19 September 2021

Down the Tubes with DTX!

DTX overview
San Francisco's Downtown Rail Extension project (DTX), officially known as the Transbay Transit Center Program Phase 2, is a two-mile tunneling project to extend the peninsula rail corridor from its existing terminus in the Mission Bay neighborhood to the purpose-built basement "train box" of the Salesforce Transit Center (SFC). The project is regionally important, as there are more jobs located within a half-mile radius of the SFC than within a half-mile radius of all Caltrain stops combined, from 4th and King all the way to Gilroy. The DTX is nearly shovel-ready, in the sense that environmental clearance is in hand and engineering is being advanced to award construction contracts the moment a key ingredient becomes available: money. Gobs and gobs of money.

Too Big To Fail

The last time the costs of the Phase 2 project were tallied in 2016, the total came to $3.9 billion in year-of-expenditure dollars assuming a 2025 opening. Due to delays, we can anticipate at least another five years of escalation at 5%, bringing us to $5 billion before any changes to the project scope. One can reasonably expect that Bay Area transit agencies' proven inability to deliver mega-projects on budget or on time is quite likely to blow up costs well beyond these figures. As a recent example, the Phase 1 project, completed in 2019, cost $2.4 billion (year-of-expenditure) or about 50% more than the $1.6 billion YOE budget of May 2010, adopted after the train box scope was added.

The DTX project's regional, state and national significance is certainly not lost on our Transportation Industrial Complex. To improve the chances of getting the Phase 2 project federally funded (after which any cost growth becomes easier to fund, following former SF mayor Willie Brown's "theory of holes"), the TJPA is undertaking a phasing study to make the project appear more thrifty. The various approaches include deferring or deleting components of the project, such as a pedestrian connector to BART, an intercity bus facility, and an extension of the basement train box. This nibbling around the edges amounts to $0.4 billion in 2027 dollars or about 8% of the total Phase 2 project cost, a drop in the bucket.

A $30 million project development study is now in the pipeline to get Phase 2 to the state of readiness required to apply for federal New Starts funding by August 2023.

PAX: The World's Most Expensive Grade Separation

If you thought the cost of grade separations is exploding, you really haven't seen anything yet: meet the Pennsylvania Avenue Extension (PAX) addendum to the DTX, a grade separation project that will approach $2 billion for two crossings, reaching the stratospheric cost of $1 billion per crossing.

Even after spending $5 billion (before inevitable cost overruns), the DTX project will leave two existing street crossings at grade, at Mission Bay Drive and 16th Street. Not to be outdone, the city and county of San Francisco has performed a methodical series of planning studies to conclude that a new grade separation project is needed. Rather than taking on the challenge of bending some design rules to keep it simple and make it fit, the favored paint-by-numbers engineering solution is a bored tunnel, which averts any conflict with a planned 27-foot sewer pipe and the sacrosanct pile foundations of the I-280 viaduct, each of which are under the jurisdiction of other agencies. The combined cost of DTX + PAX is estimated at $6.0 billion. Take away the latest (2016) $3.9 billion cost estimate of DTX and you get about $2 billion added for PAX.

Link21 Crashes the Party

Meanwhile, BART is in the early planning stages for beefing up its throughput capacity between the greater East Bay and San Francisco, with a second Transbay Tube. It's worth pausing for a moment to consider what an astonishing piece of infrastructure the first Transbay Tube already is: it carries almost twice as many people during rush hour as the entire ten-lane freeway that is the San Francisco - Oakland Bay Bridge, and at significantly faster speeds. Looking past the pandemic, long-term growth trends indicate that the region must plan for a second Transbay Tube.

Transbay Tube II is the centerpiece of an enormous regional rail program known as Link21, the scale and ambition of which dwarf the DTX. While there are many decisions yet to be made about the implementation details of Link21, perhaps the most critical decision centers on what technology to put in the tube: wide-gauge BART, standard-gauge regional rail, or both.

This question is already of great concern to TJPA, which writes in its August 20, 2021 Phasing Study:

BART and Capitol Corridor’s Link21 program is currently in the early stages of development and has not yet determined a preferred alignment, technology, or rail gauge options to meet their goals and objectives for a future transbay rail crossing. As expected at this stage of development, all options remain available for consideration. For example, Link21 may determine that a second transbay crossing best meets stakeholder needs if it provides additional capacity for the BART network only and does not provide a standard gauge rail crossing of the Bay. BART’s infrastructure and trainset design, however, are incompatible with Caltrain and CHSRA standards. Most significantly, BART operates on a wider track gauge with vehicles that may not meet collision requirements, and therefore a BART-only connection would not relieve congestion and conflicts on the DTX.

We can already see a problematic mindset emerging here, where "BART" is automatically conflated with "five-foot-six track gauge," setting up a false dichotomy of BART-or-standard-gauge.

Caltrain + BART: a Necessary Merger

The false dichotomy of BART-or-standard-gauge threatens to poison the debate around Link21 alternatives. The Transportation Industrial Complex has a vested interest in this incompatibility, as it ultimately forces multiple mega-projects to be built. Why build it right when you can build it twice and get paid twice? From the standpoint of scope and profit maximization, it would then make sense to keep DTX and Link21 as separate projects, despite their overlapping purpose and need to link the greater Bay Area megaregion together using high-capacity passenger rail infrastructure. Seamless integration is good for riders and taxpayers, but not so great for consultants and civil engineering mega-firms. That's why these firms have an interest in propagating the myth that BART and standard gauge rail will always be mutually exclusive.

Removing this false dichotomy is becoming a primary reason for merging Caltrain with BART to form a single Bay Area Rapid Transit system, although there are many other reasons. BART does not have to be synonymous with wide gauge; indeed, BART already operates a seamless standard-gauge extension between Pittsburg and Antioch, and provides day-to-day management of the standard-gauge Capitol Corridor. A new BART peninsula line, while indistinguishable from Caltrain's service vision, would suck the air out of the emerging pointless debate around the track gauge of the second BART transbay crossing. The Measure RR sales tax can serve as a dowry to integrate San Mateo and Santa Clara counties into a restructured BART district.

Link21, to its credit, places equity and inclusion at the forefront of its project development process. The contrast with DTX is jarring, as TJPA's Phase 2 project can easily be viewed as just another gold-plated white-collar rail project enabling nine-to-five technology and finance types to more conveniently access San Francisco's skyscrapers from the affluent suburbs to the south, without ever having to mix with the blue-collar working class. Bringing DTX under the Link21 umbrella, and merging Caltrain into BART, immediately defuses the classism and racism that underlies this anachronistic Mad Men commuter rail vibe.

Transbay Through Running

A stub-end terminal station suffers from fundamental throughput limits related to long turn times and the unavoidable crossing streams of inbound and outbound traffic in the station approach or "throat." For a given number of platform tracks, a through-running station configuration where all trains that come in one end of the station can exit the other end will always provide more throughput capacity, whether measured in trains per hour or passengers per hour. Trains don't have to dwell any longer than necessary at a platform, and don't foul opposing traffic on their way in or out.

With the DTX as it is, past operational analysis indicated that just 12 inbound and outbound trains per hour (8 Caltrain + 4 HSR) would push the limits of the terminus design, with near-saturated platform occupancy. If you uncork the other end of the train box (by having Caltrans clear some right of way i.e. dismantle and redevelop a couple of medium-rise buildings to the East) so that the DTX can connect directly to a new transbay crossing, everything changes. A lot of new capacity is created by virtue of not having to layover or turn trains right smack where your platforms and track real estate is the most expensive.

A recent through running operations analysis commissioned by the TJPA shows that the Salesforce Transit Center could handle up to 20 trains per hour per direction if no more than six of them turn at the station. Any more than six turning movements, and the excessively long platform re-occupancy times (as the study notes, due to the poor layout of the switches leading to tracks 1-4) will reduce throughput capacity to less than 20 trains/hour.

Broken Assumptions at Link21

The TJPA phasing study reports the following direct quote attributed to Link21 project team:

We have received briefings on the operational modeling for DTX and it would seem that even a three-bay DTX tunnel poses operational constraints. A robust service level through the transbay crossing is required to justify investment into Link21. Link21 is envisioning scenarios where not all trains that cross the Bay would continue to San Jose. At this point, there is no other location to turn trains around in the northern peninsula which makes flexibility in DTX important to the Link21 Program.

You read that right: the Link21 team is thinking of turning Capitol Corridor trains at the Salesforce Transit Center, a completely American idea (copied straight from Penn Station New York) that is operationally insane if you think about it for even a minute. In a through-running configuration, all trains that cross the Bay should stop in downtown San Francisco and get out immediately. The California High Speed Rail Authority is planning a huge yard in Brisbane, a perfect place to clean, service and layover Capitol Corridor trains. These deadhead (non-revenue) moves are much less wasteful of infrastructure capacity than treating a through-running station as a terminal.

As was remarked in previous discussions regarding San Jose, the act of parking or laying over trains at a station platform is the railroad equivalent of parking an empty truck in the middle of a bustling loading zone, and then concluding that the loading zone fails to function adequately. Just stop it, don't even think of turning trains here!

