The Sound Transit Board on Thursday received several illustrative scenarios for prioritizing ST3 projects through the realignment required this summer because of tax shortfalls and cost overruns. The scenarios bookend the range of possibilities with delays on individual projects ranging from 1 to 14 years.

Among the scenarios is a phasing approach which would keep delays on the highest priority projects within the range of 2 to 4 years. Due to subarea constraints on Board action, some variation of the phasing approach seems the most likely to emerge from the realignment process.

‘Additional Capacity’ options.

In a lengthy presentation, staff also reviewed options for additional revenue that might reduce these delays. This would lean most heavily on increased federal support, including aid for transit that is making its way through the House in the current COVID relief bill. While the political environment in DC has shifted in favor of support for transit, and the baseline financial scenario has already been cushioned with $3 billion of additional expected grants, Sound Transit wants to go further with another $2-$8 billion in aid over the life of the program. Similarly, Sound Transit hopes to extract $4 billion from the state legislature, mostly through direct appropriations.

Locally, there are three options in play. The Board could enact a rental car tax, though the $70 million impact is very small. With majority voter approval in an RTA ballot, a $24 annual head tax could raise $685 million through 2041. A more material option would be an expansion of Sound Transit’s debt capacity. That adds $1-3 billion of capacity, but that requires 60% voter approval.

Except for some federal aid, this is all fairly speculative and will not be much clearer this summer than it is today. The Board will need to make realignment decisions not knowing if any of these options will play out favorably, though it will be tempting to postpone announcing project delays by assuming the best.

Illustrative scenarios.

The illustrative scenarios should not be taken very literally. Sound Transit is required to consider all the goals of the ST2 and ST3 plans, and also to conform to subarea equity. The illustrative scenarios do neither. They benchmark what alternative capital plans would look like if the prioritization were driven by just one consideration.

The ridership scenario prioritizes projects with the largest total ridership. Weirdly, there’s no cost performance metric here, so small projects that may perform well on a rider per dollar basis fall to the bottom. The ridership metric would have delays of no more than two years on the downtown Seattle tunnel and Ballard extension, but would see the infill stations and parking projects delayed up to 11 years if no additional funding beyond the current forecast is forthcoming.

The equity scenario very nearly flips this. Infill stations perform well (as measured by the income and race mix in the one-mile station circle). Some South King and Pierce projects also perform better on this metric, with Ballard and most Eastside projects underperforming. The range of delays to projects on this metric vary from about two to ten years.

The spine scenario needs no introduction. Extensions to Tacoma Dome and Everett would see two years of delay on the Everett-Tacoma spine. The second downtown Seattle tunnel would also open within two years of its original schedule. Every other project would see nine years of delay.

The last and most complex scenario combines tenure and opportunities for phasing. The tenure criteria would give the highest priority to projects promised in ST2, and phasing would break apart most of the large ST3 projects to deliver some nearly on-time station openings in every geography.

Some of the phasing suggestions are more intuitive than others. On the Eastside, truncating Issaquah rail at Eastgate in a first phase would deliver the more valuable intra-Bellevue connections early while deferring the costly lower-ridership miles of rail along I-90. A Delridge terminus in West Seattle would serve critical connecting bus lines. An interim terminus at Smith Cove for the Ballard line seems underwhelming, but it recognizes that most of the ‘Ballard line’ ridership is in South Lake Union and downtown. Snohomish County will prefer a first phase of Everett rail extend to Paine Field rather than stop at Mariner though the latter is a promising bus terminus for commuters to Seattle. In the south, it’s not clear what advantage Fife has over Federal Way as an interim terminus, though the lower cost options for the South OMF-E require a 1.5 mile rail extension beyond Federal Way anyway. Local representatives may make different choices as they juggle Tacoma Dome Link against Sounder improvements.

Nancy Backus may have spoken for several Board members when she asked if each subarea could directly identify their priorities and have Sound Transit develop that scenario. The Board is visibly struggling to come up with a shared vision better than having each subarea move forward within their own particular financial constraints.

Given the large uncertainties about future revenues and costs, the phasing approach offers a politically palatable way forward. Sound Transit can make firm commitments to a limited set of extensions in every area with modest delays (some of which are already baked in as COVID has slowed environmental review by a year or more). Those can be supported with high confidence that they are affordable on a schedule close to the ST3 plan.

The secondary phases might face long delays if the current financial projections play out over the next two decades, but the delays may be reduced if enough funding is added or if cost savings are realized on earlier projects. As the Board grapples with the huge uncertainties in the out-years, there is no particular urgency for the Board to set firm dates on projects where work won’t begin for years anyway.

108 Replies to “Sound Transit eyes phasing of ST3 projects”

  1. What are the differences in tax costs of these various options? The board acting oblivious to the public’s interests in having those tax rollbacks occur as soon as possible. The board wants to do cost/benefit analysis without considering tax costs to which they would subject people. That is sociopathic behavior.

    1. What are the differences in tax pizza costs of these various options?

      There, fixed it for you.

      But seriously, will Link continue to serve all the pizza parlors planned, with the concomitant threats to the Region’s youth? Much depends on The Plan. Trust The Plan……

  2. If there is any change in the 253 than it should be in the form of cancelling light rail from Hilltop to TCC. BRT2 can replace that.

  3. As a firm believer in the second tunnel as by far the most valuable ST3 component, I’m relieved that in every scenario this is either in the first bucket or (in the case of equity) second-priority, but behind very small projects that take very little fiscal capacity.

    1. The second tunnel includes two parts:

      SoDo to Westlake — There is no value added here. Every station is next to an existing station. It is being built to enable the other light rail lines — essentially it is a means to an end, not something that adds value by itself (similar to the OMF).

      Westlake to Elliot — There will be three new stations. The first will be close to Denny, and similar in nature to Pioneer Square, but with smaller buildings. It is definitely justified, but overlaps the catchment area of Westlake. The second is designed to interact with Aurora buses (like the 5 and the E). It isn’t clear how many people will transfer here. Those headed downtown will likely just stay on the bus. So it essentially becomes a connection from Aurora to Lower Queen Anne, as well as Beacon Hill/Rainier Valley/SeaTac and the southern suburbs. It will also serve some people in the area, but it is probably about as bad a location as is humanly possible in that regard. It abuts the giant Aurora basket of snakes that not only take up a huge amount of space, but make it difficult (and miserable) to walk to the station (e. g. It also abuts the Bill and Melinda Gate Foundation office park, that is more park than office. Those four square blocks are likely the lowest employment (or residential) density in the area. The combined six square blocks (between Mercer, Aurora, Harrison, and 5th Avenue North) have almost nothing, yet are within the 1/4 mile walk radius (or would be, if the street grid wasn’t torn up to make driving easier). Throw in the parking lot and other landscaping around the highway, and walk-up ridership to the station will be much lower than just about any potential South Lake Union station. The last station is in Lower Queen Anne, which is a solid station — similar to Capitol Hill (but without the same level of connections).

      Overall, those three stations look good, but the cost is extraordinary for them. Sound Transit’s “High End Estimates” put the ridership at 22,000, which sounds about right. I would round up, to 25,000 (by 2040, if things go well). That is a lot of money for that many riders.

      The justification for the new tunnel comes from the Ballard line. Without the Ballard line, it is a bad value. Not the worst value in ST3, but well below most of the little projects (BRT, station infill, bus improvements, etc.).

      1. You’d be right about the south end of the tunnel, Ross, were it not for the nearly 150 feet of vertical distance between Fifth and Madison and Third and Columbia at the north entrance to PSS. There are a LOT of BIG buildings at the crest of the hill around the corner of Fifth and Madison.

        Midtown allows people on its line and those using connections from the other two always to walk downhill to and from a destination between Fifth and Third. Sure, it requires a connection one way or the other, but it’s a genuine improvement for many people. I expect that the new tunnel will achieve its highest ridership between Westlake and IDS, not north of Westlake, for just that reason. That makes the ease of transferring at those two stations essential to make the best use of the investment.

        Unfortunately, it looks like ST will build the Westlake platform under the block between Sixth and Seventh which will necessitate a one block walk assuming the best case of direct access from New Westlake’s fare-paid Mezzanine area to the east ends of the platforms in existing Westlake.

        Ditto at SoDo, but it’s harder there because the platforms are parallel. The New IDS mezzanine would have to be one level deeper than the existing IDS platforms to avoid a down-over-and-up movement between it and the existing southbound IDS platform. Unfortunately, every meter deeper the Mezzanine of new IDS lies (and therefore the platform is another meter the trackbed has to rise to reach an efficient level at Midtown.

      2. So the urban subway that serves the denser half of any Metro 8 subway is “eh, whatever,” but becomes awesome only once it crosses the ship canal? I’ll remember that next time someone argues for a standalone Metro 8 … “but it doesn’t go to Ballard, so what’s the point…?”

        I think the SLU station is designed to serve SLU, and it’s placed near Aurora mostly because that’s the logical stop spacing, and if you are going to be near Aurora you might as well be right there because Aurora+Dexter is a major bus corridor. Your Maps link exaggerates the issue – the 99 interchange is only in the way if you are north of Mercer between 6th and Dexter. Approaching the station from the other 3 directions, or simply being west of 6 or east of Dexter, and the interchange is irrelevant to the walking experience.

      3. Sorry, I don’t buy it. Consider three scenarios:

        1) You are coming from the south, and now your train is sent into the new tunnel. Are you better off? No. Instead of two stations, you have one. Even if that one station is somehow the ideal location of any potential downtown station, it can’t make up for the fact that it largely overlaps the other two stations. The two stations, combined, simply have a much larger catchment area. It is worth mentioning that as good as Fifth and Madison is, it is a block away from the freeway. That means a good chunk of any walk radius has absolutely nothing on it (although eventually it might have an apartment or two).

        2) You are on the other line, and consider switching, to get closer to your destination. My guess is only a handful of people will do this. The new tunnel will run every 6 minutes at best. Who knows what the stations will be like (how long it takes to go from one platform to the other, how deep the Madison Station is, etc.). But my guess is the vast majority of people won’t bother. If you are headed to Columbia Tower, for example, (a big building) it is a four minute, largely level walk up to Fourth. You would save one minute of walking by using the Fifth and Madison stop — time you would likely lose making the transfer (and then some). Not to mention the fact that people don’t like making transfers. Consider this scenario: You are headed to Fifth Avenue (say Marion). You are on the 7, headed north. It is 4:00 PM, so there are several express buses headed up Fifth. Do you hop out there, and catch one of them, so you can get right to your destination? Of course not. I guarantee you hardly anyone does that. They are going to ride their bus until it gets as close as possible. There just aren’t going to be many that will transfer.

        3) You are outside Westlake Station, and are headed to Fifth and Madison. Most people would probably walk. The rest will just catch a bus on Third, knowing that the bus will be a lot more frequent, and they avoid going up and down into the station. The same is true for the other downtown stations. You have to go as far as Stadium or SoDo where taking the train makes sense, and even then, most people would take the first one to appear (which is likely to be the main line).

        It all adds up to no new riders. Some riders would save time, but my guess, more would lose it. More riders from the south will have to walk further, or transfer to that other train . Again, the only reasonable justification for the new line — as currently designed — is if you need to increase capacity.

      4. Aach. I meant to close the bold tag after the word “two”. So basically one word being bolded, not most of the comment. {fixed}

      5. So the urban subway that serves the denser half of any Metro 8 subway is “eh, whatever,” but becomes awesome only once it crosses the ship canal?

