that time..

    In 1968 the high level bridge to West Seattle did not exist. Light rail would have needed to build its own crossing over the Duwamish. In the February 1968 election voters did approve money to start building the new bridge and other auto-centric transportation projects.

    It’s not mentioned in the Forward Thrust literature, but Boeing would have been the likely builder of the rolling stock. Boeing did manufacture light rail vehicles in the 1970s for San Francisco and Boston. It would be hard to imagine them losing the King County order.

    1. This was to have been heavy rail. Boeing never evinced any interest in competing for New York’s huge R-series car orders during that era. It’s not clear they would have been interested in this much smaller fleet.

      1. The plan submitted in 1970 it was modified to call for 330 light weight, electric powered rail vehicles. I didn’t know the 1968 plan specified heavy rail only.

      2. “New York’s R-series car order”

        Where does New York have light rail? There’s the Hudson-Bergen rail in Jersey City but that was surely built later.

  7. Interesting how 1990 called for 2 am service on Friday and Saturday and yet, even with the UW station and Capitol Hill stations opening, Sound Transit won’t provide service until bar close any night of the week in 2018, let alone Metro leave downtown tunnel open so the last trains from SeaTac can get tourists and residents home without getting on a bus at Beacon Hill. Seattle deserves big city service on the weekends to get students, late-night workers in hospitality (and hospitals), and people who have been drinking home safely.

  8. Great work Oran. Fascinating stuff. I really like that map — you definitely captured the style of the time. Very nice.

    1. The beautiful thing about the map is its simple geometric elegance. It conveys how to ride on the system and not what highway the system is near.

      It’s great!

    2. I imagine the appearance of stations would have been like those in Buffalo which opened in 1985 just like this but given the design and construction time would have been designed in the late 70s as this would have.

  9. Great map! What a shame this wasn’t successful – it would have been a fantastic system that would have transformed the region. With hindsight, the only key things really missing were connections to the new downtown in SLU and across north Seattle.

    As others have noted, the difference with this system is that it was designed to serve high density travel demand, not to distribute political capital. The area where rail transit demand is high is the same area where walking is possible – it’s pretty much constrained to places with higher density in Seattle and the close-in suburbs. None of the stations (other than Mercer Island) are along freeway rights-of-way where it’s hostile to walk, unreliable to transfer, and insane to put high density housing. The idea of using urban rail to connect cities 30 miles apart would not have entered any planners’ minds back then, for good reason.

    On top of that, this would have been a heavy rail system, not light rail limited to four-car trains. It would have been mostly in subway. Light rail became the great hope when the feds stopped investing in urban rail, and surface light rail in abandoned rail rights of way seemed like an inexpensive option. In contrast, if our planned system is at all successful it will be well over capacity, while still only serving less than 10% of trips – so we will need to start thinking how the rest of our growth will be accommodated soon.

    1. The drawback to this system would have been the lack of appropriate zoning laws in 1970. The plan wasn’t to overturn the suburban paradigm; rather, the plan was to maintain the suburban model but offer more bus service to it. At the time of this vote, Metro Transit did not exist and suburban transit service was almost non-existent. Transit Oriented Development (or Transit Oriented Communities) were not recognized terms in 1968.

      1. That’s not entirely true though. In the plan document pages 20-21 explicitly discusses development potential around bus and rail stations and calls out the need for coordinated planning and zoning.

        The law authorizing Metro permitted it, with voter approval, to perform “metropolitan comprehensive planning” which includes preparing a metropolitan comprehensive plan and the power to “review proposed zoning ordinances and resolutions or comprehensive plans of component cities and counties” for conformance with the regional plan.

        Another point often missed in Forward Thrust discussion is that this was a comprehensive transit plan. We would have had a unified transit authority for the metro region running both the trains and buses under a single fare system.

        Jim Ellis was a visionary.

      2. Yes, Jim Ellis was a visionary.

        The document you have linked is the plan for the 1970 vote, not the original vote in 1968. It’s a little confusing to be talking about the 1968 vote and linking documents to the 1970 plan. The 1970 plan does call for coordinated planning and zoning, but we all know it’s a long journey from “the call” and actually accomplishing coordinated planning and zoning.

        Metro Transit was created as a response to the 2 failed attempts in 1968 and 1970 to authorize a county-wide mass transit system. Neither of the failed proposals extended beyond the borders of King County.