The Bottom Line

Here are the pros and cons of merging DTX with Link21:

 Eliminates silly idea of a multi-gauge transbay tube project Could further delay DTX, since Link21 is at an earlier stage of development
 Increases SFC throughput capacity and bang-for-buck, making the enormous cost of DTX worth it
 Exposes DTX to political re-prioritization
 Provides faster Peninsula - East Bay connections than existing BART, and finally "Rings the Bay with BART"
 Greatly reduces scope and profits for Transportation Industrial Complex
 Makes more efficient use of taxpayer dollars by building  one project and building it right
 Requires inter-agency coordination and mergers, which agencies abhor
Provides seamless regional rail connection from SJ and SF to Sacramento, if Capitol Corridor is electrified

Despite the obvious political and organizational obstacles, from the point of view of a rider and taxpayer, the pros vastly outweigh the cons. The answer is then obvious: the DTX should go down the tubes of a new standard gauge Link21 crossing, with Stadler bi-level EMUs operated by BART seamlessly connecting the peninsula corridor (a.k.a. the new BART Purple Line) directly to Oakland and points beyond. DTX should be built without delay and form the first building block of Link21.


  1. What year it’s going to be completed?

    1. Never. The endless "studies" and hundreds of millions of dollars of endless graft are their own reward.

    2. TJPA is now aiming to complete DTX in 2029, which means 2032 at the earliest. Link21 is probably further out than that.

  2. When it’s going to be completed?

    1. I used to be active in a group called "Peninsula Rail 2000".

      Hilariously enough, it was, back in the early 1990s, a realistic goal that Caltrain might become something other than "Olde Tyme Commuter Railroading" (ie an S-Bahn) and that electrified Caltrain would run to downtown San Francisco (the Transbay Terminal being the only appropriately-sized site) every fifteen minutes, all day, every day, by the early 2000s.

      In any remotely normal remotely functioning first-world industralized democracy this might have been the case.

      Instead, we have Caltrain. We have MTC. We have still have fucking Steve Heminger. (Why hasn't he strangled himself?) We have a planet literally on fire. It is 2021, and literally everything is worse, and going off a fucking cliff.

  3. I have to say that I'm a bit suspicious of the whole Link21 endeavor. It seems like they started with the premise that a new transbay crossing is needed, and then set out to find some regional justification for it. This seems somewhat... backwards. It seems like BART and SF trying to find someone else to pay for their infrastructure. Maybe they should just build more housing in SF instead?

    Of course we see the same pattern with DTX. Originally it was a Caltrain/SF pipedream to relocate the 4th+King station somewhere more useful. Then HSR came along and suddenly it was a "regional" project, so it was ok for SF to piss away the local funding for the station on a $2B bus station and 'park in the sky" since the rail station was now somebody else's responsibility to pay for.

    Of course the new Transbay terminal was quite successful as a real-estate project as the many new skyscrapers that resulted are evidence to. Less successful at actually improving transportation. It's pretty clear that the continued focus for the TJPA is real-estate interests. Just search the phasing report for mention of planned development at 655 Fourth St.

    Anyway that's just my cynical view. I agree that a through-running station makes a lot of sense for Caltrain (less so for HSR as the trip time to Oakland would be the same as transferring to BART at Diridon). The cost of a new crossing would be astronomical however. The real-estate cost alone of those "medium-rise" buildings that would need to be demo'd is probably the best part of $1B.

    As far as merging BART and Caltrain is concerned, it certainly sounds good in theory. However, I can't imagine a world in which SAMTRANS lets their old nemesis BART take over *their* railroad. They can barely stand to give SF and SCC a say in how it is run as it is. The recent sabotaging of a JPB meeting to even discuss regional governance options is a good indication of how SAMTRANS react to the mere thought of a BART merger.

    1. There’s probably a negative cost (profit) to be had in rebuilding the east station throat medium rise buildings much taller. The buildings aren’t in the way; their foundations are. New towers can be built with foundation piles in places where trains can run between them, much as with Hudson Yards in NYC (way bigger and more complicated than SFC).

    2. I guess there's profit involved for somebody in the long term. It seems unlikely that profit ever gets to the poor transit agency that builds this. I think they still need to buy the property in the short term, demo the existing building, build the tunnel and then sell the lot on to a developer to build a new taller building on after construction is done. It's true they may claw back some of the cost though

      One example "middle-rise" in the way is 19-story 160 Spear Street which sold for $200M in 2015. For reference, nearby Transbay parcel F which was entitled for a much taller 750ft tower sold for $160M in 2016.

    3. 160 Spear is very much in the crosshairs. $0.2B is a rounding error, on the scale of these projects!

    4. I'm with you in being suspicious about Link21. While BART was approaching capacity, it was already planning on increasing it by 50% via new cars with 3 doors, going from 22 tph to 30 tph, and running 10-car trains at all times. The new cars are already coming in, they broke ground on the extra substations at Montgomery and Civic Center, with only new signal system remaining.

      A second BART tube's best justification is to provide service to Alameda and Mission Bay, new Seaplane Lagoon ferry service, sped up Transbay buses via direct access to Transbay have lessened the need for that.

      Extending Caltrain under the bay just makes more sense given the enormous costs.

    5. Re: 160 Spear is very much in the crosshairs. $0.2B is a rounding error, on the scale of these projects!

      You're assuming there's any SF-local or even regional appetite for anything remotely like what you propose.

      There isn't, and there's not going to be within your lifetime.

      Hey, you know that Transbay Terminal that doesn't have any trains and which is designed to make train and passenger access to the platforms impossibly slow and congested and that was hundreds of millions and years late? And you know that whole Millenium Tower thing? Yeah, well, we want billions more, to Do It Right Trust Us this time, and we're going to take out a bunch of office buildings to do it and really it's going to be great. Also, no private party would ever think of greasing the palm of the SF Mayor or, just for a totally random example from the recent past, the palm of an unexplicably-unindicted SFCTA Executive Director, in order to demolish a "rounding error" building and build something more profitable on the site, its foundations unencumbered by several train-sized holes.

      Sure, the entire political class of SF will line up behind this.

      Transbay Terminal is the end. It's terminal. It's not the beginning of anything. The grotesque misdesign mismanagement and fraud ("architectural competition"!) associated with this non-train non-station park-in-the-sky is fatal poison. Nobody wants more of this. And nobody but some train nerds are going to go to bat for "more of this" against actual commercial developers with actual plans and actual (it's super cheap! Ask Willie for his rates! You'll be surprised!) ability to buy votes and make actual money. This is the sort of thing that's a go; crayon minimum-40-years-off Caltrain through such sites to mumble-mumble-mumble east-bay-ish somewhere isn't.

  4. With any luck, link 24 does a new EIS and picks a better (non zig-zag) route for DTX (down 7th and up Howard) with a few new stations (and 4 tracks) along the way. It seems like transit in the bay area is planned based on what's the most expensive least effective option.

    1. Here here to that. If they do pick the Howard/7th alignment, might as well restore full-length 16 car train (400+ m) platforms while there at it.

    2. I gather the main criticism of Howard / 7th was that it didn’t include their sacrosanct third approach track, which the DB run-through study shows is not needed if trains run through. Imagine that.

    3. The non zig-zag route would be to proceed straight from 4th/King out towards the East Bay, bypassing SFC entirely. It is insane anyone still thinks $5 billion is a reasonable expenditure just to get trains .7 miles closer to downtown.

      First law of holes: stop digging.

    4. I believe that one of the reasons that Howard / 7th wasn't considered is because of the already permitted massive sewage pipe. Another reason was because of the Central Subway tunnels which would have caused any tunnel under Howard to come up at a gradient of a little over 3%.

      These "reasons" are quite pathetic though. The sewage pipe can easily be re-engineered and re-permitted to fit in with any tunnel/underground station at 7th. Furthermore, Caltrain and HSR are using EMUs so steep gradients are a nothing burger. If the Shinkansen is capable of sustained 180mph on 3% slope, I'm pretty sure that a Stadler KISS would be able to climb 3% at 60mph.

    5. Also re Seventh to Howard: Moscone East and Moscone West have a huge (and recently enlarged) underground connector. Any tracks under Howard are going to be REALLY DEEP, and any station under Howard is going to be REALLY DEEP and crazy expensive, with crazy long escalators. Just think about what would go into excavating a 420m long 66m wide (2x11m island platforms) station box under Howard. Oh hang on, Howard Street isn't 67m wide! Now you're talking four tracks stacked two over two a squillion feet below surface.

      Hey, Amtrak's consultants just called and have a Penn Station South to sell you, along with several bridges and tunnels.

    6. Why would you need a 66m wide station under Howard (4 tracks, 2 island platforms)? DTX only needs two tracks to Transbay, and extra tracks are certainly no use if there is only a two track through running tunnel under the Bay.

      If your answer is for some trains to bypass the station while others stop there, don't do that, have every train stop. A stop at Howard/4th access a huge pool of jobs and destinations at Moscone and Union Sq. plus the connection to the Central Subway.

      Even if you did have four tracks and two 11m platforms, this should be somewhere in the range of 40m wide (4x 4m tracks, 2x 11m platforms, 2 x 1m walls - note track centers nominally 5m but only 1.5m from platforms for level boarding) not 66m.

      As Drunk Engineer notes below, including passing movements and staging for tunnel entrance in the approaches to DTX via extra tracks makes infinitely more sense than including extra tracks in the tunnel itself. Why spend vast sums creating passing tracks in the most expensive possible place to build?

    7. "DTX only needs two tracks to Transbay, and extra tracks are certainly no use if there is only a two track through running tunnel under the Bay ..."

      The throughput limit on central-city transit is dwells of stopped trains at platforms at the most heavily used stations, not the signal headways of free-running trains on tracks between stations.

      A typical contemporary arrangment, globally, is one approach track bifurcating to serve two platform tracks across one passenger platform, allowing close-headway trains to overlap their dwell times. Whether some particular service arrives on the left or the right side of the platform is of no practical consequence to passengers and to train throughput. (Zürich Hauptbahnof has done this twice, with two adjacent underground two-track lines serving two adjacent four-track through platform new station additions. Likewise Stockholm. etc. etc. etc.)