        Nope, I’m sorry you are confused. This is not the Metro 8. There are some significant differences, such as:

        1) This requires six stations to get three. That is expensive. The same is true with the tunneling. A lot of extra work to basically replicate what is already done.

        2) The geography is different. This does not connect to Capitol Hill, it connects to Westlake. It isn’t just a matter of stations, it is a matter of where the train is actually going. As I wrote, the Aurora intercept station largely just makes sense one direction (west). In contrast, a Metro 8 subway would connect those riders to Capitol Hill and the Central Area.

        3) Station placement is better with a Metro 8 subway. I would cross at Thomas, and have a station at Fairview, Westlake and 9th (AKA Aurora). This means better spacing, especially compared to the main line. With a stop at Denny, a lot of riders won’t bother making a transfer from the main line. Put a stop at Westlake and Thomas, and they will.

        The biggest issue is the first one. The stations aren’t terrible, but the cost is. Imagine if Northgate Link required building another version of U-Link, but still without a First Hill station. You would have three stations (Westlake, UW, and a Capitol Hill Station on the other side of the park). Suddenly Northgate Link becomes a very poor value, which is the case here.

      6. You can’t just wave away overcrowding in DSTT1 without DSTT2. We still don’t know whether Lynnwood and Federal Way will cause overcrowding. And the second tunnel is predicated not just on Everett and Tacoma, but the total of all north-south circulation downtown including intra-downtown (Westlake to Intl Dist). A PSRC report said downtown north-south circulation would exceed all transit capacity (Link+surface buses) in a couple decades. Since then Seattle has done some things (split RapidRide C/D, planned other 3rd Avenue RapidRides, planned the CCC), so those add some capacity. Whether it’s enough is iffy.

        ST says DSTT1’s maximum frequency is currently 3 minutes. It would require capital improvements to go beyond that to 1.5 minutes; otherwise there would be increased risk of unreliability. The ST3 list of candidate projects had a project to do those capital improvements, but it was deselected when DSTT2 was selected.

        Finally, people from Everett, Lynnwood, and Tacoma aren’t on Link or in the DSTT now. They’re on express buses. With ST2 they’ll be transferring to Link at Federal Way and Lynnwood and in the DSTT. With ST3 they’ll be taking a train directly from Tacoma Dome and Everett to the DSTT. (And some will be going through the DSTT to get to UW or the airport or Northgate transfers.) So most of the people who will be on Link in the Everett and Tacoma Dome phases will be on Link in ST2. There will be some ridership increase when it’s a one-seat train ride from Tacoma Dome and Everett, but I don’t think it will increase that dramatically compared to people transferring at Federal Way and Lynnwood. So DSTT2 is either needed before “Everett and Tacoma” open, or it’s not needed at all. (And I prefer to play it safe and build it in case it’s needed, and to provide redundancy in case DSTT1 breaks, and to give spare capacity for future lines or unexpected population increases. Pugetopolis and the US have spent seventy years underbuilding transit. Overbuilding it would be a nice problem to have for once.)

      7. “There will be some ridership increase when it’s a one-seat train ride from Tacoma Dome and Everett, but I don’t think it will increase that dramatically compared to people transferring at Federal Way and Lynnwood. So DSTT2 is either needed before “Everett and Tacoma” open, or it’s not needed at all. ”

        I don’t think there will be much induced demand in & through downtown with the ST3 extensions; most of the induced demand should be <30 minute trips, i.e within the subareas or trips like Alderwood to Northgate or Tacoma to SeaTac; someone trying to go from Lynnwood to Tacoma is very likely going to do that trip with or without a transfer.

        Therefore, as you say with a nod towards the PRSC projections, the need for the 2nd tunnel is driven by the growth of Seattle & the region more broadly, with or without suburban rail extensions. If we do add million plus people and downtown Seattle continues to be the 800 lbs job gorilla, we will need the 2nd tunnel whether or not Link gets beyond the ST2 termini.

      8. You can’t just wave away overcrowding in DSTT1 without DSTT2.

        Yes you can. This is a common misconception about DSTT2. It isn’t needed for capacity, exactly. It is needed because otherwise, capacity for the core of our system (UW to downtown) would go *down* if we interlined the trains at Westlake. That is because frequency between UW and downtown would be worse.

        Our most crowded section is going to be between Westlake and Capitol Hill. I will repeat that, because it is crucial to understanding the situation. Our most crowded section will be between Westlake and Capitol Hill. Westlake is our most popular station. At the system grows out (and becomes more balanced) you will have more riders heading to the north than the south. Consider a northbound train. In the evening rush hour, it will add more riders as it goes through downtown. There will be people getting off, of course, but more getting on. At Capitol Hill, the tide begins to turn. By the time it hits Roosevelt, it has turned. More riders are getting off than on. This continues all the way to Lynnwood (and beyond). Furthermore, the new Ballard Line makes things worse. Right now, someone from South Lake Union or Lower Queen Anne might take the 8 to Capitol Hill, and then transfer to Link there (to go north). Now, they would be tempted to take the train south to Westlake, and join the hordes there.

        An extra line through downtown doesn’t change the situation. Unless, of course you interlined the train from Ballard. If you did that, then you couldn’t run the trains as frequently. That is the issue.

        It is an issue that goes away if you don’t share the existing tunnel. That can be achieved several ways. You can build a second tunnel, you can end the Ballard line at Westlake or just not build it. Ending at Westlake would be complicated logistically (you have to connect the two lines somehow, as well as find a place to dig out the tunnel boring machine). It is also less than ideal for riders (being asked to transfer). That is why they decided to go with a new tunnel.

        My point in all of this is that the shorter the new tunnel, the worse value it is. The redundant tunnel is essentially a fixed cost. It is the same no matter how much you build up on either end. But if you only build as far as Elliot Avenue (and SoDo on the other end) then you get much worse ridership per dollar spent. You are still paying for that redundant tunnel, and yet you aren’t getting as many riders.

        A PSRC report said downtown north-south circulation would exceed all transit capacity (Link+surface buses) in a couple decades.

        Add that to the long list of miscalculations made by the PSRC. I would like to read the report, though, if you have a link. Maybe they made a few assumptions (no new bus routes) that could easily be fixed.

        In a few years, a lot of the buses that crowd downtown will go away. All the buses from Community and Sound Transit will be gone. Express buses like the 41, 71, 77, 301, 303 and 304 will be gone as well. There are improvements that could be made to 3rd, while that leaves 1st, 2nd, 4th and 5th will very few (if any) buses. That is an enormous amount of transit capacity — far more than simply adding one train every six minutes.

        Speaking of which, if we really wanted to, we could do that with the main line. We could run trains every 90 seconds (an extra two buses every six minutes, instead of one). The biggest issue is where to send them (Stadium?). That, and it “wouldn’t give our ridership as reliable a service” ( Fair enough, although that really shouldn’t be our main focus; San Fransisco, Boston (or New York for that matter) are not spending money trying to ensure that the trains are super reliable. You might have a bit of a delay — but that is a small price to pay.

        Again, that seems highly unlikely — if we need extra capacity it will be between Capitol Hill and downtown (and we’ll have to live with whatever reliability issues that involves). For service within downtown you just run more buses.

      9. RossB, why do you expect bus service on 5th Avenue? If ST3 is complete, perhaps there will be very frequent Seattle oriented service on 1st and 3rd avenues and no routes on 2nd, 4th, 5th, and 6th avenues. SDOT can place PBL on 2nd and 4th avenues.

      10. “Maybe they made a few assumptions (no new bus routes) that could easily be fixed.”

        That was the point of the report, to show what would happen if no additional transit were added. The area of the study was Westlake to Intl Dist (and maybe further north into Belltown/Denny Triangle), not Westlake to UW. The people who are currently on endangered north-south buses don’t disappear, they move to Link. It’s about the north-south circulation in the downtown area as a whole, which includes both Link and the express and local buses. The argument is that both Link and the buses would be overcrowded so there’d be no transit available at some times.

      11. Ross, it would be very difficult to interline using the existing tunnel at Westlake. You probably could do it with some very fancy tunneling using an over-under tunnel from a block or so north of Denny and Westlake over to Minor and south to the old Convention Center Station. The upper level could break through the support wall to join the southbound track at the curve from the southbound tube to the ramp down to platform level. Then, if the space between tracks is wide enough, the northbound would diverge just east of the platforms at Westlake to a new track between the existing ones that would begin ramping downward immediately while the existing tracks rise up the ramp to the tunnel curve.

        That’s it. That’s the only even slightly conceivable way. It would be extremely difficult, if possible at all, to place a northbound divergence between University Street/Symphony and the curve into Westlake.

        In the first place, I very much doubt there is enough distance for the northbound diverging track to drop quickly enough to underrun the Westlake Station box. which has a south wall somewhere very close to the foundations of the buildings on the south side of Pine. The blocks are 240 feet long, and the diverging track can’t start diving until after the frog of the turnout from the main line. If you put that frog right under Union Street you’d have 460 feet to drop about 35 feet in order to underrun the station box safely. That’s about a 7% grade, which is entirely within the operating limits of LRV’s, but much greater than any gradient ST has countenanced in Link heretofore.

        It would be a roller coaster ride.

        But the second thing is that you’d have to dig up Third Avenue and create another “station box” just north of Symphony for the northbound divergence. That box would have to be deeper than the existing tubes in order to accommodate the diving track, at least on the east side of the trackway. We can hope to have a wall between the two directions to make the excavation narrower, but they might have cross tunnels which would complicate that. In any case, construction would be quite the shock to the Third and Pike intersection, though it could certainly be timbered over for temporary use by the buses above.

        The diverging track would have to be accommodated within the envelope of Third Avenue to avoid the basement of the buildings on the east side of the street. Since the tubes are far enough apart for the “bus bypass” in the stations, it would be a tight fit.

        It would probably be fine to break through the north wall of the extended box at the curve for the southbound track convergence, but d.p. notwithstanding, Sound Transit is not going to put a level crossing in the middle of its most heavily used trunk line.

        So, probably no direct interlining to a tunnel under Third at Westlake, but possibly, depending on the clearance between the tracks east of the Westlake platform, interlining via the rather tortured “detour” above.

        However, as I’ve written recently I think it’s possible to do without the second tunnel at all if downtown growth has clearly flattened by say 2024 or 2025 and shows no likelihood of reigniting. In that case ST3 for North King should be shifted to focus on the capital improvements to the existing tunnel and main stem to get to 2 minute headways. I doubt that 1.5 is truly achievable with the relatively poor internal circulation of LRV’s with the narrow platforms in the DSTT but two should be achievable.

        Those improvements would consist of:

        1) revamping ventilation within the tunnel, especially between CHS and HSS, and shortening the signal blocks throughout the main stem between SoDo and Northgate.
        2) rebuild the tail track at Northgate to become a two-track stub with scissors for reversing. That can be done by removing it after Lynnwood comes on line and rebuilding it above the revenue tracks, or by building a new revenue track just east or west of the revenue tracks and repurposing the track bypassed to become one of the turnback pair.
        3) overpass Holgate and close Lower Royal Brougham at the track crossing. Buses could still access Ryerson from the busway and buses to the Greyhound station would come from Sixth South.
        4) over- or under-pass the four busiest streets crossing Martin Luther King Jr. Blvd and make the others right on and right off only.