      3. This post was originally based on the 1970 plan and was written a few months ago (the map itself actually was done in 2016). We just realized that the 1968 anniversary was coming up and so inserted the reference to the vote at the beginning.

      4. The plan may have called for regional planning of land use, but I think the politics of the 1970s would have yielded very similar land use to what happend around suburban BART, metro, and MARTA stations. Seattle’s politics wouldn’t have been that different than its peers at the time.

    2. Actually, as noted by some others, there would have been a station about a block from Amazon at Westlake and Lenora. And two First Hill/Central District stations.

    3. “it was designed to serve high density travel demand, not to distribute political capital.”

      No, it was designed to serve low-to-medium density travel demand. In 1969 the only buildings above four stories were in downtown Seattle, college campuses, and a few scattered ones before the zoning caps banned them (Beacon Hill tower, Madison Park tower, University Tower). Bellevue may have started planning its highrise downtown, but it was over a small area as evidenced by what did get built. So don’t imagine urban villages at all these stations; that’s an anacronism. Only if the subsequent land-use planning had been decisively pro-urban would that have happened. Theoretically that planning might have been like Vancouver and DC, but knowing this region’s mood that sounds unlikely.

      1. Mike, the Rice era urban village policy did not change zoning at all. The zoning for density existed long before that. All the urban village policy did was to recognize that the places zoned for density would be where all the growth would occur, so it committed to prioritize the city’s resources into providing the urban infrastructure needed for them to be high quality urban neighborhoods. The rapid transit system was designed to serve that zoning.

      2. But the zoning was still mostly four stories, or where was it higher? Broadway didn’t get six-story zones until a decade ago. Bellevue Avenue was six stories earlier but was almost completely residential-only. It’s possible to have density with four stories as in Edinburgh, Paris, and Boston, but not they way they were built in Seattle.

      3. High rise does not always equate to high density. Mid rise can be much denser as evidenced by Paris as just one example.

      4. In Seattle a lot of 4-story zoning remained unbuilt because it was not enough of an increase over the existing single-family houses and 1-2 story commercial buildings to be econonomically attractive, especially with the large setback requirements that were then required. And the four-story buildings that were built just didn’t have that many units, again because of setbacks and the desire for open space and parking and a lack of desire for density.

      5. Height ≠ density. Paris, Brooklyn, San Fransisco and even Manhattan all have neighborhoods full of short buildings that are densely populated. Relatively speaking, places like Capitol Hill and the Central Area are (or were) similar back in the late 1960s.

    4. “surface light rail in abandoned rail rights of way seemed like an inexpensive option”

      Unfortunately Pugetopolis doesn’t have legacy rights of way in the right places like Portland, Vancouver, the East Bay, and many other cities did. The Interurban ran on Linden and the Duwamish to Lynnwood, Kent, and beyond. The Burke-Gilman route bypasses the population centers in the U-District, Wallingford, Roosevelt, Maple Leaf, Northgate, and Lake City. I don’t know if there was ever rail to Des Moines and Federal Way (west of the Interurban), or if they depended completely on Highway 99. The Eastside had a single low-volume north-south track, and some kind of Renton-Snoqualmie line. But our postwar development completely ignored these and concentrated people where there had never been rail, and now we have to retrofit it with new rights-of-way where houses have already been built. It’s either that or move the population centers back to where the legacy ROWs are, but that’s not going to happen. Federal Way wants its Link please because it’s entitled! (it has an I-5 exit!), and doesn’t want to move its population to Auburn and Kent. The Eastside does not want to evacuate and move to Seattle. Both of these areas are approaching a million population (south King County is already above 800K) so high-capacity transit cannot be punted any longer.

    5. The “limited to four-car trains” thing with Link isn’t quite the problem it sounds like. A single Link car has two sections and is about as long as two cars in older systems like the New York subway or the Chicago L. BART, the crazy outlier of US transit, runs 10-car trains with big wide cars and manages about twice the capacity of a four-car Link train. My understanding is that BART makes full use of that capacity crossing the bay; the Bay Area has twice our regional population and nowhere do we need to funnel so many people across a single chokepoint. We have four different ways across or around Lake Washington with frequent transit service and six across the Ship Canal. I guess transit between West Seattle and the mainland is mostly concentrated on one bridge, but that’s a small enough transit market that the choice was between a single high-capacity line aggregating transfer from around the peninsula (the ST3 plan) and a BART-in-the-East-Bay-style fanout using buses (as shown above).