      And if you're not modelling level of train throughput at which CBD station dwell delays aren't the limit on your several-tens-of-billions-of-dollars capital project then why the fuck are you even thinking about spending billions in the first place? Under 8tph is "are you joking why WTF just stop it" territory. (Meaning, all of the USA USA USA.)

    8. SEPTA runs 15 an hour through Jefferson during rush hour. Last time I checked Philadelphia is still in the U.S.A.

    9. Richard, thank you for the informative reply.

      This raises a question, however. How do subway systems maintain such close headway using just one platform track per track entering the station? Even BART manages to operate 24 tph through SF using two tracks and one island platform, despite 10 car trains that are as long as 8 of the KISS derivative cars that Caltrain is getting (or two 100m FLIRT combines).

      Even if you have to have two platform tracks per running track, this still doesn't preclude a Howard alignment. Alternatives are:
      1) No station, Moscone is only 1 km from Transbay, this is subway level spacing and not required for commuter rail let alone intercity service.
      2) Four tracks but one platform, in a 4-3-P-2-1 configuration. Some trains would serve the station using tracks 3 & 2, while others would address the throughput issue via "false stops" on tracks 1 & 4 (slowing through the station but not stopping). With a 9m platform this fits under Howard's 29m width.
      3) As I noted even with 11m platforms the station width would be only 44m, and this could be shaved with 8 or 9m platforms. There is no reason the station has to fit fully under the street. Moscone West is a hard barrier on the north side of Howard, but to the south there are a mix of buildings that are 2-7 stories; undercut them a few meters or just remove them to build the station then redevelop the land on top after.

    10. I should also point out I am in favor of the current general alignment because:
      1) The zig-zag doesn't lead to a longer tunnel because of the taxi-driver problem (straight then right is the same distance as right then left on a street grid).
      2) As Richard notes below the zig-zag should actually be a shorter tunnel if you are smart and trench the existing ROW to get closer before the TBM starts.
      3) 4th & King is far enough from Transbay that a station there would serve meaningfully different neighborhoods (South Beach and Mission Bay) while still connecting to the Muni T. A station at Moscone is really too close to Transbay.

    11. Richard, I don't think putting a station underneath Howard would be necessary. I was thinking of putting an underground station at 7th and Townsend, near where Amazon just bought up a huge property. A station there wouldn't need to be very deep. From there, the tunnel would then start descending deep enough to pass under the Central Subway tunnels. Maybe the Central Subway tunnels can act as a vent shaft for emergency alightments.

      I'm not against the "zig-zag" because of the length of the tunnel. I'm against the current alignment because of how slow it is. Three tight, 90 degree turns are limiting trains to under 40mph. Just from eye balling on Google maps, one smooth curve from 7th after passing underneath the 101/80 freeway onto Howard would provide trains with only one braking point. Speeds would definitely be higher that 40mph on that curve. No need to accelerate, then brake, accelerate, then brake, then accelerate again.

      Personally, stacking tracks is what I originally imagined when Prop 1A was passed. I admire the Japanese rail model so I would prefer that Caltrain and HSR have dedicated tracks and therefore dedicated train levels at Transbay, but hey that's just my opinion.

    12. Hi Mister DudeSF,

      I guess I can't agree with any of your premises -- don't attempt to put a station anywhere near the SF CBD, meaning anywhere near by by far the greatest ridership site anywhere north of downtown Los Angeles); prioritize meaningless increases in speed over serving passengers at stations; put concrete before organization and build insanely duplicative redundant facilities -- so I can't really engage with this.

      Since the late 1980s there's only been one feasible and suitably sized station site near the SF CBD. Since 2000 there's only been -- because real estate developer payoffs to Willie Brown and the SF Planning Department destroyed superior alternatives -- one feasible route to connect the Caltrain line to that station site.

      I don't see the choice as being PTG's DTX with the utterly bat-shit fraudulent multi-billion "Pennsylanvia Avenue" scam versus herp derp Link21 herp derp regional herp derp more studies herp derp.

      I see it as cut-cut-cut-cost lemons-from-lemonades not-entirely-insane revised DTX (meaning just go back to what we always proposed, not what PTG's criminals have inflicted on us at a "design" cost measured in the hundreds of millions, and climbing) in a decade vs nothing for 40 or 50 years.

      I'm intimately familiar with nothing for decades on Caltrain, nothing at all, and I'd prefer to die without another couple more decades of the same but, hey, that's just like, my personal preference.

  5. Caltrain doesn't need to serve the existing but awkwardly-located 4th and King station. Instead, replace the existing SF stations with stops at Oakdale, UCSF / Mission Bay at 7th / 16th Sts, and SoMa at 4th / 5th Sts. Past SFTC, the tracks can link up with the existing CC line / proposed wBART line via Emeryville to Hercules.

    Long-term, a southern branch could be built to Bay Fair via Oakland JLC, and linked to a re-gauged Dublin / Pleasanton branch. This would allow for one unified standard gauge line between SF and the Central Valley via ACE and Valley Link, and more BART trains to run to Hayward / Fremont / SJ.

    1. I like the idea of a re-gauged Dublin/Pleasanton branch. This makes much more sense as a route for Altamont HSR than tunneling under Fremont + Dumbarton. The only reason it was ruled out originally was the cost of a new transbay tunnel I think. Run HSR through SF and terminate at Bayshore maintenance facility / park+ride. Serve SJ via BART transfer or a spur via Alviso.

  6. Instead of a second tube (a very expensive underwater expedition), perhaps BART should implement passing tracks (on land) so that it can have express trains, and increase the utilization of the existing tube (which is only busy doing commute time and only in the "SF commute" direction).

  7. OK. So everything about this is wrong.

    Unless you never, never in anybody here's lifetime, wish to see Caltrain run anywhere in SF besides Fourth and Townsend.

    So ... The problem: PTG (primary "desigb" criminals)/TJPA/PCJPB/SFCTA/CHSRA have an utterly shitty plan that involves far too much tunnelling, has blown out its budget several times over, is decades late, has descoped into meaninglessness, and has nothing at all to show but tens of millions of dollars (may be hundreds of millions now -- it's too depressing to go and look and add it all up) for "studies" and "planning".


    This time for sure.

    PS I've been around the Bay Area long enough to have seen least four, perhaps more, iterations of MTC-shitfest consultant porkfest do-nothing-ever "Link21" type studies come and go. This is no different. Same house, same call.

    But this time for sure! MTC is going to study integrated ticketing! Everybody's up fo studying an integrated regional rail network! It's going to be great! Just sign this blank check and sign up for doing nothing for another 20 years. You're going to love the thick binder of map-crayons we're going to deliver ... eventually.

  8. Re regional rail to the East Bay: the issue is that THERE IS NOTHING TO LINK UP WITH.


    You run Caltrain through a new transbay tube to Oakland ... THEN WHAT?

    Have any of you people actually seen the freight/Amtrak route and tracks from Oakland to Benicia? THERE IS NO THERE THERE.

    You run Caltrain through a new transbay tube to Oakland ... THEN WHAT?.

    Having anything to connect to on the Oakland side is so many decades away that SF DTX seems realistic, timely and cost-effective.

    I mean, maybe you can turn the very-soon-to-be-underwater Alameda Naval Air Station into a parking lot and turnback for Caltrain so that Clem gets his through-running in SF. A few billion for a bunch of deadheading, but hey, through-running! But beyond that? So many hundreds of billions and so many decades away that ... THERE IS NO THERE THERE

    Oakland is totally "there" in terms of regional centrality. It's awesome. It's where I'd chose to live if I weren't settled where I am. It's the core of BART. There's so much potential for urban development and realistic (surface!) transit improvements, some of them even haltingly happening. Maybe some decade the hateful crimes of the city-destroying freeways will be homeopathically band-aided. But none of this changes the fact that there are zero standard-guage high-passenger-train-frequency-compatible corridors for a standard-guage non-freight high-frequency trans-bay rail line to connect with. Nothing!

    Get out there, ride BART or your bike or around and look around Oakland. Hell, even Google Earth can sort-of show you. Stop with the crayons!

    1. That's my worry too. One idea I had would be for CC to run underground between Coliseum and Emeryville stations to bypass the slow street running on JLS. Since I'm dreaming, there would be an underground station below 11 St BART stations. Trains heading south from there would branch west towards 2nd Transbay Crossing to SF Transbay Terminal (and it wouldn't be a terminal anymore).

      This would provide a better CC connection to BART than the slow and circuitous one via Richmond, Berkeley, and Oakland.

    2. The Bay Bridge is the second busiest in the country, with at least a quarter million people crossing it every day, apparently going to and from nowhere....

      There is very little from Richmond to Benicia, but plenty from Richmond to Fremont. The inner East Bay is 1.5M people in a 6x55km corridor. This is the size of Santa Clara county (but closer to SF) or of SF + San Mateo county (albeit a slightly longer corridor). If it is worth having Caltrain it is worth bringing a line to the East Bay.

      There are two rail corridors south of Oakland, and one north (although more poorly sited). There is existing passenger rail service on two of the corridors today. Major improvements would be needed if a standard-gauge tube were built (electrification, high platforms, an ~1.5km viaduct through DT Oakland) but complaining that a high-passenger-frequency line doesn't exist today is foolish when Caltrain doesn't meet this standard yet. The existing lines are about where Caltrain was in infrastructure/service when it started, no reason they can't be brought to where Caltrain is/will be.