        These improvements would allow trains to operate at a faster speed than the cars along King Boulevard and at full speed along the busway.

        Then turn the trains from Tacoma at the improved Northgate tail track for a total transit time between TDS and Northgate within two minutes of that between TDS and Ballard.

        As money savers don’t do West Seattle; instead, turn the Everett trains using the turnback loop at Forest Street and make bus improvements for them.

      12. Regarding Capacity:

        1) This is unnecessary, according to Sound Transit. The ventilation shafts are fine. They signaling is fine. Here are a couple quotes from
        Sound Transit’s signaling system is designed for a minimum 90 second headways.

        I specifically asked about the deleted Montlake vent shaft, often cited in STB comment threads as a constraint on tunnel throughput. Ms. Olson denied that it was one; in fact, when ST deleted the shaft they placed a signal at the midpoint between Capitol Hill and UW. This allows two trains in the three-minute segment between the stations, which means it has the same fundamental 90-second limit as the rest of the signaling system

        It is really a question of where to turn back at the north (Northgate sounds good), where to turn back to the south (SoDo seems appropriate) and whether it is OK that we “wouldn’t give our ridership as reliable a service.”

        If they managed to run the trains every two minutes and if they manged to interline (two big ifs) it would still be messy. Here is how I would do it: Pair Ballard Link with East Link (which makes the most sense geographically — as it is essentially East Link to West Link). Send one North Link train to SeaTac, the other to SoDo. It would be like this every two minutes:

        North Link to South Link
        North Link to SoDo
        Ballard Link to East Link
        North Link to South Link, etc.

        That could be done without improving either the East Link or South Link sections. If we could get those headways down to 4 minutes, then it becomes a lot easier — just alternate the trains every 2 minutes like so:

        North Link to South Link
        Ballard Link to East Link

        That assumes that 4 minute headways is enough, which sounds quite reasonable. These are four car trains, which can carry a lot of people, even during rush hour.

      13. Ross, I like your idea to run East Link into Ballard Link if the only thing built were the over-under tunnel and it went all the way to Ballard. Otherwise I guess it would be Redmond to Smith Cove.

        I like it because there is much more likelihood that folks from the East Side would be heading to SLU than those from the South End. They’d prefer the University connection I believe, and it would be very popular in Northeast Seattle because of the airport service.

        The Redmond-Ballard running time would be the shortest of the three operating segments. It would probably be in demand for operators.

        So the four operating segments would be TDS to Northgate, SoDo to Lynnwood, SoDo to Everett and Redmond to Ballard.

        This could absolutely be built within the existing cost projections if the “ramp-down-in-the-middle-of-of-the-ramp-up” is possible.

        Thanks for the refinement.

    2. I’m somewhat skeptical that the 2nd tunnel is really needed. It doesn’t add anything to the walkshed, so it’s only value is capacity. Surely, there is something ST can do to allow trains to run closer together in the existing tunnel that’s cheaper than the $several billion it costs to build a whole new tunnel.

      1. The second tunnel (or whatever profile it takes) is needed between Westlake and LQA. The southern part may not be needed except the tracks need to tie in with the system somewhere to enable maintenance of the vehicles.

        Since we are already talking billions more, it seems logical to evaluate what it would take to have branch tracks north of Westlake. Why spend an extra $5B without seeing what else could be done? Ignoring alternatives to a $5B cost increase (not the project cost but merely the cost increase) is fiscally irresponsible.

        Have you ever been to the Green Line platforms in Park Street Station in Boston? The directional platforms there have tracks on either side, then merge together once out of the station. That means two trains can be closer together between stations because the frequency constraint is not the tracks between the stations but the time a train must sit with its doors open at a station.

      2. On top of this, isn’t it the approach to the 2nd tunnel that requires the second SoDo and ID stations? I suspect nobody would argue that those two stations add anything to the walk shed! If anything, they may make the situation worse because poor transfer experience and construction impacts in the case of ID. Of course, if the second tunnel were going to First Hill, and/or if the second SODO station were on 1st Ave., (“actual stadium station”) it might make up for this:(.

        THAT SAID, these decisions have been made, and the second tunnel has become essential to complete “the spine” as currently envisioned, because capacity. And quite possibly, even the portion of “the spine” from Lynnwood to Federal Way. (Too bad transfers and station access are not necessarily designed with high capacity in mind, but I digress…)

      3. The second tunnel has a Madison station and serves the library, which DSTT1 skips over. University Street Station should have been a couple blocks further south or there should have been an in-between station. (And if Westlake Station had an underground passage to the west side of 4th & Pike where the eastbound bus stop is, it could serve the northern part of the University Street station walkshed.) DSTT2 finally serves Madison, the biggest arterial between Pike and Yesler. That’s not nothing, and would be a loss if it didn’t happen.

      4. One possibility is to do the following:

        Build the system from Ballard (or Elliot) to Westlake as planned. Then build the tunnel from there to SoDo, but without any additional stations. Essentially every other station becomes like Graham Street (it will be added later). This saves the cost of building the Madison Station (likely very expensive) as well as the additional platforms at I. D. and SoDo.

        You would have winners and losers. Those at the south end get their one seat ride to the UW (they come out ahead). People who transfer from the north (e. g. Northgate) at Westlake wouldn’t even notice. Folks from the northwest (e. g. Lower Queen Anne) would have to transfer. If the train ended at Smith Cove, it would be about the same for Magnolia and Ballard riders. Either way, the main reason they would transfer to that train is to get to Lower Queen Anne and South Lake Union. If they are headed to south downtown, they will stay on the bus (either way). If they are headed to the airport (or similar destinations) then they just stay on the bus longer (and transfer at Westlake or University).

        The main people it would hurt are those closest to Westlake. The Denny stop, for example, would hardly be worth using. I suppose that is true for folks wanting to transfer (from the north) but this just makes it worse. For Lower Queen Anne it becomes similar to the monorail (not just for those riders from the north, but for those riders in south downtown). It is obviously not a great system, but it would be a lot cheaper.

        Ridership would definitely take a hit, but ridership per dollar spent might actually go up. If ST is seriously considering half-ass solutions (like ending the train at Smith Cove, or Delridge) than this should definitely be on the list.

        The best temporary solution, of course, is a rail-convertible bus tunnel. This would be more expensive that what I proposed above, but far more useful. It would be cheaper than a full build out, yet have as many, if not more riders.

      5. Al, read my long post a few above yours. There may be a way to branch just east of the Westlake platforms on that long cut-and-cover incline up to the tunnel portals.

        I did not mention it, but given that the track would have to loop over to Minor and this is an urban stub, it seems that there could be a station somewhere near Minor and Denny, serving the part of SLU that the Convention Center station used to serve, though better because it would be several blocks to the north.

        The Denny and Westlake station would have to be a block farther north, though, which is a bit of a drawback. Ross might approve, though, because it would be farther from Westlake.

      6. The Denny and Westlake station would have to be a block farther north, though, which is a bit of a drawback. Ross might approve, though, because it would be farther from Westlake.

        I would definitely prefer it. Westlake is close to Denny Park (two and a half square blocks of nothingness) and closer to the Aurora stop. Terry or Boren is much better. The east side of Boren is where the biggest residential building in Seattle will soon be built ( In general you just want to spread out the stops as much as possible, maximizing coverage. If you could add yet one more stop (as you suggested) then it would be great, but it would add to the cost. If you could pull it off (the interlining along with the extra stop) then my guess is you would have a line with a lot more riders, but not necessarily cheaper. Interlining without the extra station, though, should be cheaper (you save money by not building the Madison station).

    3. Martin, if the new tunnel is to go only to Smith Cove, not a bad idea since the stretch from Mercer Place to the north end of Third Avenue is pretty gnarly using either route, to get the best usage from it there will have to be extra shuttle trains between SoDo and Smith Cove. ST needs to understand that they can’t have a CBD tunnel with six-minute headways and get much value from the Midtown station. There aren’t enough people from the south end that work in the financial core. It will get most of its ridership from transfers at Westlake and IDS.

      1. Midtown will also be the primary transfer to Madison BRT. However skewed the “finance district” is toward north/east commuters, First Hill should be skewed towards southend commuters?

        I think transfers within downtown to get to another downtown stop will be small; ST went for the Midtown station at 5th or 6th because it wanted both tunnel to serve the full CBD directly. What you are envisioning would have been much more likely if Midtown was further east, but that’s not the system we are building.

      2. Midtown will also be the primary transfer to Madison BRT.

        Only if their train goes there anyway. No one is going to transfer from a train to a train to get on to the Madison BRT. They will walk a couple blocks, or take a different bus (e. g. the 60 or streetcar if you are coming from the north). You would get people from the south, but only because the train is rerouted there. This is not a good thing. They gain the quicker connection to Madison BRT as well as SLU and Lower Queen Anne, but lose the one seat connection to Capitol Hill, the UW and Northgate. The latter are much bigger destinations.

        The big connection for Madison BRT are buses like the 1, 2, 3, 4, 5, 24, 26, 28, 33, 40, 62 and E. If Link goes to Ballard than some of these go away. If it goes to Smith Cove, they all remain, and all become a better bet for most of those riders in getting to Madison.

        I think transfers within downtown to get to another downtown stop will be small; ST went for the Midtown station at 5th or 6th because it wanted both tunnel to serve the full CBD directly. What you are envisioning would have been much more likely if Midtown was further east, but that’s not the system we are building.


      3. I meant specifically the south end, as that was Tom’s comment.

        SLU + Denny + First Hill seems a bigger combo that Cap Hil + UW + whatever

      4. They won’t transfer for two short downtown stations when DSTT1’s Seneca entrance is one block from a RapidRide G station (and you can ride it westbound through the turnaround at 1st to go east, like with the 10 and 11 on Pike/Pine). But at least one line will stop at Madison and the library, and that’s better than no lines.

      5. AJ, you’re right that it’s likely that quite a few of the folks who work on First Hill come from the South End. So your point that it’s a good transfer to Madison BRT is a plus. Thanks for pointing that out.

        But, as I wrote elsewhere on this thread, if downtown really does go flat as Daniel and Q prophesy, then no second tunnel of any kind, bus or rail, is needed. Simply turn the South End trains at an upgraded Northgate turnback. It will be within two minutes of the time to Ballard from Westlake.

      6. SLU + Denny + First Hill seems a bigger combo that Cap Hil + UW + whatever

        No it doesn’t. Capitol Hill is bigger than Lower Queen Anne (although not by a lot). Denny is extremely close to Westlake. Likewise the difference between getting off at Seneca and taking the bus versus getting off at Madison and taking the bus is minimal. Like Denny, there are other ways to get there (e. g. from Beacon Hill I would take the 60).

        That is not true with the UW. Link is pretty much it if you are downtown or anyplace south (yeah, you can get there via the buses, but that is dramatically slower). Meanwhile, UW is huge. Even with that very poorly placed station, and access from only one direction, it is our second most popular stop. For a southbound train, it has the most boardings (11,000). That is more than University, Pioneer Square and I. D. combined. You have bus connections there as well (Kirkland, Wallingford). That also doesn’t count Northgate, with the campus and the major medical center (and whatever they decide to build at the old mall). Sending the train over to Smith Cove will be a degradation for those on South Link. They’ll get over it, but it is still worse.