      The only reason train capacity is even worth worrying about here is that we’ve decided peak-hour commuters from all of Snohomish County, the Northgate transfer-shed, the 520 transfer-shed, the UW transfer-shed, and Capitol Hill all go on a single set of train tracks! If transit is popular enough that it becomes a serious problem, we’ll have the political will to fix bus routes that relieve the crowding using existing road space.

      1. “The only reason train capacity is even worth worrying about here is that we’ve decided peak-hour commuters from all of Snohomish County, the Northgate transfer-shed, the 520 transfer-shed, the UW transfer-shed, and Capitol Hill all go on a single set of train tracks!”

        Almost. If we didn’t build the train capacity, everybody from the north would be funneled onto I-5 and 99 which are severely overcrowded peak hours and can’t handle any population growth. The alternatives of more highways or more parallel train lines were both rejected.

      2. Link train cars are 95 feet long. BART’s is 75 feet long (admittedly wider though). CTA’s are around 48 feet long. Thus, one Link car is about 1.27 BART cars or 2 CTA cars. Of course, the actual capacity comparisons vary based on seats, configuration and driver/cab space. As a rule of thumb though, a Link four-car train is like a 5 and 1/3-car BART train or an 8-car CTA train.

      3. Exactly Al. Worries about capacity are really overblown. It might get a bit crowded between Capitol Hill and Westlake once in a while, but if it ever becomes a big problem we can run express buses or spend the money improving headways.

      4. Maybe so.

        Still, ST has not publicly presented an analysis of a peak hour line and station capacity analysis post-ST3. An analysis should include elevators, escalators and stairs — particularly at transfer points. It really needs to be part of the background reporting before any of these ST3 stations can be planned, yet I still cannot find one. I don’t think ST has published even a post-ST3 station-by-station average weekday boarding table, so that we can even guess where station capacity may be an issue. The absurd thing about it is that ST has the data on what the station boardings are forecasted to be because they put in segment forecasts for all the ST3 segments. I guess it’s because no elected board member has demanded this basic information.

        I’ve commuted on light rail systems where I couldn’t board the train because it’s too crowded. It’s very frustrating, especially when I have had a meeting to attend!

        I’ve been on platforms where escalators couldn’t get me off before another train arrives. The crowd surges are pretty scary! I’ve heard personal stories about people getting pushed off platforms and in front of trains to their deaths because the platforms were so crowded!

        Any analysis also needs to inform how if there is an equipment failure like a broken escalator, they have methods and procedures in place to handle the accompanying crowd surges.

      5. Not all of Snohomish – the west edge can use North Sounder, and eastern Snohomish can use 405 buses and transfer to East Link to get to Seattle. But generally, yes.

    6. “Light rail became the great hope when the feds stopped investing in urban rail,”

      Light rail was an international phenomenon. It started in the 1970s when the Europeans modernized their tram technology for a new system for cities the size of Bellevue (Bielefeld) to Seattle (Duesseldorf): a surface rail with a downtown tunnel. They also used it for more grade-separated systems like Link, called premetro, with the intention of later converting it to heavy rail when the population rose enough to justify the expense. But after it was in use they found that it was adequate even for growth so they didn’t need heavy rail after all. The same technology started appearing in the US as San Francisco renovated its MUNI Metro and the first light rails were built (e.g., Portland, San Diego, San Jose), although unfortunately they didn’t build downtown tunnels, since the priority was low capital costs so drivers would agree to pay for it.

  10. > $385 million in general obligation bonds. The remainder of the $1.15 billion cost would be picked up by the federal government.

    Wow, imagine how great (and fast) ST3 could be built if the federal government would foot 3/4 of the bill. Maybe someday we’ll decide as a country to invest in our cities and towns rather than our military.

    1. The nation will (hopefully) be two by that time. We need to throw the boat-anchor of the Confederacy and Copperhead Midwest into the Gulf of Mexico.

      1. History suggests different outcome, Richard. When Lincoln said that our country couldn’t survive very long half-slave and half-free, it wasn’t rhetoric.

        Forget Vladimir Putin. The Secessionists already had a solid alliance with the World’s biggest military powers, Britain and France.

        Who’d always considered the United States a brainless experiment. Along with a fair number of Americans, south and north. We’d be lucky to be in Canada.