      Also, as you note, Oakland is the central rail hub for Northern California. There are existing services going beyond where an E Bay Caltrain (or Clem's 'actually BART') would make sense. Ridership on Cap Corridor and the San Joaquins would soar with direct service to Transbay and beyond to Stanford, etc. HSR Altamont-Oak-SF makes more sense than through SJ (DT Oak has far higher employment and residential density than DT SJ or anywhere in SV).

      I look around Oakland and see high-rise office towers and lots of mid-rise (some high-rise) housing development. If that doesn't justify commuter rail service to the regional center, what does?

  9. Re PAX: The World's Most Expensive Grade Separation.

    Everybody knows this is insane, but it is extraordinarily profitable.

    As has been stated over and over, the first law of holes is stop digging.

    There can and should be zero new tunnelling between the DTX interface at the portal of the existing Tunnel 1 in SF (south of 16th Street, below the Mariposa Street overcrossing) and the intersection of Fourth and Townsend Streets where the two-track-only cut-and-cover tunnel under Townsend starts.

    Following approximately the existing alignment under Highway 280, it is perfectly feasible to trench -- not cut-and-cover TUNNEL, and absolutely not, as some insane eliminate-strawman-alternatives clowns at SFCTA claimed, drive a god-damned TBM -- between the Mission Bay Viaduct highway footings, with only two or three foundations causing any sort of trouble at all.

    It is perfectly feasible to site a thee-track, three-420m-platform trenched and not covered station on the northern edge of the Sixth/Townsend/Fourth/King site, occupying less than 40% of the existing "rail yards". No tunnels! No insane mezzanine levels. No out-of-control fire safety pork. Just a station, open to the sky, with rain canopies over the platforms, connected by escalators and elevators to the streets above, serving passengers.

    As for the "mandatory" three-track minded tunnel from Mission Bay to Transbay -- the people who came up with this are at the same time the very stupidest and most corrupt excuses for humans you could ever meet. It's purest pork.

    I mean, just think about it: under what circumstance are three trains going to be moving in this section, and if so, why, meaning what scheduling mistake did you make? A third track is simply a nosebleed-cost parking track, underground.

    The place to sort things out is at a three-track shallow-trenched comparatively-low-cost Mission Bay station. Dispatch trains into the short, massively-expensive tunnel only when they have a place to arrive at the terminal. Hold trains as necessary in the open air to make this work -- don't park them in a three-track tunnel under Second Street.

    The whole scam is both criminal and insane.

    This is not how anybody else in the world does things.

    Solely America's Finest Transportation Planning Professionals. They should be in prison, all of them.

    1. It definitely makes more sense to build the station in a trench, and they could still build on top of it, maybe let a developer pay for structural columns and a cap (could help offset the cost). And maybe they don't get to have an underground parking garage underneath, but boohoo, it's directly over a transit station.

      What do you think of including an extra platform or two in said trench to allow some trains to turn back without blocking the tracks into the Transbay? Might be useful for evening events at the arenas.

    2. "What do you think of including an extra platform or two in said trench to allow some trains to turn back without blocking the tracks into the Transbay? Might be useful for evening events at the arenas."

      I thought that was an idea (not a good idea, but something perhaps nearly "free", so hey, whatever) 15 or so years ago, but I was wrong.

      Besides not really being geometrically feasible and leading to bad compromises, and besides the large cost and complications of shoring up an unnecessarily wide trench in bad soil just for an "emergency use" (meaning useless, in practice) terminal track short of The Terminal, this buys into the quite literally insane concept permeating Caltrain "planning" that Giants baseball events are the peak load drivers and the construction window blackout veto driver and that, fundamentally, the SF CBD is just negligible shit compared to a few baseball games a year, because hey, we don't go to the actual hyper-dominant regional CBD and never will, herp derp, how about that Googleplex?

      I mean, just look at how BART, which is pretty well run, all things told, runs things -- run a ton of trains, all the time, to places in SF people want to go and will actually use, and then some "special event" blips are just running a few more trains at some times on top, not the overriding determinant.

      (For decades Caltrain daily riders have been totally fucked over by Caltrain staff prioritizing uninterrupted Giants baseball over regular "service" when any sort of construction anywhere between SF and SJ has been considered.)

      Caltrain should be running 6tph to and from Transbay most hours of the day every day. If some of those trains happen to pick up some drunk entitled suburban baseball fans and cater to a far-right sportsball oligarch's economic interests while on the way, sure, great, whatever!

      Anyway, a basic 3-track 3-platform Mission Bay station -- especially with 420m platforms and mid-platform crossovers splitting two of those platform tracks into double berths for four 150m (6 "car") Caltrain sets -- just happens to function pretty damned well as an emergency turnback location, while serving its primary function as a traffic-flow staging and smoothing point for traffic to the SF CBD.

    3. @Adam,

      It misses the point to build on top of the trenched station because it then becomes an underground station, which leads to much high costs relating to ventilation, fire alarm, smoke control, etc. Richard's whole intent is to minimize cost by taking advantage of existing ROW and avoiding the "subway" costs that come from tunneling underground when no ROW exists.

    4. Richard, since the route is mostly fixed at two tracks starting with the four existing SF tunnels, wouldn't the place for a three track station for the purpose of dispatch be at Bayshore station?

    5. "Richard, since the route is mostly fixed at two tracks starting with the four existing SF tunnels, wouldn't the place for a three track station for the purpose of dispatch be at Bayshore station?"

      There also, but for different reasons -- eg a shuffle to put a non-stop train ahead of of stopper if the total person-delay were lower, etc. But a shuffle to run wrong-track is going to be very rare given (what has to be) pretty constant bidirectional travel.

      A Mission Bay traffic buffer ahead of Transbay has far more to do.

      First, you don't want to put trains in a tunnel (for crazy made-up bullshit US fire safety reasons for sure; but also because a $$$ tunnel is a terrible place to park unmoving trains.) Temporarily holding a train at a nice station with platforms and open air beats the hell out of a dark hole deep under Second Street.

      Second, even if the approaches to Transbay hadn't been fatally sabotaged, for no reason at all, by some the very stupidest, incompetent and unprofessional subhumans on the planet (especially Parsons Transportation Group, who are still sucking down millions of public cash for DTX "design" -- endless criminality), there are still going to be of cases where "wrong track" (left-hand) running through the Mission Bay to Transbay tunnel section is going to result in freer traffic. Basically, with only flat junctions in the approach to a multi-track station, you have route conflicts, but sometimes you can use a second interlocking (set of flat crossovers) further down the line to resolve some of these conflicts.

      So ... what you want at Mission Bay is, first, an island platform with two platform tracks interchangeably serving all stopping outbound trains. This allows trains to be dispatched through the tunnel to whichever side of that Mission Bay platform improves traffic flow, without messing up passengers waiting to leave SF. (Just another example of why island platforms should be used uniformly everywhere along the line, and always FSSF.

      (Note that this doesn't matter for inbound traffic -- ridership from Mission Bay to Transbay is of zero importance ... and even if if not, guaranteed another train is coming within 10 minutes.)

      The second thing you want is at least three platform tracks so that you can indeed buffer inbound traffic there a little to smooth out terminal throat track conflicts further down along, and, again, to not stall trains in a harder-to-evacuate $$$ tunnel.

      The last optional (but basically free, so do it!) thing you can have an Mission Bay are platforms long enough (the standard 420m everybody uses, and that we can have here and Redwood City and SJ Cahill Street -- just like grown-ups are doing in places like Madrid Atocha) to double-berth shorter (150m, 6-car) Caltrain sets with space between them for mid-platform-track crossovers, sort-of giving you four service platforms in the space of two, and superior possibilities for turning some trains short of Transbay in exceptional (never regularly timetabled, that would be insanity!) circumstances.

      In short http://www.pobox.com/users/mly/TTT-200909/200909-Mission-Bay.pdf

      * 420m open-air trench stretching from Sixth to Fourth Streets (under a restored Fifth Street!)
      * Three through tracks; three platform tracks
      * Island platform between northernmost tracks; side platform for the southmost
      * Full universal any-track-to-any-track three-to-two-track interlockings at both the outbound (under 280) and the northbound (under Townsend Street) ends;
      * Scissors crossover pair right in the middle of the adjacent platform tracks for extra fun.

      You're welcome. $200 million plus 10% of costs saved ought to cover my services.

    6. Richard,

      Thank you for the informative reply.

  10. I fail to see how any of these plans (including putting a station in a trench) is better than the current arrangement. The existing 4th/King station is perfect as it is. The access is all at-grade, so absolutely no elevators or escalators or mezzanine levels. And it certainly doesn't lack for platform space. If you want a European-style central station, this is as close as you will ever get.

    As for the existing at-grade crossings, just do a standard road under-crossing (and by standard I don't mean the option "studied" by SFMTA which entailed running 0.6 mile tunnels 10 stories down). Then take the $10+ billion you saved by not digging tunnels or trenches and spend it on some actually useful project.

    1. Put a signature / iconic shed over the top of it and done

    2. "If you want a European-style central station, this is as close as you will ever get."

      I can't even.

    3. The trend in Europe over the past 50 years has been to move away from the "European-style central station" (i.e. at-grade terminal) in favor of underground through running for commuter/regional rail. See: RER, Crossrail, Hirschengraben/Zurich HB SZU (then later Weinberg Tunnel/Löwenstrasse Station), Stuttgart 21, Munich Stammstrecke, Stockholm Citybanen/City Station, Oslo Tunnel/Nationaltheatret Station, Milan Passante, Frankfurt City Tunnel, etc.