    4. I agree the second transit tunnel would appear to be the key piece to ST 3 for N. King Co., certainly if it could handle buses. Plus all five subareas are paying for it. The risk for cost overruns — above the current estimate of closer to $3.5 billion — is a worry for such a deep tunnel, so responsibility for cost overruns among subareas will be key because some subareas will have limited additional revenue for cost overruns, and can’t rely on a HB 1304 levy.

      Seattle will argue it paid a disproportionate amount for the spine based on subareas created decades ago, which is true, and the four other subareas will argue: 1. There is adequate capacity for the spine in one tunnel; 2. They didn’t get tunnels; and 3. You can’t get blood out of a turnip, which are also true.

      1. I think Tom Terrific simplifies what at least I have said.

        For some time what I have said is ST 3 costs were always underestimated in the N. King Co. subarea in order to pass ST 3 in the four other subareas, and any decline in revenue — let alone a pandemic that may permanently affect work patterns — would expose the cost estimates, because ST was planning on Seattle continuing its amazing growth, despite a terrible city council and last five mayors, and so revenue would come in higher than estimated.

        Next when ST admitted the $4 billion cost underestimates for West Seattle and Ballard I said I was not surprised, but the real issue was loss of general fund tax revenue for decades if work patterns change, and the true cost of the second transit tunnel that all five subareas have to pay for, and some claim they can’t pay more.

        When ST revised its estimates for underfunding ST 3 to $11.5 billion I said that still does not count the tunnel, and since $11.5 billion is about the total for ST 3 in Seattle it means what I also suspected: ST is using ST 3 revenue to complete ST 2. Ok. That is an important bargaining point, because the one subarea that does not benefit from the spine is N. King Co. (and east King Co.).

        Finally, I tried to tell Seattle Subway HB1304 was written and designed to complete ST 3 in N. King Co., except the amounts are too great for a local levy, although ST now wants $9 billion. From where?

        Someone above said Seattle should have gone with its own transit levy, except with subarea equity it did. The real problem is when the subareas were first drawn Seattle had all the revenue, and so it was unfairly burdened with completing the spine to Snohomish Co. and S. King Co. so it can reach Tacoma. Who in Seattle wants to reach Everett or Tacoma, especially on a slow train?

        There is no benefit for Seattle to have transit to Everett or Lynnwood or S. King Co. or Tacoma, and as you move out that far you lose population and ridership density, and car culture takes over.

        I also wish some would better understand population vs. ridership density, and I think Ross understands this the best.

        There is a very small area of population density in this region for rail, although that does not equal ridership density. If Seattle had funded its own levy it would have run rail from Northgate to UW to Capitol Hill to downtown to First Hill and maybe Beacon Hill. Even West Seattle and Ballard. But not to Lynnwood or Federal Way. Not because of population density though, but because of ridership density.

        There just isn’t the population density for rail outside this area of quasi-density I outline above. The idea you can create population density through TOD, or millions will move to the area and live along the spine as the PSRC claims, is crazy, even pre-pandemic. Snohomish Co., S. King Co. and Pierce Co. should have paid much more of the spine because it benefits them, not Seattle.

        Transit, especially rail, follows density, both population and ridership, it does not lead, or create its own ridership density. Claiming The Spring Dist. will support ST’s crazy ridership projections for ST 3 decades from now is just not realistic.

        East King Co. decided to run rail to Seattle, and pay for all of it, because being connected to downtown Seattle (and not Ballard, Tacoma or Everett) benefits the eastside (depending on Seattle’s politics and economic vitality). Although East Link will be a nice tourist run, like the ferry to and from Bainbridge, and covers lovely scenery, no one lives along it, and it doesn’t even access downtown Bellevue.

        But, it does connect to downtown Seattle directly, and the route was relatively efficient because it uses existing rights of way with almost no tunnels, (although running rail to Redmond was political, not rational, and driven by Microsoft). So East Link was a rational economic decision for the eastside, not Seattle. East Link will get half the ridership estimated, but the eastside paid for it, not Seattle, and it runs directly to downtown Seattle. So good project overall, for both cities.

        Finally you have ridership density, which is not the same as population density. Again ridership density is basically the route in Seattle I describe above, because Seattle is the huge jobs market, its roads are congested, it has a huge university, and rail works really well with commuters during rush hour (depending on first/last mile access). But to claim Seattle has population density for 90% of its neighborhoods just is not accurate.

        I have to shake my head when I see ST, the Board, and some on this blog promising they will complete the “spine” Seattle built, but will (and did in ST 1) gut rail in the areas in Seattle that make sense. Ok, maybe Ballard and West Seattle are a little extravagant as designed, but compared to running rail to Lynnwood or Tacoma they make great sense, certainly for Seattle.

        Right now I think ST needs to focus on the second tunnel, because without a second tunnel new rail lines don’t work. Seattle has to go to the cheap subareas — Pierce, Snohomish and S. King Co.’s — and tell them everyone has to chip in on the true cost of the tunnel because you should have paid for most of the spine, and then Seattle can figure out the other ST 3 projects in its subareas it wants to fund.

        I doubt East King Co. will balk because it has the money, doesn’t know where to spend it, and if I were at the S. Bellevue station I would much rather be able to take rail to West Seattle or Ballard than Issaquah, because you know I am driving to Issaquah, because you need a car to get around Issaquah, and eastsiders like to drive on the eastside. There is zero chance I am taking rail from S. Bellevue to Tacoma or Everett.

        If Seattle truly knew how to play hardball it would stop work on the “spine” and claim it doesn’t have the money and the other counties will have to complete it. Then use that chip to start the second tunnel that hopefully can be done for $3.5 billion, and then Seattle can figure out the rest of its ST 3 projects.

        The spine was a stupid idea, but we can’t let that defeat rail in the areas it makes sense, and that will need a second tunnel, although just the thought of the risk and construction is frightening.

        For God’s sakes stop making the spine the priority when it does not benefit N. King Co. at all.

  4. Here’s to hoping the infill stations get fully canceled as opposed to merely delayed up to eleven years. We need to be looking at a faster system with fewer stops per mile, more like BART. I know this view is incredibly unpopular here, yet it is the one that supports the RTID as a whole, and as such the direction we should take. The less Seattle centric ST is, the better. Metro Transit is already Seattle centric. It doesn’t need a second transit agency doubling up on that focus

    1. Here’s the thing, some of the infill stations make sense to add, like 130th because they serve a beneficial purpose to transit and connecting people to other modes of transit or places people want to go. In the case of 130th it provides a good transit node for access to transit in the North End of Seattle. Connecting neighborhoods like Bitter Lake, Haller Lake, Broadview, Pinehurst, North End of Lake City and major arterials like Aurora Ave or Lake City Way.

      1. Postponing 130th St. by 11 years is just ST shooting itself in the foot. In the long run, it costs far more money to have the construction crews come out a second time, disrupting a running train line, that to just do it right the first time.

        Even if it means borrowing a little bit more and paying a little bit more interest, ST should just do it. It will pay for itself.

        The other infill stations are different of course, so postponing them is more justified.

      2. The other infill stations are different of course, so postponing them is more justified.

        Except that Graham Street Station is likely the most cost effective station (in ridership per dollar spent) of any of the projects. I don’t mean that it would be in the top tier, I mean it would be the best. Number One. There are other things to consider, but it performs well on those as well. Building the most popular pieces first means that you have better fare recovery, and thus lower costs in the long run. It means that you can justify more frequent service, which means you get higher ridership, and even better fare recovery.

        Delaying Graham Street Station would be Sound Transit shooting itself in the foot, but with a different gun. It still hurts, and it is still stupid.

      3. I lived in Lake City while Mosqueda fought for the 130th stop. I voted for her despite that. 130th makes no sense for Lake City residents. 130th doesn’t connect to Lake City Way. 125th does, as does 145th. The latter would be more useful as a North End connector.

      4. 130th at I-5 blends into 125th straight to Lake City Way so I’m not sure what you’re talking about.

      5. You would get great ridership out of 130th if you would build a gondola from Bitter Lake to 130th to Lake City. For the same reason a pedestrian bridge is getting built at Northgate station.

      6. I think A Joy is trolling, or maybe a bot. The statements are contrarian, and nonsensical. There is no evidence of critical thinking, analysis, or reading comprehension. This latest comment is a good example. How could you live in Lake City and not understand that 125th becomes 130th when it crosses the freeway? Even if you have never ventured that far, wouldn’t you look at a map before posting something like that? You just assume that the city council member is wrong about her geography. You assume that previous comments — on this exact thread — mentioning Bitter Lake and Lake City are wrong. You never even bother to ask yourself “Where is Bitter Lake anyway?” or “Where does 130th even go?”. Because if you did, you would realize that Bitter Lake is due west of Lake City, with the connecting road being 130th/125th. I mean, just look at how you get from the actual lake to the center of Lake City: You go right by where a station would be.

        It is just a weird statement for a human to make.

    2. You might be the only person out there who thinks BART is a model worth emulating. The vast majority of ridership occurs in the SF-Oakland-Berkeley core. The rest of the system, as fast as it is, is a very poor value, and the low ridership bears that out. Link is already headed in that direction, but 130th and Graham St are absolutely higher value than most stations in the suburbs. Deleting them would be a conscious choice to make Link suck more.

      1. Yep. But some people prefer to ignore facts, and just imagine that BART performs better than the Paris Metro. Of course you don’t need to leave the continent to see why BART was a bad design, you can just compare it to DC Metro:

        This is not really a debate in transit circles. You will be hard pressed to find any transit expert, anywhere, saying that BART did the right thing. It is like vaccines — the medical establishment is not torn over the issue — get your shot (when you can).

      2. Pre-pandemic, BART recovered 75-80 of its operational costs with distance-based fares. That’s hardly a total failure. BART clearly has done some things right.

      3. No one said BART is a total failure. We are saying that the qualities that make it BART-ish are what they did wrong. Specifically the long stop spacing and the lack of lines in the urban core (roughly defined as San Fransisco, Oakland and Berkeley). This is why it performs so poorly compared to other, similar systems (like DC Metro). I can explain in more detail, but again, this is not something that is disputed amongst transit experts, even ones (like Alon Levy) that laud suburban systems (like the S-Bahn). The previous link gave a good explanation.

      4. The primary problem with BART is land use, not alignment or stop spacing. BART+Caltrain+Muni gets comparable ridership to the DC Metro; the fact BART, Muni subway, and Caltrain are completely different technologies and agencies has nothing to do with alignment or stop spacing.

        If ST has great land use around its stations – in Seattle & elsewhere – it will have excellent, all-day ridership.

      5. It seems important to point out that BART built the Muni Metro tunnel even though VART doesn’t run through those stations. Before BART, the J, K, L, M and N Muni trains ran on the surface east of Twin Peaks and Duboce Park. This means that BART built Castro, Church and Van Ness stations.

        Recent BART extensions were funded by local county sales taxes that San Francisco didn’t collect. Alameda, Contra Costa, San Mateo and Santa Clara voters separately provided the local match money for the various projects. San Francisco funded rail extensions were instead used to expand the Muni Metro rail network to Mission Bay, along Third Street and soon across the core as the Central Subway.

      6. The primary problem with BART is land use, not alignment or stop spacing.

        Bullshit. Come on, man, you are in denial. Again, no transit expert, anywhere, believes that. They are all in agreement. BART was a mistake. They should have had better stop spacing, and more lines in the urban core — like D. C. Metro. There is not a single transit expert anywhere who sings the praises of BART, and says “if only this, or only that”, because it simply isn’t true. Because it is weird. It is the first of its kind.