        Midwest contained Wisconsin and Michigan, which lost a lot of men on right side at Gettysburg. Same with Eastern Kentucky and Tennessee. Lose the mid-west, and Canada would just be smaller.

        There was never a vote for Secession. And main reason the South lost was that ordinary British and French military men wouldn’t fight for slavery. And the Northern States got more racist after the war.

        So average transit blogger on the world-famous Orange Blossom Special might be indignantly demanding riddance from those pathetic, ignorant, lazy Washingtonians sneaking in to take their IT jobs. And steal their Amazon headquarters too.

        Careful what you wish for.


    2. “Maybe someday we’ll decide as a country to invest in our cities and towns rather than our military.”

      Maybe we raise taxes and invest in all three. This is a transit blog, not a US Naval Aviation blog but I should share I’m not too happy at the Trump tax cuts when we have a massive maintenance backlog in US Naval Aviation.

      I’m also flaming mad at the Feds walking away from ST2, especially as ST2 is to build out some of this map outside of Seattle. It’s like Trump wants to give New York City and Seattle money to build rural roads and give corporations tax cuts.

      1. It’s amazing when Trump is from New York City and builds skyscrapers in Manhattan. How can he not know that urban transit is essential for Manhattan to function and for his skyscrapers to function? Or that New York is the biggest jobs engine in the country and subsidizes the rest of it including those rural areas?

      2. Here’s a better way to put it, Joe:

        Proposed fleet of 2,457 F-35 jet fighters drawing very bad reviews in the trade. No bargain at $160 million each, let alone $400 billion whole order.

        Somebody who knows: If somebody puts a low yield warhead on Bangor,
        which use would save more American lives: These planes, or equivalent money for transportation, firefighting, rescue, and hospitals?

        The Interstate System came in under The National Defense Highway Act! If freeways count as National Defense…why doesn’t public transit? Follow-up question: Why haven’t I heard any elected official ask that question in my life?

        Mark Dublin

      3. The defense purpose of the Interstates was a sham: it was an excuse to justify the federal government intervening in the economy in such a large way, because nobody would question “national security” in the early Cold War and McCarthyite era. We’ve never had the miltary take over the freeways for troop and tank transport, and even if it did have to it would not take up all the freeway capacity for a significant amount of time, or require four-lane highways in Montana and North Dakota. They just needed to modernize the existing federal highways and perhaps build wide two-lane roads where there were gaps, and keep the rail network intact which they let decay after they decided to build the Interstates and airports instead.

        Why public transit wasn’t/isn’t considered a national security necessity is a good question. A robust transit network would make the country more resilient in the face of a variety of unpredictable challenges, including the reduction of foreign energy supplies. That’s what Europe has been doing since the 1970s but the US went the opposite direction. As to why the US didn’t include transit in the national defense mobility plans, see above about the Interstates and airports. Private intercity buses and airlines were to be our national mobility solution. I’d almost include public metropolitan buses, but the Interstates were not originally intended for metropolitan travel but for long-distance travel, and they did not foresee the rise of greenfield communities and shopping centers at metropolitan freeway exits, and assumed all the existing trains and buses would continue running as they still were in the 1950s.

      4. For every minute of the time since current resident at 1600 Pennsylvania announced his campaign, his eveyr word has had one thing in common with every other- in addition to being lies.

        Every single word is a precision-calculated diversion from something also in a category if its own. Each having two unvarying characteristics of their own.

        One, major damage to the Republic. And number two, so boring the media won’t touch them. Cabinet Departments- Energy, Interior…massive vacancies left vacant. Think anybody had to be forced to leave?

        Others, filled by people who’s main purpose is to turn their agencies over to the industries they’re supposed to regulate. The rules of Capitalism have turned places like Texas away from coal and toward solar and wind energy.

        So the Administration is working its rear end off seeing to it that the owners of these dying industries walk away with the last multi-trillion dollars, at taxpayers’ expense. See (closely) “Infrastructure.”

        So let the poor Russian fake-twitterians go! ESPECIALLY if they were hired by the Administration to turn Jill Stein against Hillary. Because every jaw-dropping revelation keeps the cameras off another obsolete fossility getting its permit!

        And let’s face it. It’s not FOX News’s fault that spectrum wide, whole viewing audience knows civil servants are BORRRRRRING!” Giving the Resistance a painful new mission:

        The Anarchists will never get their morale back from being robbed of their target. So anybody willing to die rolling their eyes and yawning in defense of the Republic….