      Note that Stockholm built Citybanen for better access to Centralen even though it already had at grade through service at Stockholm Central Station. Berlin was of course a pioneer in this respect with Stadtbahn dating to the 1880s and Nord-Sud built in the 1930s.

    4. The projects you mention are generally through-running services. They generally don't terminate at a badly designed station at the city center. And their cost was order(s) of magnitude less expensive (Crossrail might be an exception?). I don't see any similarities between those projects and DTX.

      I would add than in the case of Milan Passante, the suburban network does not connect to Centrale, and there are no plans to do so.

    5. This blog post specifically addresses DTX in the context of through-running using the proposed Link 21 transbay crossing.

      It is true that Passante does not connect to Milan Centrale, however, Republica and Porta Venezia are closer than Centrale to the historic core of central Milan, while Porta Garibaldi is closer than Centrale to the CBD of Pora Nuova. That is the point: Passante (the "downtown" tunnel) does a better job for commuter rail of getting people closer to where they want to be than the "European-style central station" that is Centrale.

      It is true these projects cost less than DTX is projecting, but there are ways to lower costs, people in these comments have provided several already. More importantly, arguing that the cost-benefit analysis is poor is far different than arguing the benefits are lower, as you did by suggesting that 4th & King is perfect or better than Transbay would be. No one in Europe would propose a location/layout like 4th & King for a new commuter rail station, and where commuter rail does use similar stations they are actively building DTX style infrastructure to replace them. See for instance RER E: lines that used to terminate at Gare D'Est were taken in a tunnel across the city to currently terminate at Hausmann the way DTX only would at Transbay. Extension for through-running is underway, but the point is to take Transilien J trains that terminate at St. Lazare the way Caltrain does at 4th & King and bring them into center Paris.

    6. As was previously commented, Caltrain run-through service makes no sense as there is nothing in the East Bay to hook into. The Amtrak service will always be a joke operation so long as UP controls the ROW.

      Past history suggests that Link-21 will either be vaporware, or else used as a pretext for BART to cannibalize Caltrain DTX funding.

    7. Right now, the Capitol Corridor is the only thing that makes sense to use a new tube

    8. "Right now, the Capitol Corridor is the only thing that makes sense to use a new tube"

      Exactly right, except for the "sense" bit.

      SF-Oakland-Hayward? Seems there's already a perfectly servicable high-frequency train line there.

      SF-Fremont(-Livermore-Tracy-Sacramento/Los Angeles)? I have an actually useful tunnel to sell you.

      Fremont-San Jose? Hard to undo the criminality. Prosections, starting with Steve Heminger would be a good start.

      SF-San Jose? Seems there's a train line there already.

      SF-Oakland/Alameda? Maybe you get one East Bay station, connected to nothing else. (Exercises for the reader: show exactly where you've found the site for a rail passenger terminal station of a size and ridership potential remotely commensurate with the costs of a brand new SF-Oakland/Alameda rail tunnel and approaches.)

      Oakland-Richmond? Seems there's a pretty good passenger train line there already.

      Richmond-Crockett/Martinez? You and what army?

      Crockett/Martinez-Benicia? You and what army?

      Benicia-Davis-Sacramento? UPRR would like a word.

      SF-Sacramento? I have an actually useful tunnel to sell you.

      Oakland-Sacramento? Would have been nice, but endless profits for Steve's very very special buddies at WSP (née Parsons Brinkerhoff) and Tutor-Saliba were far more important than the core of the urbanized region, or the earth's environment, or anything else.

      A non-BART SF-Oakland/Alameda Transbay line has nothing at all to connect to this side of mutiple hundreds of billions of brand new rail line construction in brand new rights of way. It's simply a nonsense crayonista fantasty (for rail fans) or rent-seeking (for "transportation consultants") to waste a minute of thought of this at any time after around 1999.

      Link21 is just another consultant-enriching MTC attention-diversion scam. Many of us have seen exactly the same agencies do the same thing several times already over the decades. EXACTLY the same thing. The answer is always "more studies" and "oops even more BART extensions drank your milkshake".

    9. A standard gauge transbay tunnel would require new rail line construction (for the electrification if nothing else) but by no means in new right of way. There are three very serviceable ROWs heading out of Oakland, one to the north and two to the south. There are two further lines heading east branching off of these.

      Cherry picking small case studies and finding fault is literally failing to see the forest for the trees. Look at the totality: 1.5M people just to the west of the Oakland Hills, 3.2M in Alameda, Contra Costa and Solano counties. Several million people is more than enough to justify multiple rail lines (not just one BART line), plus regional service to other areas, plus the major connection (the tunnel) to DT SF that is the ridership driver making it all work. One could use your "case studies" to conclusively prove we should shut down Caltrain ("SF-Millbrae? Seems there's a train there already." "High platforms preventing freight traffic? Port of SF would like a word." "SJ-Blossom Hill? You and what army?")

      The SF and Sac CSAs combine for over 12M people with a GDP over $750 Billion. The idea that an area this size is adequately served by an unbranched 77mi commuter line and about 130 mi of subway/regional rail hybrid ("we have one line from SF to the East Bay, we're done!") is not tenable, as is the idea that the three major cities/employment centers of that region should not be all directly connected by rail.

      I believe your "actually useful tunnel" refers to a Dumbarton Tunnel, which is absurd. Dumbarton crosses shallow water with no shipping and existing rail ROW right up to the approaches where there is no difficulty building a viaduct bridge (already built 4 there before, easy) so there can be no justification for the expense of a tunnel. The Dumbarton Bridge sees only 80k passengers per day. SF-Oak crosses a major shipping channel and the Bay Bridge sees 300k pax per day. All of your arguments of "there is nothing there" regarding Oak/E Bay evaporate and ring hollow when you propose a major piece of infrastructure (an underwater rail tunnel!) connecting the not-even-second-tier regional centers of Fremont and Redwood City but depreciate the same for connecting the two largest downtown employment centers in all of Northern California.

      Note there should absolutely be a Dumbarton rail crossing, via bridge, ideally sharing tracks and a station in the vicinity of Newark with the Oak-SJ line for an "actually useful" cross platform transfer between lines.

    10. I've said nothing at all about the desirability or otherwise of an additional SF-Oakland rail connection.

      I do have opinions about "herp derp through-running Caltrain to, ugh, something herp-derp defer DTX for another 40 years and meanwhile hourly headways any six limited stop services are cool herp derp do some more studies herp derp and then a miracle occurs herp derp "standard rail, because standard" herp derp.

      Re Dumbarton: building useful things sometimes works differently from the way people first imagine it might, especially if they do anything more than glance at a map and pronounce.

    11. Richard,

      An additional BART tube as a rail connection would be useful, but would do nothing to address regional transportation needs. The mistake of BART was trying to be a subway and a commuter rail system, thus the spectacle of ripping up tracks to lay new tracks to get to San Jose when the tracks ripped up could have been running 'CalTrain East' decades earlier instead (only one example of many). There is no need to continue this fundamental mistake. A new BART tube for better core subway service is fine, but it doesn't allow use of existing track throughout the East Bay and beyond to provide service to DT SF, the place the most people want to go.

      There are rail lines with passenger trains running on them many times a day from Oakland right now, all of which would see vastly higher ridership if they went to San Francisco. Through running wouldn't connect to "something" as if it were vaporware, the infrastructure is there, just poorly used.

      I agree that through-running should not delay DTX, Caltrain should get to TBT as soon as possible.

      Dumbarton would be a very useful crossing for CalTrain and CaHSR, but there is zero justification for it being a tunnel, ever. Actually less than zero, a tunnel would be a negative transportation investment from the perspective of vastly higher upfront costs, higher operating costs FOREVER (paying to run, maintain, test, and eventually replace pumps, lights, fans, and fire alarm systems), and less operational flexibility (no crossover tracks to route around damage; incidental maintenance - like on those fans and lights - causing track shutdowns due to clearance issues, as opposed to fan maintenance at all or unavoidable maintenance like signals happening trackside in the open air without blocking trains).

      You are completely right that PAX should be trenched rather than tunneled until DTX starts at 4th&King. Using existing ROW and tracks/stations in open air are much less expensive than tunnels. How could you possibly justify a tunnel at Dumbarton instead of using existing ROW, at grade, with not even a trench needed? PAX runs right through a city, the Dumbarton ROW runs through absolutely nothing you need to go under to avoid, and except for the burned out trestle the entire ROW is intact down to the ties and rails. Why would you waste money trading cheap ROW upgrades and a simple new bridge for expensive new tunnel? (You honestly sound like a transportation-industrial complex consultant on this one, advocating for the more expensive option when the easy cheap one is right in front of you.)

  11. @Drunk Engineer, as I previously commented there are three rail lines heading out of Oakland, plus 1.5M people and 0.5M jobs in the inner East Bay alone. Include everything beyond the Oakland Hills and possible conversion of BART's Concord and Livermore lines and there is plenty to connect to, both in the operation sense and the demand sense. Close to 300,000 people cross the Bay Bridge each day (or did before Covid) which is about 20% more than enter SF from the south on 101.

    Would you oppose additional Hudson River tubes (ARC, Gateway) because the Portal Bridge has only two tracks therefore "there is nothing in NJ to hook into"?

    1. "possible conversion of BART's Concord and Livermore lines"

      No no no just stop it!

      Sure, the Dublin line should never have been built. But it's there, glorously sited in the middle of a freeway, with civil structures designed around BART train dimensions and BART train loads; deal with it, try not to throw a cent more than absolutely necessary at it, we tried and tried and tried to stop it, we couldn't, it's a black hole of capital misallocation and endless subsidy, but it's there, it's not going anywhere (hah hah, literally!)