        It doesn’t matter how fast the train is. It doesn’t matter if you build up around the distant stations (and they have). You simply aren’t going to get decent ridership *from those suburban stations* unless you have a very strong urban system. San Fransisco doesn’t. Oakland doesn’t. They have extremely slow buses and trains that still manage to carry more riders than BART! It is a textbook example of what *not* to do. Keep in mind, this wasn’t cheap. This wasn’t the S-Bahn (which has much better urban spacing by the way) that takes advantage of existing rail in the suburbs, and simply allows those riders to transfer easily. It cost a bundle to go very far out into the suburbs, and pick up relatively few riders. Again, just nine stations between San Francisco and Berkeley account for half the ridership of BART.

        I notice you didn’t write “The primary problem with BART is land use, not alignment or stop spacing like such and such”, because there is “such and such”. There is no system, like BART, that has ever had the success of a common, everyday, normal subway system. None.

        If ST has great land use around its stations – in Seattle & elsewhere – it will have excellent, all-day ridership.

        Again, bullshit. ST’s rideship will be driven by Seattle, just like every other transit system in the world — including BART! You just can’t get away from the fact that not that many people want to ride a train for forty minutes, then get on a bus that takes them another 20 minutes to finally get to their destination. Ridership of Sound Transit is entirely dependent on the quality of service *within Seattle* (and to a lesser extent Bellevue) not how deep it goes into the suburbs, or how built up it is around the suburban stations.

      7. BART+Caltrain+Muni gets comparable ridership to the DC Metro

        OK, let’s compare them:
        DC Metro — 816,000
        BART — 420,000
        Muni (rail) — 157,000
        Caltrain — 67,000

        Hmmm, OK. Yeah, let’s throw some other systems in there, while we are at it:

        Muni (bus) — 500,000
        AC Transit (East Bay) Bus — 215,500
        WMATA (DC bus) — 340,100
        Ride On (DC Suburban Bus) — 68,500

        MARC Train — 30,000

        A few things jump out at me:

        1) Commuter rail doesn’t account for much. This is the case, everywhere. The only place where commuter rail is large is where the inner city transit system is much larger. The commuter rail network in New York carries around a million people. The bus system carries over 2 million (the subway over 9). San Fransisco is no different, despite having arguably the greatest commuter rail service in North America. The suburban part of BART has less than 200,000 riders. The Muni bus system (operating in San Fransisco) has 500,000. Caltrain is a rounding error.

        2) In DC, people take the Metro. In the Bay Area, they take the bus. This isn’t necessarily bad. If you know me, you know I like buses. They are an underappreciated part of every cities transit system (even Paris has great bus service). But there is one important fact worth considering: San Fransisco has one of the slowest bus systems in the world. To quote

        With a fleet average speed of 8.1 mph (13.0 km/h), it is the slowest major urban transit system in America and one of the most expensive to operate, costing $19.21 per mile per bus and $24.37 per mile per train.

        Yes, people use it. People use Muni more than they use BART and Caltrain combined. They use the bus part of Muni more than the other two combined. They use the bus part of Muni more than twice as often as the suburban part of BART.

        But it sucks. It is godawful slow. People use Muni because they have no other choice. It is slow, expensive, yet carries the bulk of the riders. Subway service within the urban core is underfunded, and yet huge amounts of money were spent on BART — especially on parts that carry very few people. That is my point. If you spend money on things that aren’t important (long extensions into the suburbs) instead of things that are (lots of lines and stops in the urban core) then you end up with system that isn’t as good. You can claim that it doesn’t matter — that you can spend money on anything (whoopee!) — but it does.

      8. Wiki says 816k?

        Ross, if you can hand wave away all political and financial constraints facing transit construction in the Bay Area but take land use as a given, then I think I’m entitled to hand wave away land use constraints but acknowledge political and financial constraints facing transit alignments.

        I’m not saying BART is good, but simply that is is underrated today because Bay Area’s catastrophic land use issues significantly limit BART’s potential. Would SF be better off if they could figure out how to build Muni tunnels at a cost & speed closer to international norms, or if they had the political power to just take the suburbs’ money and spend it on urban transit? Yeah, for sure.

      9. My guess is Wikipedia numbers are based on peak values. DC Metro has had major maintenance issues that predated the pandemic, and really cut it into ridership.

        I’m not saying BART is good

        Looks like we are on the same page then. Good. End of discussion :)

        but simply that is is underrated today because Bay Area’s catastrophic land use issues significantly limit BART’s potential.

        First of all, I don’t think the land issues are as bad as you say. There is plenty of development around various stations in the suburbs. Second — don’t you know the land use going in? Why build a line to a distant suburb, hoping there are big developments around the station, when there are huge developments that already exist, that are closer? In fact, there is density even on the way! Look at a density map of Oakland ( Now consider the fact it is 2.7 miles(!) between Lake Merritt and Fruitvale. You would think that there is just nothing there — maybe a swamp or a lake. You would be wrong. There is density between there comparable to that at Lake Merritt and Fruitvale, they just didn’t want to add a station. Those two stations, by the way, don’t have huge ridership — and this is why. BART only works for a handful of trips in east Oakland, despite all of the density. They just take the bus.

        Imagine, for example, if Link just had one station in Rainier Valley at say Columbia City. Getting there would be faster, as would getting to SeaTac, and other places in the suburbs. And ridership *at every single station* would go down. Speed is nice. Being able to walk to the station is much nicer.

        Here is an interesting essay about American mass transit ( in a section about local and express trains Alon Levy writes:

        In any discussion of BART extensions, people bring up the fact that BART can’t skip stops – never mind that its stop spacing is extremely wide owing to its function as suburban rail. The average speed on BART is 57 km/h per the National Transit Database; the RER A, which is the express service here, averages around 50. At BART’s speed, the single longest express segment in New York not crossing water, the A/D between 125th and 59th Streets, would take 7 minutes; in fact it takes about 9. If anything, BART errs in having too few stations in Oakland and San Francisco.

        You will notice that at no point does Alon mention the lack of development around BART stations. It isn’t an issue. But the issue of inadequate stop spacing is called out.

        One of the things that often comes up with BART is the comparison with the S-Bahn. Here is a relevant essay I ran across (same author): One of the big misunderstandings about the S-Bahn is that it is regional rail. It isn’t. To quote the author “Service to suburbs that are not directly adjacent to Berlin the way Potsdam is provided by hourly RegionalBahn trains, which do not form a neat network of a frequent north-south and a frequent east-west line through city center”.

        I wish Alon had listed the stations within that map, just to give it even more perspective. Here is a map (not to scale): You can see how many stations are within the doghead shape ring (the Ringbahn). Many of the lines are part of the U-Bahn, but the Ringbahn itself (clearly visible on Alon’s Map) is part of the S-Bahn. From the tip of the nose to the back of the head (the widest part) it is about ten miles. Turn the ring sideways, and it fits just fine between Northgate and Columbia City. When the S-Bahn does venture out of the ring (the “city center” as Wikipedia calls it) various lines go to the equivalent of Renton or Shoreline. But even when they do that, they make lots and lots of stops. From Oranienburg to the Ringbahn, there are 17 stations, or more than one every mile (nothing like BART, even in the urban part of BART).

        There is really no case for doing what BART did. It is very difficult to have too many stations per mile (Paris has the most, and is hugely successful). But you can do the opposite, and BART did. They spent a fortune building a regional system and neglected to actually build a mass transit system that justifies that kind of expense or frequency. They clearly don’t have enough stops, nor do we.

    3. A BART-del Norte aficionado, then? Should we also make it 5’6″ gauge so we can pay twice as much for railcars forever?

    4. A Joy: if intercity service is what you want, how about freeway based express bus routes: Tacoma Dome, Federal Way TC, Westlake via Spring-Seneca.; three Lynnwood lines: Everett, Everett Boeing,, and Marysville. Just the opposite, Link should be more urban and less BART-like.

      The ST Board should consider cost-effectiveness or rider minute per dollar. The infill stations would look good. NE 130th Street should be considered in tandem with frequent Metro service between Bitterlake and Lake City via the station.

      1. The whole point is getting mass transit users out of freeway traffic, as well as moving them quickly and efficiently. Link fails to move mass transit users today, with the current number of stops. Infill would only make that worse. Freeway based express bus routes being needed, or even an idea, shows the failure of Link overall.

        Link takes longer to get from downtown to Seatac Airport than a peak hour 174 in traffic. That simply should not be an acceptable thing. Yet as it stands today, not only is it acceptable, but transit “experts” and STB aficionados want to add more stops between them and force Link to take even longer. Light rail should be extended to Olympia, even without Thurston County being part of the RTID, before we start shooting the system in the foot with infill stations.

        BART is very imperfect, admittedly. But the concept of extending light rail farther rather than making it an overglorified local subway is not one of those imperfections. But here you’ll hear people pushing for luxury stub lines like UW->Ballard before even completing the spine to Tacoma and Everett.

        Link isn’t just for Seattle. That was the deal made when the RTID was formed. We’re decades into construction now. Reneging on that deal is a bad faith non-starter.

      2. At least BART can run at 80 mph. Link peaks out at 55 mph. Link is significantly less strategic for long distances than BART is because BART is 45 percent faster.

      3. Link’s 55 mph speed limit is an artificial limitation. It might require different cars or power stations, but the rail itself is still capable of faster speeds. Unless there’s a state or federal law I am unaware of.

      4. Why is there no research to describe how to make Link faster? The extensions north of Northgate and south of Rainier Beach could really benefit from faster trains. The improved travel times should also entice more riders. ST needs to quantify this.

        Then ST would know what requirements are and an idea of the costs involved. It seems like the least ST could do to inform the board on this issue.

        I expect lots of public disgust about train speeds once Link to Tacoma and Everett open (if it ever happens).

      5. Link is limited to 55 mph by ST’s specs. ST designed the initial segment with tighter curves and didn’t require higher-quality train specs that a corridor intended for a 65 or 85 mph upper limit would have. That’s surprising given ST’s intention from the beginning to go to Tacoma and Everett. But it’s just like the thinking that envisioned south Link being surface (35-40 mph) from Intl Dist to SeaTac and presumably Federal Way and Tacoma. Did anyone think to calculate the travel time? No, they couldn’t have, or they’d realize that was non-viable for a 35-mile line that’s competing with a 65 mph freeway and 89 mph Sounder..

      6. The board in its pre-ST3 deliberations did mention maybe seeing what it could do to retrofit the initial segment for higher speeds, presumably in the Rainier Beach-SeaTac segment. I haven’t heard anything more about that, or what it would take. I gather it would involve making the curves less sharp and strengthening the guideway, which sounds like it would require rebuilding it.

      7. A Joy: in your response, you intended Route 194, not Route 174. In the I-5 of 2019, pre Covid, I-5 was jammed and slower.

      8. No, I intended the 174. It crawled from Pioneer Square through Sodo, not really picking up speed until it reached the back side of Boeing Field. Not that it is relevant. Link is slower than the 174 or 194 were at peak.

      9. The whole point is getting mass transit users out of freeway traffic, as well as moving them quickly and efficiently.

        That is what bus lanes are for. That is what commuter rail is for. That is not what a mass transit system is for. Again, no transit expert in the world would agree with you. None. Zero. Zilch.