        Forget destroying the Establishment. Your mission is to BECOME IT! 2012, you be in here with a degree in Public Administration…And know how to use it!



        For real early starters.

    3. You can’t just divide the country at the Confederacy and hope for unity. People with both visions live all over the country. Even the reddest counties have Democrats and liberals and vice-versa. While there may be a difference in conservatism in different regions, with the old south focused on white supremacy and the west more interested in individual libertarianism, it’s a continuum rather than a sharp break.

      1. That’s very true, Mike. And the localized demographic shift in more red states is that the more blue metro areas within them are where the population and economic growth is occurring. Most of NC growth is around Charlotte or the Triangle. Most of TN growth is around Nashville. It’s not unlike what is happening even here in WA, with the Seattle metro area driving the growth rate. There are many other states where a more blue metro area is the hot growth area of the state in the Midwest and the South.

        The other side of the picture here is that many small towns in the south that boomed in the 1980’s and 1990’s or the Midwest small towns that boomed in the 1950’s and 1960’s are no longer booming (and many are even shrinking) — and those locals are eagerly looking for someone to blame. They are often easily persuaded to blame things on political opponents and on other minority groups they perceived as favored than on the societal implications of corporate economics or robotic technology.

  11. Very fascinating! Some observations:

    The original line that was actually built doesn’t share a single station with this blue line south of downtown, but does share 3 stations with bus connectors to the blue line. (speaking of which, it would seem silly if the Rainier Beach bus didn’t at least interline with the airport bus)

    The West Seattle busway is quite close to the future West Seattle segment, especially considering that initially the WS line will only operate to Stadium station. I think a busway really would have been a better choice for WS considering that buses can be as fast as trains on dedicated ROW, and we already knew that travelers from WS will come from many different directions (STB has talked at length about how there is no one obvious path for Link in WS) and indeed Link is really going to be kind of a stub line in ST3.

    The most important thing here I think is the Bart-like trains that can go 75 mph, versus our slow 50 mph trains (at best), that easily get passed by cars on I-5 off-peak. The slow trains are I think an even bigger mistake for a long-distance Link system than the at-grade MLK section.

    Splitting the green line makes sense, and with the kinds of demand on the Ballard line in the future as well as limited throughput on MLK, I think this makes sense if ST is able to connect the tunnels correctly. Which they won’t bother to do. Aw, well.

  12. I’ve always liked the idea of splitting East Link so one line goes to Eastgate as shown in this plan map. Of course, Mercer Slough impacts are an issue — as is possibly the effect of having two trains going in opposite directions on the Lake Washington bridge is.

    1. All along there has been the possibility of an Issaquah-Seattle line. What finally pushed it to Issaquah-Bellevue was capacity in the DSTT and perhaps on the I-90 bridge, and the growth and jobs on the Eastside that increasingly demands intra-Eastside travel. But there’s still the possibility of an Issaquah-Seattle line in the future; e.g., if it merges with East Link at South Bellevue, if the I-90 limitation is unfounded, and if capacity in DSTT2 is assigned to it. I don’t see any of these as likely in the next several decades, but it’s still theoretically possible if demand emerges for it. However, the Eastside will probably prioritize north-south transit higher, Bellevue-Lynnwood and Bellevue-Renton — or even that dubious UW-Kirkland crossing (at 520 or Sand Point).

    2. What are these headway limitations on the I-90 bridge? I’ve heard chatter and speculation in the comments, but I don’t think STB has done a proper post about this (Which is surprising given the huge implications).

      Is it just that we don’t yet (and won’t until after it has been operating for a while) know whether the I-90 floating bridge can safely accommodate two opposite-direction four-car trains passing each other?

    3. I don’t know whether the limitations are true or not, thus my hedging. One story is that only one train can be on the bridge at a time, but I think that was debunked. I assume the limitation is comparable to Rainier Valley (6-minute frequency each direction), but that’s just an assumption.