      As for the Concord line? You know, the most heavily used section of BART outside the SF-Oakland-Berkeley core? Stop with the nonsense crayons! (Some day the Hawyard Fault will go and the Oakland hills tunnel will collapse or be years out of service -- the answer will be to rebuild another tunnel for BART trains from the BART tracks on one side of the former tunnel to the BART tracks on the other side.)

      BART's here. Deal with it. Don't repeat mistakes and don't build more of it. It carries hundreds of thousands of riders. It's essential. It's much better operated than you may think, especially given the infinite insanity of local politics. Make the best possible use of it. (I'd start with halving fares in the cost-effective, suburban-subsidizing core urban areas, and tearing out the pure-Cubic-rent-seeking-private- profit faregates, myself.)

    2. I agree with you that BART is not nearly as bad as some of its detractors claim (the "it's not standard gauge" red herring being particularly bad since no one seems to claim that the Toronto Subway or the Market-Frankford line in Phila. are somehow flawed).

      Civil structures designed for BART should have no trouble with M8 size equipment used by the two busiest commuter rail services in the country.

      Saying that Concord is the most heavily used line outside of the core isn't terribly useful, its a little like saying "the warmest place north of Montana in the winter" - guess what, it's still cold. The ridership outside of the core is abysmal. If Union Sq were Times Sq then the farthest reach of the NY Subway (the one tiny little line all the way to Far Rockaway) would barely reach Walnut Creek, let alone Concord, Pittsburg, or beyond. Out that far is commuter rail territory, trains should be stopping once in DT Oakland with transfers to local destinations instead of stopping places like Rockridge or W. Oakland. Let BART focus on the core between Millbrae, Richmond, and maybe Hayward, focused on SF-Oak, as a subway should be. The NY Subway took over many mainline rail lines that were too close to the city in the early 1900s, no reason an alternative service cannot take over BART lines too far from it.

    3. I've nothing else to say about this.

      There's nothing stopping anybody from ignoring what actually exists in the real world, and enjoying a box of crayons. I'm not going to be reading it or writing about it though.

      Meanwhile I'm free to ignore hideous political and institutation and corporate reality and posit perfectly simple and feasible and obvious and cost-effective things that won't be ever be considered let alone built, but not because I choose to fail to observe the actual built environment around me. I prefer my imaginary model train set over the alternate-history ones, especially having watched too many totally fucked-up actions taken to create current-day reality, but that's just a personal choice, and one without much traction.

  12. A new Caltrain station in the Bayview: Here’s the plan

    SF planners are starting community outreach about building a new Caltrain station somewhere in the Bayview neighborhood to restore service to the historically underprivileged neighborhood for the first time since 2005, when the Paul Avenue station was decommissioned due to low ridership.

    The City is studying 3 potential station locations at Williams, Oakdale, and Evans avenues. In surveys dating back to 2005, residents had indicated support for the Oakdale location, but planners wanted to see if that still made the most sense for the community. The study will consider transit, bike and pedestrian connections and opportunities for new housing surrounding the new station.

    If all of The City and Caltrain’s plans come to fruition, the longtime commuter railroad would look more like an urban rapid transit line, with stops in the Financial District, South Beach, Potrero Hill, Bayview and Bayshore, connected to communities down the Peninsula.

    1. SF Department of "Planning".
      SF "County Transporation Authority".
      We are truly blessed, here in the best of all possible worlds!

      https://sfplanning.org/project/southeast-rail-station-study (there's nothing here, other than an implied and unbreakable promise to incinerate plenty of cash on Moar Studies.)

      Moar studies! MOAR!

      Here are some studies from 2005 and 2014 (don't be fooled by the "2019-03" in the URLs) to get you started:
      BAYVIEW-OAKDALE CALTRAIN STATION STUDY / Design Feasibility Assessment and Station Concepts Final Report / February 2005 / Prepared for The San Francisco County Transportation Authority by HNTB Corporation
      (If you find yourself looking for a broken link to "https://www.sfcta.org/BayviewOakdaleCaltrainStudy-Final.pdf.pdf", the above is link is that file, at least until some Web Designer breaks all the links yet again.)

      Caltrain Oakdale Station Ridership Study / March 2014 / The collaborative efforts of many individuals made this study possible. Gratitude and thanks to ..." (Excuse me, I need to pause ... I'm coming over with quite the kumbaya feels; gratitude floods my soul, especially but not exclusively towards "HNTB Corp.")

  13. Maybe $7-10 billion (or more) for DTX + PAX is just not worth it. There's undoubtedly a ton of bang in serving the financial district directly from the peninsula, but wow that's a lot of buck, so much that maybe the bang-for-buck value just isn't there. If that's where it ends up, better than waiting for another 40 years, how about never?

    1. "Maybe $7-10 billion (or more) for DTX + PAX is just not worth it."

      Perfectly agreed.

      Which is why killing PAX and killing everybody in any way associated with it is the first step.

      What we have is a project and a project team and a corrupt consultant mafia operation that cannot be supported and cannot be allowed to continue. Agreed.

      That doesn't mean that a grade-separated connection from the existing Caltrain tunnels to the shitty mushroom farm basement under the $2.5 billion Transbay Useless Park in the Sky facility is completely unsupportable.

      Straw man alert!

      * Trench from slightly north of existing Caltrain Tunnel 1 under the 280 viaduct and grade separate 16th, Common and 5th streets. No PAX, no TBM, no problem.

      And no, contra TJPA/SFCTA lies, this does not imply suspending all service to SF for "several years".
      The construction phasing may in fact be comparatively straight forward, as these things go, and it is possible that service interruptions may be no longer than weekends-long.

      * Trench Mission Bay station, as shallow as possible, nearly entirely withing the existing Caltrain terminal site, NOT UNDER a worse-than-useless mezzanine UNDER a utility-crammed (so much sewer!) city street.

      The "joint development" returns of placing buildings on the entire existing terminal rail right-of-way while forcing the station under a street under a mezzanine are negative hundreds of millions of dollars. There is no way that this "other people's money" (bilk "transporation" budget several hundreds of millions in order to score maybe max ten million from a real estate developer) comes close to passing a laught test. (See also, absolutely everything to do with the Transbay Terminal, Salesforce Tower: absolute total unmitigated reaming of the public budget, out-of-control give-aways to the developers and construction mafiosi.)

      * Cut-and-cover TWO TRACK tunnel for the shortest feasible distance under Townsend Street -- very roughly just the block from Fourth to Third Streets.

      America's Finest Transporation Planning Professionals (inexplicably-unindicted ex-SFCTA executive director! ex Amtrak Capitol Corridor honcho! more of the same!) sand-bagged their three-track budget-inflating "studies" way beyond any laugh test. There's no way in 100 years that the "redundancy" of dead tracks in the very most expensive to construct sections could ever even remotely be commensurate with the tens of hundreds of millions of dollars of extra tunnel construction bloat.

      * Actively investigate actual technical feasibility and cost-effectiveness of relocation (laterally transferring load of crazy unused park in the sky) of PTG-"designed" service-killing structural columns at the entrance to Transbay. It may cost, it might be fiscally impossible, it probably is; or it might both be an enabler of higher sustainable train throughput and a disabler of three-track connecting tunnel insanity, perhaps paying for itself this way.

    2. A trench or tunnel anywhere near China Basin just seems like a really bad idea:


    3. That sea level rise map seems like a strong argument for a bored tunnel along 7th & Howard.

    4. Fortunately Highway 101 is safe (within SF, for a while, on this map, and for much less time than you may think) so you all get your wishes: no meaningful Caltrain ridership, ever, and lots and lots and lots of studies meanwhile.

      Enjoy the pyrocene!

    5. The sea level rise map is in no way an argument for a tunnel along 7th & Howard.

      First, it is not a sea level rise map, it is a map that shows flooding with *maximum* sea level rise in 100 years *and* a worst case storm surge. In other words it is not showing areas that would permanently underwater anytime soon, or permanently underwater ever.

      Second, if those blue areas actually precluded a tunnel then it becomes a map not to build DTX at all. Half of the TBT is "underwater" and unusable by that criteria, so there is nothing for the "dry" tunnel on 7th to connect to. To say nothing of the fact that huge swaths of the financial district, Mission Bay, and SoMa are unusable so there would not longer be any point to bringing Caltrain downtown since hundreds of thousands of jobs and residents are somewhere else.

      But its all academic since that level of flooding is unlikely to happen. The map represents worst case sea level rise of about 17mm per year for a century. There is more than enough time to build the necessary sea walls to protect low laying areas in SF (unless you think tens of thousands of residents and business interests representing tens of billions worth of real estate will give up and do nothing). Passing on infrastructure because of the chance of a disaster a century away would be like New York not signing the dual contracts in 1912 because Hurricane Sandy was coming in 2012. Even that actual flooding disaster didn't mean route choices in lower Manhattan were wrong - the New York Subway saw its highest postwar ridership in 2014-2015, just a few years after the Hurricane Sandy's flooding.

      The route for DTX, like everything else, should be decided based on what provides the best service to the highest demand areas, at the lowest reasonable cost, on a *day-to-day* basis. One-weekend-a-year festivals, special sporting events, and once-a-century disasters need not apply. (Note, this does not mean ignoring the risk and leaving out fire alarms or sump pumps, just that you pick the best route and mitigate the risks, instead of picking a risk free but lower-transportation-value route). In this light Richard's trench-as-far-as-you-can-by-using-ROW-to-4th-&-King with a station close enough to serve the booming Mission Bay, South Beach and SOMA neighborhoods is clearly the best DTX route.