        It really shouldn’t matter why, but I’ll go ahead and explain it as best I can. First off, you get way more riders in the city. It doesn’t matter what you do in the distant suburbs — you can build an outstanding system to the distant suburbs — a lot of those people are going to drive. BART is a great example — those trains are blazing fast, and yet almost all the ridership is within a handful of stations in the city. Second, travel on the freeway is very fast, while travel in the city is not. This means that a subway system in the city competes much better with driving than a subway system in the suburbs. Even suburban riders take advantage of this. This is explained quite well in this excellent comment that envisioned the DC Metro as designed by Sound Transit. The key quote is this:

        One might think Baltimore and Manassas would at least be happy with this outcome, as all-day 10-to-15-minute service provides the ability to flee these profoundly struggling and profoundly boring (respectively) cities on a moment’s notice. But perhaps not. The likelihood that any given Baltimorean or Manassahol can travel all the way to D.C. without planning ahead is vanishingly slim. The likelihood that all their needs, once there, will be within easy reach of the “spine” is even slimmer. And after nearly an hour spent on the train, will they really choose to fight the last three miles of their journey on a run-down and unreliable bus? More likely, they’ll begrudgingly get in their cars and fight the traffic apocalypse – growing ever worse because the real DC Metro was never built – all the way to the parts of the District they actually wish to reach.

        The point is, if you start taking away stops, then the system loses value. Take away Mountlake Terrace, for example, and riders in Lynnwood have a faster trip to downtown Seattle, but Mountlake Terrace riders end up driving. You also lose the handful of people who would go from Lynnwood to Mountlake Terrace. You never gain ridership by taking away stops, you only lose them. This means more people driving, whether on the freeway or otherwise. Is that really what you want?

      10. Link takes longer to get from downtown to Seatac Airport than a peak hour 174 in traffic.

        Now you seem to be arguing against yourself. Link to SeaTac *is* an express. There is only one stop between Rainier Beach and SeaTac — a distance of over 7 miles! Even the stops in Rainier Valley are extremely widely spaced. You are essentially arguing against light rail to the distant suburbs, because no matter how few stops you have, you can’t possible beat the speed of a good express bus.

        Either that, or you are arguing for something absurd. Keep in mind, SeaTac Station, despite being an express, only gets about 5,000 boardings a day. Many of those are likely to be to Rainier Valley and Beacon Hill. It is worth noting that between Beacon Hill and Tukwila there are 4,500 *southbound* riders. Obviously a lot of them are going to SeaTac. If it just ran to the airport (along the freeway the whole way) you would *lose* riders to SeaTac, not gain them. Ridership overall would be thrashed.

        You are espousing some sort of weird system where we spend billions of dollars so that a handful of suburban riders get a fast ride into a handful of place, while the vast majority of people — in *both* the city and the suburbs get crap. Sorry, no thanks. You need to stop taking a freeway centric view of the world. Transit doesn’t work that way.

      11. I’ve ridden the 174 between downtown and SeaTac. Link is faster. Especially considering faster times loading passengers downtown and at the airport. The 174 and 194 took several minutes just waiting for passengers at the airport to get on with all their luggage.

        In fact, by the time you consider the extra loading time, plus traffic in the airport pick up/drop-off lane, Link starts to approach the travel times of the 194. If you’re staying on Link to Capitol Hill or UW, you save a transfer, and your trip becomes a lot faster.

        Also, recall that prior to Link, the 194 stopped running on Sundays around 5 PM, so anytime you travel somewhere for a weekend, by the time you land, get off the plane, and get your bags, there would likely be no 194 to take. With Link, I can book flights that land as late as 10 or 11 PM on a Sunday and know that I can get home with minimal waiting.

      12. A Joy: Route 194 did serve the DSTT, SODO busway, and I-5. Route 174 use East Marginal Way South as Route 124 does today. Both extended to Federal Way. You also missed another acronym: RTID was the regional transportation investment district to build highways; it was defeated in 2007.

      13. I am well aware of the 174 and 194 routing. Nothing you have said here disputes my words. The 174 and modern 124 routes are not identical though. The new 124 routing causes the bus to make a slow and odd jog through Georgetown.

        I should have been using the acronym RTA instead of RTID. You got me there. Mea culpa.

      14. The 194 was 10 minutes faster than Link to the airport without traffic. But you never knew when traffic would delay the bus and make you miss your flight. Link adds Rainier Valley, Beacon Hill, and TIB for just a 10-minute penalty. No bus could do that. There was never a bus between Rainier Valley and the airport before Link. Link’s pre-covid schedule was every 8 minutes peak, 10 minutes off-peak every day, and 15 minutes between 9pm and midnight. The 194 was every 15 minutes daytime, 30 minutes until 9:30pm, not at all after that, and I don’t remember weekends, but asdf2 said it ended at 5pm Sundays.

        The 174 was half-hourly and took longer, and everyone had to take it when the 194 wasn’t running. I remember several times flying back in the mid or late evening and having to take the 174. The last time I took it it stopped at every single stop along the way except the Boeing complex, which was frustrating. And that was late evening, not at 4pm.

        So that 10 minutes extra travel time yields a lot of benefits, and is canceled out when I-5 bogs down or stops completely for a collision. And then people would really miss their flights.

        Having said that, the travel-time penalty wouldn’t exist if ST had designed Link better. Grade separation in Rainier Valley and SODO would have speeded it up. Light rail is capable of 85 mph performance if it’s mandated in the track and train specs. ST could have compromised with 65 mph to match the freeway.

      15. “Infill would only make that worse.”

        Each stop adds 20 seconds. All three add up to one minute. Only people traveling between TIB and 145th would experience all three in one trip. That one minute is dwarfed by the train-to-train transfer time at Intl Dist, which could be 5-12 minutes in a worst-case scenario. It’s OK to add a few infill stations, we just don’t want dozens.

      16. “The whole point is getting mass transit users out of freeway traffic, as well as moving them quickly and efficiently.”

        That’s part of a larger general theory. Mass transit should be attractive enough to be the first choice for over 50% of the population, and it should have enough coverage and performance to be viable for the largest cross-section of people and trips. The post office provides universal service, and private operators provide additional choices. The same with a mixed healthcare system like Canada, libraries vs bookstores, or municipal Internet vs private Internet in some cities. The public service should be robust enough that average people use it, not just the poor.

        Every transit service has a sweet spot: a distance range where it’s most competitive, and beyond that it’s increasingly less competitive. For 55 mph light rail with stops every 1-2 miles outside downtown, the sweet spot is trips 5-15 miles between village centers, and high-volume destinations like downtown, stadiums, malls, and transport hubs. North Link is competitive all the way to Everett (starting from Westlake), while south Link is competitive to Seatac (due to surface segments and a detour). It becomes increasingly less competitive beyond that. At the same time, the wide stop spacing and small number of lines means it doesn’t serve many in-between trips and parallel trips, so buses are needed for those.

        The south end problem is primarily that its cities are further away. Lynnwood is the distance of north Kent, and Everett would be just south of Federal Way. South Link’s surface segments and detour add to the travel time. And North Seattle has UW and bustling urban villages that attract short Snohomish trips. The south end doesn’t have anything comparable, and what it does have is mostly further east in the Kent Valley and Renton.

        Since the point of mass transit is to move a lot of people efficiently and conveniently, it should be grade separated or exclusive lane regardless of whether it’s light rail, express bus, or arterial bus. The longer the distance the more separation is needed to avoid travel times over 30-60 minutes.

    5. A Joy, I actually think you have an important point that gets lost in agency biases of others. There is some value in a 80 mph rail line that is 40 miles long and stops only every 2-7 miles except in the central core where the stops are about 1/3 to 1/2 mile apart.

      The RATP in Paris, CrossRail in London, GO Transit upgrades in Toronto and Caltrain’s CalMod program (that evolved from the Baby Bullet operations) are a recognition that all-day long-distance frequent rail has a place in a large urban area.
      The ST Board approach has evolved into believing that light rail is best for every kind of trip unless it’s just too expensive. It’s terribly ironic given the early successes of ST Express and Sounder.

      Lots of people here on STB like to hate on BART — but the challenge of connecting across a wide bay created a unique transit line design and governmental challenge. And let’s be clear that pre-pandemic, BART recovered most of its cost in fares and it is very hard to find an urban bus system that had achieved that in the US. Finally, SF skyscrapers would not have been possible without BART bringing in employees.

      1. LINK isn’t an 80 mph service though. For the types of trips you are talking about, what is needed is better quality Sounder service. Link will wind up being a degradation of service over the express buses for those types of trips.

        Or, in the case of Germany, they serve that market with the equivalent of Cascades trains.

  5. The ridership scenario prioritizes projects with the largest total ridership. Weirdly, there’s no cost performance metric here, so small projects that may perform well on a rider per dollar basis fall to the bottom.

    Which makes the metric complete BS. It is simply a measurement of how large the project is. There is nothing wrong with noting this, but prioritizing large projects over small projects and then stating the have better ridership performance is a lie.

    To show how absurd it is, imagine if the infill stations are grouped together, and considered one project, like Ballard Link. Suddenly their “ridership performance” improved dramatically.

    No other way to sugarcoat this: This is a Sound Transit lying about the performance of various projects. At best this is incompetence. At worst it is scandalous. Either way, it is shameful.

    1. I see the same thing in the software development space all the time. The big shiny new feature that takes 6 months has more impact than the small paper cut bug that takes 2 days. So, the small papercut bug is always at the bottom of the priority list and never gets fixed.

    2. The entire point behind Link is something that is all the time, so not commuter rail, yet is better than bus lanes. If sheer number of city riders was going to be used as a metric, then Seattle should have gone out and built the system on its own. You take Everett and Tacoma’s money, you prioritize getting to Everett and Tacoma. To do otherwise is called a bait and switch. And that is no way for a public agency to operate.

      Downtown to Seatac was the metric I was using because it is the metric I personally know best. My family has lived there for three generations. I doubt Link would compare any more favorably at any 194 stop, but I’m willing to entertain disagreements.

      You and I have very different definitions of value. That is clear. I am fine with that. I value getting people the furthest distance in the fastest time. You seem to value the most passengers, regardless of distance or time. I would argue such a value system is a much better fit for standard busses than mine. Busses can work for what I perceive to be your values. We know no number of bus lanes or express busses will work for mine.

      I am espousing a system that gives us something we don’t already have. A system effectively promised by where taxing lines were intentionally drawn. I feel you are espousing a system that merely improves on what we already have and have had. Link needs to be a vision of the future, not a vision of today.

      If you think a distance of seven miles denotes an express, you must have hated the 194. Not a single stop from the Busway to the Airport. That to me is an express. The seven mile stretch between Rainier Beach and TIBS? A single stretch in an otherwise very non-express system.

      I was under the impression that Seatac Station’s boarding were on a significant rise recently. I’d also like to see evidence for your “Many of those are likely to be to Rainier Valley and Beacon Hill.” assertion, given the fact that unless you have access to ORCA information you have no way of knowing one way or the other. And at any rate it is philosophical, as number of boardings isn’t The Holy Grail for me.

      I find the accusation that my view is “freeway centric” to be a Strawman Fallacy. My view is spine centric, and to be more specific unbroken spine centric. The former is literally what the RTID and Link were founded upon. I understand if you think that idea is a horrible one, but you lost that argument back in the 1990s. It is far, far too late to back out of it in favor of some “Seattle Uber Alles” approach.