      We did debunk two myths about the DSTT and U-Link tunnel that had been circulating on STB for years. Originally there was a planned ventilation shaft in Montlake where the Hop-In mini-mart is, but that land was owned by Kemper Freeman who was trying to block East Link so ST decided not to condemn it and eliminated the ventilation shaft. The myth was that the lack of the ventilation shaft limited downtown-Northgate trains to 3 minutes, and that the DSTT was also limited to 3 minutes. But Martin later talked to ST and discovered that when it eliminated the ventilation shaft it split the signal block into two, so that allows trains to run up to 90 seconds apart. And the DSTT can run that too although it would require capital improvements, which were considered for ST3 but not included in the package. Although ST cautioned that going below three minutes could lead to less reliability because of the logistics of coordinating so many trains.

      1. Interesting. Makes sense that the passing trains limitation was probably debunked, since that seemed to be a design priority of the project. They also simulated two passing trains during the 2003 nighttime load testing of I-90.

        I’m not sure why there would be a headway limitation unless it’s to specifically prevent more than one trains being on the bridge in the same direction. If that was not prevented, then you could potentially have four 4-car trains on the I-90 bridge at once, which sounds scary. A sinking of the I-90 bridge with four trains on it would make Amtrak 501 look like a fender-bender.

        If that is the case, then the possibility of trains running at full speed is good news for frequency. Hopefully block signals will be placed as close to the stable edge as possible so as to allow the most throughput. Hopefully they will also be collecting bridge movement data over the course of years to determine feasibility of bridge upgrades. They’d be fools not to, really.

    4. I don’t like splitting East Link before the line gets to Bellevue. Take makes the headways in Bellevue and Redmond too low. Splitting at Wilburton, to send trains to Kirkland and Redmond, is more intriguing.

      It makes the mistake of assuming East Link is just a commuter route to get people to into Seattle. It’s also a way to move people around within East King. As an Issaquah resident, I’d be just as keen on a 1-seat ride to Bellevue and Redmond as I would a 1-seat ride to Seattle.

  13. This map highlights the concept of branching rail lines. There have been other posts that have considered doing this. ST seems to prefer to define their operations to not doing this, except for the double line between Downtown and Lynnwood and for a small distance in Bellevue.

    Do you prefer to ride a train –waiting for up to 6 minutes — and always have to transfer (and wait for a second train after walking to another platform, perhaps on a different level), or would you prefer having to wait for a train up to 12 minutes but not have to make a train-to-train transfer?

    I’ll note that ST has never publicly shown the impacts of these two different approaches in any of the ST3 planning.

    1. 10 minute maximum until 10pm! That’s the threshold where travel stops being spontaneous and starts to require planning, and some trips become unfeasible. In New York, London, Moscow, or St Petersburg you never wait more than 2-5 minutes for a train in the daytime. In St Petersburg it rises to 10 minutes after 8:30pm. New York reaches 20 minutes only after midnight. That’s the kind of transit that makes every trip convenient and maximizes one-car and zero-car households. Link has a good minimum standard: 10 minutes until 10pm, 15 minutes until 2am.

      In contrast, MAX and BART are 15 minutes on each branch, and depend on doubling up to reach 5-7.5 minutes, but that means you have to live in a limited area to get ultra-frequent transit and it’s often just not feasible to live there. MAX’s tripled segment is along the Banfield Freeway with limited walksheds, and if we look at Metro’s pre-2016 bus routes, ultra-frequent service on Capitol Hill and Fremont only got to Bellevue Ave and 34th Street, the barest edge of the neighborhoods. And both MAX and BART have dropped to 30 minute Sundays during various budget-cut periods.

      So only branch if you can run each branch at 10 minutes minimum.

      ST has not been clear about ST2’s frequency. It may remain 10 minutes on each line like I recommend, or it may drop to 12 minutes.

  14. Nice, Oran – thanks!
    That was one of the first elections in which I voted after turning 21 the year before. Took 40 years of waiting for me to get on a train in Seattle.

  15. My general thoughts are:

    a) For those who wanted “urban villages” connected, this was your plan.

    b) Happy ST2 & ST3 will build most of this map out. Not happy our kids will need a ST4 to connect to the Renton factory.

    c) Worth noting this map didn’t have a Ballard to UofW connection. Something if we got this system, I’m sure we would have built and now our kids will with ST4.

    d) Great graphics work Oran!

  16. Oran: what software did you use to make the map?

    I’d love to be able to make some transit maps of my own!

  17. Uh….2020. Some very, very dishonest person in MS-13 told me the wrong year. And MD was born in Texas too. Evidence of which is that Texas has declared war on Coal. Next question, No Comment.

Comments are closed.