  14. "Passing on infrastructure because of the chance of a disaster a century away would be like New York not signing the dual contracts in 1912 because Hurricane Sandy was coming in 2012."

    The new South Ferry station was done more recently (on 9/11 basically). And it wasn't very long after that for it to get demolished by Sandy. Fixing that destruction cost $340 million and 5 years.

  15. OK, so Transbay to Howard to Seventh is less completely infeasible and less totally insane than I imagined.

    In particular, the Howard to Transbay transition ends up potentially hugely less insane than the $4+ billion disaster that the "professionals" at PTG, Arup, TJPA, CHSRA engineered/sabotaged at the approach to the terminal platforms, where structural columns have been placed exactly and unfailingly where they will most restrict and inhibit train movement. Death remains far too kind a fate for anybody involved in any way.

    But it turns out one can almost make tatre au citron from this $4 billion pile of poisoned rotting lemons by zig-zagging under Howard Street instead. Using perfectly UIC-standard 1:12 turnouts curved to a perfectly European-legal 200m/333m curved geometry and with perfectly Germanic/etc-legal 7m spacings of curves and turnouts, it's actually possible to recover a maximally conflict-free throat configuration and hugely improve train throughput over disaster engineered by the criminals at PTG (who are still, as we speak, sucking down tens of millions of public funding for further "design" of their never-ending disaster. The First Law of Being Defrauded: Stop Throwing Money at Fraudulent Parties.)

    It's kind of amazing that it crayons out in any manner, but it kinda does.


    The real problem is in elevation, track vertical profile.


    1. ...


      The real problem is in elevation, track vertical profile.

      Ignoring (hah!) for now all Fun Utilities under Howard Street -- all of which would be equally messed with by DTX tracks heading down Second Street -- we encounter two massive horrible expensive obstacles.

      The first is that Moscone Center conference stuff and truck loading docks fill the entire block under Howard between Third and Fourth Streets. The base of the basement slab is something like 35 feet below the street, meaning any Magical Tunnel Boring Machine creating a magical Caltrain connection under Howard must place tracks at least 30 feet deeper, meaning ~40+ feet deeper than the platform track elevation in the $2 billion Transbay Terminal underground mushroom growing facility.

      The second hideous obstacle is the Fucking Central Subway below Fourth Street. It's very deep (about 100 feet down) and diving in order to pass under the BART/Muni station under Market. So any Magical Tunnel Boring Machine creating a magical Caltrain connection under Howard has to be more than 30 feet deeper, with that vertical proximity truly stretching it (but hardly unprecedented ... including just two blocks north!)

      Basically getting under the Fucking Central Subway under Fourth is tough. Getting under Moscone (Third to Fourth) almost comes for free given that.

      A bonus of the depth is that the Magical Tunnel Boring Machine is going to be deep all the way under the many parcels and (mostly low-rise) buildings under the curve from Howard to Seventh. Helpful in that way, but of course an evacuation/access morass.

      But ... maybe not impossible? Maybe? (Geotechnically? Zero clues.)

      It does require a steep downgrade. At least -2.75%, probably more. Oh, and that -2.75% has to continue through the sharp curves (mine 200m radius, PTG's 650 world-class "feet") which give an effectively 0.35% steeper "compensated grade" of -3.10%, which is steeper than Caltrain's own "design standard", but is hardly without precedent.

      Every single day of the year hundreds of trains, including Stadler KISS trains with lower power/weight than Caltrain's over-specified over-cost version, run up and down the 4.0% grade connecting to the new-ish "Löwenstrasse" underground level of Zürich Hauptbahnhof. (The "40" trackside sign visible at 9m41s into the linked video means "4.0% down grade".)

      So it may be vertically geometrically feasible as well. Maybe.

      A further challenge is that the dive to get under the Fucking Central Subway has to begin within the existing $2 billion Transbay Underground Mushroom Farm. In fact, we need to lose about five feet of elevation by the time the tracks exit. Well the slab underneath the mushroom farm happens to be exactly five feet thick. So quite the excavation into and through the sealed-up $2 billion basement under a useless park in the sky. Plus there are three (miraculously only three!) structural columns that need relocation, using Civil Engineering Magic.

      But they're talking $2 or $3 or $4 or $6 billion to build their stupid tunnels, so no problem -- that buys tons of Civil Engineering Magic.

      We do lose a station at Mission Bay. Too bad. Muni exists. $2 billion of criminally inept design and construction by criminal unprofesssionals has its real cost and not all eggs can be unbroken. The best we can hope for is criminal prosecution and ostracizing.

      But SF would still have a CBD station, 22nd Street and Bayshore. (And no, a below-grade station south of Seventh and Townsend is neither realistic nor desirable. It's basically "serving" a freeway hellscape, with no ridership potential in any way commensurate with costs.)

    2. Always appreciate your in depth analysis Richard! What about a station (around 4th) connected to the central subway and Moscone? Or is that too close to Transbay to make sense? In my infinite money tunneling fantasy you could have passing tracks so Caltrain can make the extra stop(s) and HSR(lol remember fantasy here) would be able to pass on by without getting slowed down. Agreed on 7th/townsend station not making sense today, but in a word where 280 stops at Mariposa and the freeway land is reclaimed and upzoned there could be a case for a station. Given SF/bay area politics I doubt any significant upzoning and freeway removal would ever be approved (can't mar those Potrero Hill views any further!) but we're in fantasy land anyways.

    3. Richard,

      You were completely correct above that the cheapest way to get Caltrain to Transbay is trenching the at grade ROW all the way to 4th & King. While 7th to Howard may be technically feasible, there is no reason to spend the money tunneling so deep when you can get basically a kilometer of free tunnel using the tracks alongside Townsend.

      Losing easier access to South Beach/Mission Bay is a big deal. There is no way to compare them to 22nd St or Bayshore in terms of ridership from residents and jobs served. You can say Muni exists, but the main way to get to Mission Bay on Muni is...

      ...the T line, which stops right at 4th and King. As you point out the elevation issues make a station connecting to the T under Howard difficult or too expensive. Taking the longer view, SF should someday have two primary subways under Market and Geary (meeting roughly at TBT) and two secondary: the T from Fisherman's Wharf to Mission Bay and the N going due east from the Sunset under 16th before turning NE under Brannan to get downtown. These two lines would meet basically at 4th & King, meaning from most major neighborhoods in SF you could catch a 1 seat ride to Caltrain and/or CAHSR - if you plan for a DTX station at 4th & King.

      Do Geary/Central Subway extensions/N-Judah subway plans exist? No, of course not, SF is stupid in its transportation planning. But the whole point of planning is to develop an integrated transportation SYSTEM, not just isolated lines or single stations, so selection of the DTX route and secondary stations needs to consider integration with existing or future Muni lines.

    4. Richard,

      This is great work and I would not be surprised that TBT columns cause throat issues, but I'm confused. The "not a problem" approach track layout you show in your PDF (202110-Howard) would seem to work equally well with a 2nd street approach (it even has dashed lines continuing that direction as well as the red lines continuing to Howard). It seems the issue at TBT is making the first 45 deg of turn to the south - once out of the existing station box it seems going 45 deg further south to 2nd or 45 deg west to Howard is irrelevant.

      Am I missing something?

    5. Hi, I don't want to hijack Clem's blog (install yer own damn wordpress software and start writing, loser) by over-posting comments here, but here I go.

      "What about a station (around 4th) connected to the central subway and Moscone?"

      Several things.
      First: Muni already exists. (And at this distance, so do feet.)

      Second: Mainline rail stations are big. I claim -- with perfect unimpeachable unrefuted immaculate authority -- that Caltrain should size stations for minimum 300m trains (12 car, 990 feet) but even BART-ish length 200m is big. A station near 4th would be only a couple train-lengths away from Transbay. Stopping every block doesn't work for buses, stopping every couple blocks doesn't work for metros, stopping every couple blocks doesn't work the regional rail.

      Third: Fairly aggressive vertical rail grades have to be used to get down and under the Fucking Central Subway. This sort of steep up-and-down is incompatiable with passenger safety on platforms, where you don't want anything rolling away. Also ADA illegal.

      Re 7th/Townsend: again just consider how long a 400m (or 300m, or even just 200m "8 car") platform is. 300m north is Bryant and 7th, the magnificent urban nexus of Fucking Highway 80 and the County Jail. 300m south is the SFPUC sewage outfall into Mission Creek and a near-future Amazon block-sized loading dock. And yes indeed, all of it soon enough inundated.

      Re: "Losing easier access to South Beach/Mission Bay is a big deal"

      Humans fundamentally can't reason about orders of magnitude, which is why this beautiful planet and untold billions of creatures that experience pain are doomed to misery and painfu extincion.

      South Beach is a something-or-other deal. Something other than "big". Whereas the SF CBD is actually, locally, a "big" deal. (Not "Tokyo" big. Not "Shenzhen" big. Not "Paris" big. Not "Seoul" big. But big enough, by super rough order of magnitude accounting, by our provincial standards.)

      Anyway, you don't "lose access". You ride Muni to somewhere around First and Mission, or you ride Muni to somewhere around 22nd and Indiana. It's not perfect, but here we are. Here we are.

    6. Re "The "not a problem" approach track layout you show in your PDF (202110-Howard) would seem to work equally well with a 2nd street approach (it even has dashed lines continuing that direction as well as the red lines continuing to Howard)."