      1. You take Everett and Tacoma’s money, you prioritize getting to Everett and Tacoma.

        You are taking money from Lake City and Bitter Lake too. If we are going to prioritize anyone, it is those folks, who can have a station that would be far more useful to them, and far cheaper to build than one in Tacoma or Everett.

        My family has lived there for three generations.

        What difference does that make? What an odd thing to write. Then again, it wouldn’t take too much effort for a bot to figure out that nativism is a great way to troll.

        I was under the impression that Seatac Station’s boarding were on a significant rise recently.

        They are not. It is one of the few stations where ridership went down (before the pandemic). Look it up.

        I’d also like to see evidence for your “Many of those are likely to be to Rainier Valley and Beacon Hill.” assertion, given the fact that unless you have access to ORCA information you have no way of knowing one way or the other.

        Again, look it up. Sigh. OK, here it is: Look at the boardings for a southbound train. A thousand at Beacon Hill. A thousand at Mount Baker. 600 at Columbia City and Othello. No, Sound Transit doesn’t release actual trip data (stop to stop) but it stands to reason that a substantial number of those riders are headed to SeaTac. Even if they aren’t, it doesn’t matter.

        You are basically saying we should screw over the people in Rainier Valley, so that you can have a slightly faster trip to SeaTac. Screw over the people in north Seattle so that a handful of people in the northern suburbs will have a faster trip.

        Sorry, but I don’t want that. I don’t believe in favoring a small group of riders, while making things worse for everyone else. I believe in building a transit system that will provide the most benefit to the greatest number of people. I guess you are right — we do have different philosophies.

  6. In the universe of government-funded projects, it must be a very rare win-win for the option with the greatest equity impact to also be the cheapest. Unfortunately, as the author noted, Sound Transit planners aren’t considering “cost per” in any of these scenarios, which frames the conversation in a way that de-emphasizes an in-fill station that is both eminently affordable AND improves access to the communities who have borne the impacts of at-grade rail. The people of Central Rainier Valley should be outraged, but by now they are inured to broken promises. What’s 14 years of additional delay when they’ve been waiting since ST1? I suppose the much less efficient, though comfy, solution of last-mile demand-response is some consolation.

    1. Agreed. It is quite likely that Graham Street Station is the most cost effective project in terms of ridership per dollar spent. When this analysis was published, it was second: (scroll down and you can see a couple tables). Since then, the cost of Ballard Link (number one at the time) has skyrocketed, while the estimates for Graham Street have gone up. Thus instead of being at the top — not the top tier mind you, but the number one project when it comes to ridership — it is considered one of the lowest in terms of performance. It is not just absurd, it is scandalous. At best Sound Transit is incompetent. At worst, they are corrupt. This presentation makes Donald Trump seem honest. It is a lie, and should not be tolerated.

    2. Doesn’t Graham Street require widening Martin Luther King Blvd for the length of the platform(s)? When Central Link was constructed ST did not widen the median between the tracks there for a center platform which would require the least space. So that means that there will have to be two side platforms.

      To the west McDonalds has a fairly large parking lot; some of that can probably be taken without killing the business. The U-Haul self storage building also has a large parking lot around it so is probably not required. But to the east, the Starbucks is quite close to the street and the businesses to the south directly front on it.

      It seems that the platforms would have to extend south to Angel Place.

      Now maybe all of this auto-oriented low-rise would be quickly replaced by new mid-rise development like at Othello, but the roadway is going to have to be moved over one lane in both directions, and that won’t be cheap or easy.

      This is just another example of ST’s blinkered refusal to consider potential future expansions. It’s a very amateurish agency.

      1. I don’t believe that a Graham St design has been decided. Yes, the space for the platform should have been reserved. Even on this latest discussion, it seems reasonable to buy the needed right of way even if the station is pushed beyond 2040. Phasing can be defined by project development steps (rather than by only interim end points). Since the station is “voter approved” in ST3, it seems more than legally consistent with ST3 to buy the land for the station even if the station has to be deferred. In fact, not buying the land by 2030 could be called “inconsistent with ST3”.

        As to the station design, it could widely vary. The easiest is to site the station north of Graham with side platforms and take the fronts of a few buildings (maybe taking away the southbound left turn pocket at Graham for part of a side platform). Another would be to move one track (and the adjacent street) to enable a center platform so only right if way only on one side of the street would be needed. More complex options like a grade separated aerial or subway station could even be considered (requiring at least a few years of single tracking during construction).

        Regardless if the money is there in a timely manner or not, ST should at least determine what should be built at Graham Street by 2024.

      2. Al, thanks for respondin. I don’t think that putting the platforms north of Graham is a good idea unless the built environment mandates it. [That is, there is a big new building built just south of Graham which comes right to the ROW line, making any widening impossible.

        I say that because Graham is closer to A.aska than it is to Othello, so Edmunds is almost in easy walk of Graham. The platforms extend to Edmunds. Putting them south of Graham spaces them more evenly.

        But I wouldn’t die on that hill.

      3. Uhhhh…. Graham St is clearly closer to Myrtle (Othello) than it is to Edmunds (Columbia City).

        Maybe you’ve confused Graham and Orcas?

    3. I agree with this point. Total ridership without looking at a denominator is manipulative and useless. Riders per hundred million? Riders per mile? Riders per station? Riders per aggregate travel time saved or GHG reduction?

      It’s a criteria created to favor the big ticket items because they have wider support. It’s a populist choice rather than a strategic one. It actually puts the eventual project more at risk because further cost increases are going to be more sizable.

  7. I believe that the illustrative “phasing” option should include a rethinking of train tracks and vehicle types.

    Take the Issaquah – South Kirkland corridor. It (East Main or South Bellevue) could be served by a partially single track system with two tracks at stations. It couldn’t run at 6 minute intervals but is 6 minutes reasonable? Plus, the project is so far ahead that more funds may be available.

    Even the spine extensions north of Mariner or south of Fife could be done this way.

    Similarly, technology changes could mobilize faster trains at the extremities on closed single-track systems. Operated like EBART, the time to transfer becomes awash if the technology is switched from 55 mph Max speed to 80 mph max speed.

    Finally, I would note that the cost increases on the other extremities are minor compared to the WSBLE project. These kinds of cost savings seem more easily recoverable without stopping stations short of their ultimate ends.

  8. South Kirkland-Eastgate Link before 405 BRT North: I did not see that coming. It’s more important to connect Bellevue to Bothell, Lynnwood, and eastern Kirkland, where the distance is much longer and biking and walking is less feasable than between Bellevue and Bellevue College. The latter can be addressed with an improved 271. Bellevue to Wilburton will have East Link. Wilburton to South Kirkland P&R is so unimportant it shouldn’t wag the dog. It would make more sense if it extended further to downtown Kirkland, but it doesn’t.

    I’m glad DSTT2, Smith Cove, and/or Ballard get priority in some of the measures. There’s something unclear in the projections though. In the exiting schedule, WSJ-SODO was to open in 2030. DSTT2-SLU-Ballard was to open in 2036. So what’s the target timeframe for Smith Cove-Delridge? It combines parts of the two segments. Is Delridge-SODO postponed to match DSTT2-Smith Cove (originally Ballard)? Given that Delridge-SODO requires a bridge (or tunnel), how much does that delay DSTT2-Smith Cove? That’s an important issue because another alternative is Smith Cove-SODO (or Smith Cove-Intl Dist), and then either Delridge followed by WSJ or Delridge+WSJ together.

    I’m also glad ST is considering an interim phase at Mariner, and not making Ballard wait for WSJ.

    Where is PT1 BRT? The tables have SR-162 bit Pacific Ave is SR-7. SR-162 is a small highway from Puyallup to Orting. What is ST planning to do there.

    Putting RapidRide C/D at the end is idiotic. The purpose of those investments is as a stopgap measure until Ballard/West Seattle Link open. If they occur afterward, they’ll miss the greatest need, and there will no longer be any RapidRide C/D lines. Would it then be a contribution to nearby RapidRide lines? Metro Connects has four potential lines that could be beneficiaries: Alki-Burien, the 120, the Fred Meyer to Fred Meyer line (Ballard to Lake City), or the Smith Cove-Magnoli-145th line. There doesn’t seem to be any Metro service between downtown and Ballard, or between Uptown and Ballard, that could receive RapidRide D funds.

    130th should open with Lynnwood Link of course.

    ST can’t just “borrow a bit more” for 130th. The whole reason we’re having this debate is ST will be constrained by its debt ceiling for a few years. Accelerating 130th in the scenarios that have it last would require postponing something else earlier. It would require a 60% voter majority to raise the debt ceiling. It’s inconceivable that voters would have a record-breaking yes vote when many people are less enthusiastic about ST3 and its taxes now than they were in 2016, and when we have a recession that could drag on for years.

    It’s surprising that the gap between the last phase being delayed 14 years vs 2 years is “just” $3 billion dollars.

    1. These examples are illustrative. I don’t think any staff are seriously proposing C/D to be after WSBLE.

      I read 405 BRT north to be the 85th freeway station; the rest of the line can open pretty much as-is when the bus base opens. Also, these numbers are years of delay vs the baseline schedule. I think South Kirkland-Eastgate Link would still open later, even if the magnitude of delay is less that 405 BRT (in that example). I think that also answers your Delridge questions I read the table as Ballard & WS sliding back equally.

      Good call on PT1 BRT … maybe that’s moving forward so it’s out of scope for alignment? The Ortig project is a Sounder spur/feeder, and is only an HCT study.

      1. I don’t think any staff are seriously proposing C/D to be after WSBLE.

        No, staff just came up with BS numbers. It still boggles the mind that they did this. Who was thinking “I know, let’s just look at how many riders each project gets, while ignoring the cost. Then let’s rank the projects in that order, and suggest that the bigger ones should be done first”?

        It is incredible really. It is like saying the Mariners had a really good season last year, because they only lost 33 games. “Lowest in franchise history! Hurrah!”. Uh, yeah, except you only won 27 games. “Oh, details, details”.

        Lies, damned lies, and statistics.

  9. The big cost increase is from the WSBLE project. This cost has increased so far that it needs to be rethought. The other project cost increases including Tacoma Dome pale in comparison.

    There are ways to cut costs. The Smith Cove to Delridge phased option is one. However, I remain skeptical of even this revised estimate as the second tunnel details are pretty much not discussed. I can’t believe that two four-mile tunnels with six subway stations by skyscrapers won’t have upcoming engineering challenges at greatly increased unforeseen costs.

    Shortening the tunnel is the most obvious alternative. In particular, the part around the ID is through awful soil. Multiple options need to be created and evaluated about how this can be more useful. With a design change at SODO station to have cross-platform transfers, the platform adjacency needs at ID become less essential (mainly just an East-South transfer). If the second tunnel begins at Fourth and Washington with the ID platform above Fourth Ave at Jackson, several billion could be saved. A similar but probably more modest cost saving is possible if the LQA – Seattle Center segment is aerial.

    Another alternative is to rethink Midtown station. The maps make it look “convenient” but it’s going to be extremely deep. The elevator and escalator rides are going to be time consuming. If the tracks are moved closer to Pioneer Square and/or University St stations then those mezzanines can tie in the second line — and the Westlake transfer wouldn’t be needed (transfers at University St instead) so that platform could be moved closer to Belltown and Amazon HQ. The Midtown area and First Hill could then be served by a incline (or funicular) to one of the existing stations.

    The bigger point is that — unlike the other projects — this project cost is so increased that the project needs complete rethinking about alternatives. aerial. A much wider alternatives study is now needed — from U-Bahn street trams to use of the I-5 express lanes tunnel to using corridors west of Third Ave. The current EIS needs to be halted and the planning work needs to go back to the drawing board a la 2013-2015. It’s crazy for ST to move ahead with the project as schemed in a backroom in 2015 given how it now has a cost increase higher than the entire cost of other major Link extension projects!

    I know it disappoints many people at the thought of this — but this much additional money can’t be created out of thin air.

    1. Yeah, you are right Al. It is quite likely that a short tunnel would also be very expensive. It would also be crap. I keep pushing for a short bus tunnel, because at least it wouldn’t be crap. You would actually add a lot of value for Ballard, Magnolia and West Seattle, as well as more value for the stations in between (much better frequency, and fewer transfers). But it would still be expensive.

      I’m not sure if there is a good way to rethink things though. Building a bus tunnel that can be converted to rail still makes it an interim project. It still gets us along that path (much as the original bus tunnel did). But building something else (e. g. skipping South Lake Union entirely, and going to Belltown, which is really what they should have planned in the first place) is a different story. Likewise Ballard to UW.

      ST3 was a very poorly designed project. These cost overruns just make it worse. Now the agency is cooking the books as best they can to save the big parts, while throwing away the few bits and pieces that are actually still a good value.

      I very much fear that we’ve blown it. Ten years ago, I was very optimistic about transit in the region. I felt like with a little work, we could be like Vancouver. Northgate Link, and even Lynnwood Link, was really going to happen. Of course they would add the 130th station, that part was obvious.

      A little (real) BRT here and there, a little more rail (Ballard to UW, maybe even Metro 8) and that’s about it. Now, I fell like we are doomed. ST is trying desperately to save some very poorly designed projects, while ignoring things that would actually make a really big difference. Metro can’t seem to recognize a good restructure (like the 61) even when they come up with it. They are reverting to a hub and spoke system from the ’70s, only this time to secondary locations (South Lake Union and First Hill). The 130th station may be 20 years away, while light rail to South Kirkland is fast tracked. Yeah, eventually they will build it all, I guess, but San Fransisco and Oakland are still waiting for the type of system that D. C. built about 50 years ago.

      I feel like 20 years from now we will be nowhere near as good as Vancouver is now, let alone when they build the Broadway/UBC Line. I guess there is always Uber (maybe they’ll have a union by then).

    2. I think that “corridors west of Third Avenue” have been eliminated because of the tangle of tunnels in the southwest corner of downtown. There is some big sewer pipe which isn’t shown on most maps, and the BNSF tunnel is “not that deep” down there, since it is essentially level between its two portals.

      A cut-and-cover stub tunnel ending at Columbia or so might be doable, but it wouldn’t be very popular. Also, the west side is more residential. Those folks just walk to work.

  10. The entire ‘realignment’ process needs to get thrown in the trash. The delays to date are already an existential threat to ST3 and it’s only going to get worse if they don’t push forward with environmental analysis and route selection. Sound Transit is shooting itself in the foot by even considering these scenarios. There’s no need to change the scope of the project or delay the start of construction at this time. There are no issues with financing for the rest of the decade.

    The West Seattle – Ballard DEIS should have been released already but we’re still waiting for it. Just release it already and move ahead with selecting a route, finalizing the EIS and acquiring ROW. Then start construction of the entire line on schedule. If construction has to be paused or slowed in 2029 or 2030, then so be it. At least the project will be on the path to completion.

    This realignment nonsense is already causing the delay of the START of construction, since everyone is distracted by it while critical deadlines are missed and property values continue to increase.

    I get it, they legally have to plan for contingencies. But why do they have to do it now? We’re in the middle of a pandemic and the assumptions about 2030 are likely to be wrong. Move ahead with everything right now and revisit the realignment in 2025.

    1. To clarify, I’m not opposed to cost-cutting measures on Ballard-West Seattle. But these should be considered within the existing environmental process. Such as moving the guideway out of the path of the expensive new construction in West Seattle. That’s easy to do, just use more of the existing ROW (directly over Fauntleroy) rather than routing over private property.

  11. It is too bad that the money and planning for the ST3 second tunnel was not ready last year. With the unfortunate, and unanticipated problems of dowtown during Covid times, this would be the best time to build. Less traffic, and less impact. Even with a 5 year total disruption, 2 of those years would be during a time when less people would be negatively affected. In 2025 or 26 or whatever, I am assuming traffic and jobs to be back to normal. Or I hope.

    I seem to remember a STB recorded blog about a while ago. Timing was also discussed. I believe it was close to this: Vote for a massive transit project during good economic times. Collect the taxes for a couple of years during the good times. Then wait for a recession, and purchase the materials, housing and labor at a discount due to the economy. I know it was not meant to be translated that way. And I know that was not the messge trying to be given. But the result is the same. Unfortunately transit in Seattle needs to be built during a hardship.

    1. Construction can’t start until ST2 is finished because most of the money doesn’t come until then. ST3 is using all of the ST 1/2/3 tax streams, but the first two are busy until ST2 is finished and the ST1 bonds are paid down more. So only a third of the money is going to ST3 now. So it’s mostly going to planning, which costs a tenth of what construction costs, and is a prerequisite anyway. There were going to be low-cost early deliverables now (RapidRide C/D improvements) but they got swallowed by covid.

    2. For the new Downtown light rail tunnel in San Francisco, Stockton Street was closed for seven years. The pandemic effects on tripmaking will mostly ease by fall, which would mean the opportunity was only 18 months.

  12. How about we priorize connecting the areas/ cities/neighborhoods over putting up the artwork.

    1. Artwork doesn’t cost that much. The underlying law is “1% for art”, and I assume ST sticks close to that. It seems like it’s more because it requires selecting an artist and a piece during public hearings, which takes up half of the hearings’ time.

    2. The problem I have with the ST art program is that it’s random. It should be more strategic.

      1. The program should be useful in wayfinding. Entrance art and platform art should be related by exit path rather than each level being installed with a separate concept. Rather than platform and entrance being two separate installations, it should be exit A or exit B as the separate installations.

      2. Other cities use the art requirement to make essential architectural components more attractive. The San Francisco jail used the art funds to pay for unique exterior windows for example. Rather than drop an abstract sculpture in a station plaza, why not switch the clear station windows for artistic stained glass? Rather than hang an object from the ceiling, why not create inspiring Chihuly chandeliers? Why not create a lighting system that also helps with our winter light deficits?

      3. Finally, I really bristle that the program doesn’t promote local artists and themes. I am hard pressed to find many (if any) ST installations that couldn’t have also been appropriate for LA or Atlanta. Our region is rich with artistic talent and natural beauty and unique wildlife as well as a uniquely Salish heritage. Our stations should reflect what is unique to our region.

      1. #2, please! Good art should be beautiful and useful. Wayfinding would be great. A piece of art could provide shade or a wind break. It shouldn’t just sit there.

        There’s a reason older systems are filled with murals and mosaics. Whomever runs the program at ST thinks novelty equals creativity. Most of the art commissioned by ST is trash.

  13. “Weirdly, there’s no cost performance metric here.” That’s because transit agencies only care about ridership, not the cost of that ridership by route. Otherwise, Sounder North would never have opened, as it costs a ton per rider, while duplicating other Seattle-bound service from each station via that ancient form of transit called the express bus (canceling it would provide $ for light rail to Mariner, a far better use of public money). The problem is, many transit providers have an influx of the bulk of their revenues, i.e. public tax money, specifically sales taxes, that usually grows faster than inflation, while those lobbying for more service tend to believe that they’re entitled to have a heavily-subsidized ride no matter what time of day nor how scarce the ridership is: all that matters to them is that they have a subsidized ride when they need it, no matter the cost, and even if they live outside the taxing district. With many committee meetings under the shadow of secrecy, fiscal prudence and practicality is missing in these agencies, many of which don’t even have internal audit departments, and the focus isn’t always on service for transit-dependent riders who cannot afford their own transportation (vs. those who can afford their own transportation, but choose not to buy it).

    It will be interesting to see which option flies. IMO, it should be to open the spine first, which is the routing that follows I-5 and I-90. The best fit for that is a hybrid of Spine – Project Performance and the Tenure/phasing option. North: Opening Lynnwood-Mariner, the equivalent of the Northgate extension, but without any tunnels, would open up the connection to Swift Green bus rapid transit to Paine Field and Boeing, and riders would prove they are smart enough to transfer. ST could even subsidize extending that BRT line from Seaway Transit Center to downtown Everett in the interim: this could open in less than 2 years due to it sharing the Swift Blue line’s routing and only needing buses and drivers. The light rail dogleg to serve 2 stations: a declining Boeing plant and to Evergreen Way, with no stop at Paine Field, is not worth shelling out an extra $10 Bn and an additional 5 years’ construction time and should be re-evaluated for worthiness. Another benefit of opening the Mariner segment is moving the armada of buses carrying passengers to rail from Lynnwood to Mariner, saving some $. On the Eastside, I’d prioritize Issaquah to South Bellevue over the duplicative and concept-violating Eastgate to South Kirkland segment (the ST Board has said that there was to be no light rail along I-405, and while they rarely change their mind, in this case they should to resubmit and redirect the route south from South Bellevue to Tukwila International Station, then phasing out BRT in that segment). For the central section, only the second downtown tunnel is sort of in the spine, while Ballard and West Seattle are clearly spurs that duplicate RapidRide BRT. If there was a priority, I’d say it’s not in the first three phases, then it would be Smith Cove to Westlake Station and Delridge to SODO, saving connecting those two until a later phase (though I’d much prefer light rail between Ballard and U District than downtown). For the south segment, the priority should clearly be to extend rail to Fife and then to the Tacoma Dome and the Sounder to DuPont.

    1. Disappointed by todays ST Exec board meeting… It looks like ST will only push out the schedule rather than taking a hard look at reducing the plan.
      I like the Transit Rider’s suggestions but I don’t think many people will get on a short ride from Delridge to SODO just to wait for another transfer; RapidRide-C won’t even stop at Delridge. You could get a gondola built for less than another Duwamish bridge and serve the Junction directly rather than stopping at Fauntleroy Way and reach the International District without an extra transfer.

  14. ST has a while to do a reset of ST3. Delay is probable. Some important elements are unknowable: first, the shape of the recovery from Covid and recession. If the ST3 projects are delayed significantly, ST could make other changes to mitigate the delay. Candidate changes: improve headway on Link to reduce waiting; improve headway on truncated routes connecting with Link (e.g., routes 512, 522, 545, 554, 556); restructure routes 574, 590, and 594 to provide a short-wait spine via Federal Way TC; speed up bus routes via all-door boarding and alighting. Could the DSTT signal system be improved to allow shorter headway, if that is an issue? Could North Sounder be deleted with the Snohomish County funds shifted to bus and Link and the rolling stock shifted to the South Line? The ST3 parking could be zeroed out. I hope the infill stations are implemented soon (e.g., NE 130th Street, South Graham Street, and Boeing Access). So, ST could spend money on operations in response to the capital cost crisis.

    Note that ST has done this before; both Sound Move and ST2 were reset.

    Seattle can allow buses to flow well on the low level West Seattle bridge during Covid and on the high level bridge after it is repaired.

Comments are closed.