      The massive, insurmountable, entirely self-inflicted problem is that the "dashed lines" -- those "designed" at a cost to the public of multiple hundreds of millions of dollars by Parsons Transportation Group, Inc. -- all run into each other, in the very worst possible ways.

      Again see Clem's "Transbay Update" from 2012 (2012!) for some background on "route conflicts" parallel arrival/departure paths and so on.

      Without some various degress of grade separation you can never have "a train can arrive at platform track at the same time as a train can depart from any other platform track", but you can get as close to that as is topologically possible. And, in fact, at the approach to Transbay, getting as close as possible to perfection at grade was easy -- a trivial exercise in primary school geometry -- and completely feasible.

      Until the "experts" (experts only in the USA USA USA!) at Parsons Transportation Group inc., ARUP USA inc., the Peninsula Corridor Joint Powers Authority. the Transbay Joint Powers Authority, Parsons Brinkerhoff inc. (now DBA "WSP, inc."), and the Californian High Speed Rail Authority (a fully-owned subsidiary of of WSP, inc.) became involved.

      Then it became completely infeasible.

      Because the World Class transportation consultants at PTG -- who are still, to this day, sucking down millions of public dollars a year for failure after failure after abject failure -- deliberately and with perfect AutoCAD-aided malice -- decided the position structural columns for a stupid aerial park exactly and precisely where tracks for trains and escalators for train passengers needed to go.

      I type "death is too kind a fate" often enough. That's because death is, indeed, too kind a fate.

    7. I updated http://www.pobox.com/users/mly/caltrain-hsr.blogspot.com/vertical-profiles/DTX.svg with the dive-dive-dive under Fucking Central Subway vertical profile. (The drawing is a bit busy, what with including four different profiles, but I think one can spot the relevant addition easily enough. Full sub-metre geometric info in "tooltip" mouseovers of the various lines and curves as usual)

      It looks like a roller coaster, and it sort of is by US Olde Tyme Commuter Railroader standards, but again it's hardly without precedent, and violates no first world (ie not Olde Yyme US Commuter Railroad = copy of US freight railroad "standards") in getting there -- again compare with with the 3.7% dive to get into Zürich's beautiful newish underground station platforms and the 16m dive the Weinbergtunnel takes under an existing train tunnel on the other end.

      As a bonus for the "Mission Bay must have a station regardless of cost" fans, the grades can be jiggered to have a nice 350m long level section from south of Townsend Street as far as the bucolic sewer outfull and garbage truck parking lot (soon to be an Amazon warehouse and trucking zoo) before climbing at a 100%-Caltrain-"legal" 2.80% (3.00% "compensated" for curvature) to match the EXISTING tunnels.

      You're welcome.

    8. PS Here's some nice Zürich Löwenstrasse / Weinbergtunnel project info and construction porn -- in English even, delivered to a dismal dispiriting we-don't-do-foreign-things-like-that-here-don't-lecture-us USA-USA-USA do-nothing know-nothing APTA audience.

    9. A Seventh and Townsend (Townsend-to-Channel) station turned out to be less totally utterly batshit insane than I suspected. (Less technically insane to sketch out that I imagined in advance -- but not in any way less bat-shit insane for real-world rent-seeking limitlessly-corrupt pig-fucker consultants to actually "design" or construct, especially given the actual real-world adjacent land uses.)

      Anyway, here's your pointless multi-hundred-million-dollar intermediate station, laid out in sub-centimetric detail. As always, you're welcome, and you know where to send the invoices. It's actually pretty cool, TBH! (That "cool" assuming we can get the Seventh Street-via-Howard Street-to-Transbay Caltrain line under the FUCKING CENTRAL SUBWAY at all, which honestly I don't know, lacking geotechnical details, but whcih I suspect is ... (whisper the word!) ... not infeasible.)


  16. At the latest TJPA ESC meeting they changed the DTX configuration to be 2-tracks almost all the way to the station i.e. until "approximately mid-way between Harrison and Folsom Streets" saving $200M in cost (yay!). Unfortunately, at the same time they propose expanding 4th/Townsend to have 2 dedicated outside HSR platforms as well as a center Caltrain platform serving only 2 tracks. This adds $150M in cost vs. the previous 3-track single-platform Caltrain-only design. Does anyone understand why they want HSR to serve 4th/Townsend??? It makes no sense to me, especially for $150M extra cost.

    1. "Does anyone understand why they want HSR to serve 4th/Townsend??? It makes no sense to me, especially for $150M extra cost."

      I think you answered your own question there.

      "Oh hi! We are WSP inc., DBA the "public" agency California High Speed Rail Authority, a wholly-owned subsidiary! We have An Idea! Perhaps you can study our idea? Perhaps you can approve our idea? Perhaps you can undertake many cycles of design revisions of our idea over a decade or two? Perhaps there will be many Stakeholder Workshops. Workshops are fun! Anyway, it is an idea, and it comes from people who know how to spend other people's money, and expensive shitty ideas from serious people who consume serious amounts of money deserve study!"

      "Oh hello, sure we can, WSP inc., DBA the "public" agency California High Speed Rail Authority, a wholly-owned subsidiary -- we are PTG, inc., DBA the "public" agency Transbay Joint Powers Authority! We just so happen to be tottaly super awesome at studying utterly shit ideas. In fact, we have come up with some of the shittiest ideas anybody ever has any time in history, anywhere on the planet, and cast them into $2 billion of concrete. We just LOVE shitty ideas. In fact, we have an pretty-much perfect and long-term record of only promoting the shittiest ideas ever. That's how we roll! That's how we land the big "competitive" "public" contracts at "public" agencies, after all. We're not as big and as shiutty as WSP, but with time and effort and lots of other people's money, well, a girl can dream. Anyhoo ... perhaps there are other super shitty ideas that we have not yet billed anybody for. Do you have a new shitty idea? Great! We'll be happy to study your shitty idea. We will bill the public a few thousand dollars an hour so that PTG inc. can study shit from WSP inc. Also, in addition to your Stakeholder Workshops, how about several rounds of Executive Committee meetings of Truly Important Revolving-Door Figureheads? Does that sound like a good idea, WSP inc.? That sounds like a lot fo fun. We think it sounds like a good idea!

      Let us study all of these fine ideas! Together!"

  17. They need a station in "Downtown" San Francisco but closer to LA in order to meet 2 hour 40 minute trip time requirement. 4th/King just happened to be closer to LA than the Transbay...

  18. (broken into several comments due to blog limit)

    All of the valiant efforts by Clem, Richard, and others to put lipstick on the pig that is the Salesfarce Transit Center seem to hinge on one big assumption: that intercity trains to San Francisco (CAHSR, etc) must come via the Peninsula. I submit that if a new transbay link is going to be built anyway, intercity trains absolutely should not travel on the Peninsula, and Salesfarce can and should be written off. We can build an S-Bahn/RER/Crossrail for the Bay Area, and make the most of HSR, by cutting our losses on these headaches.

    "Blended" service on the Peninsula will permanently hobble both intercity and regional rail. Caltrain, as Clem argues, ought to become a high-frequency regional rapid transit line a la BART (again, as Clem argues, this need not include ripping up the existing tracks and laying wide-gauge tracks). And more than 4-6 intercity trains per (peak) hour to San Francisco CBD ought to be possible, instead of stopping some at Diridon Pan-Dimensional. Problem is, on the Peninsula, these goals are mutually exclusive. Caltrain obviously needs the ability to run express trains of some sort in order to be viable (witness the surge in ridership after Baby Bullet was introduced). The line is just too long for many people to want to ride local-only the whole way. Meanwhile HSR needs a minimum of intermediate stops. Thus we end up with at least three speeds of trains (I mean this in terms of average speed due to different numbers of stops--they probably all have the same maximum speed of ~110 mph): HSR (1-2 stops max between SJ and SF), express (either 5-6 stops a la Baby Bullet or something like Clem's Silicon Valley Express pattern), and local (all stops). Three speeds means we'd be constrained somewhat even if the whole corridor was four-tracked--which we now know is not going to happen. Given that large parts of the Peninsula won't be four-tracked, we will be *very* constrained in terms of passing, even if the Salesfarce bottleneck is removed by implementing the Deutsche Bahn through-running concept. Even if the whole Peninsula is grade-separated with level boarding, and even if a clever service pattern can be designed to make maximum use of passing on the corridor's four-trackable segments, it will continually run into problems because long-distance trains from LA or beyond won't always be possible to schedule as tightly. A few minutes of accumulated delay on a train from San Diego will be enough to foul up the delicate dance at rush hour. This is why DB would never run a bunch of ICE trains in mixed traffic with a high-capacity regional rapid transit line with several stops, like the Munich S-Bahn central spine or the S-Bahn tracks of the Berlin Stadtbahn (running in mixed traffic with low-frequency RegionalBahn trains, such as on the Stadtbahn's other set of tracks, is a different story). The "blended" Peninsula concept might have been workable when it was assumed that the whole corridor would be four-tracked, but now that we know it won't be, this concept would be detrimental to the future of both regional and intercity rail in SF and on the Peninsula.

    1. This comment has been removed by the author.

    2. Salesfarce is a horribly-placed station, requiring glacial speeds along ridiculously tight curves and/or a lot of demolition in order to be somewhat usable even if through-running is added (and its passenger-circulation problems are probably just unfixable). The details of why Salesfarce is a horrible station have been exhaustively covered by Clem, Richard, and others, so I won't go deeper into it. We keep trying to find minimally-terrible ways to make it work for the simple reason that there doesn't seem to be a better option in the SF CBD--and indeed there is not, if long-distance trains are required to travel on the Peninsula and foul up regional trains. As Richard pointed out in the comments, a minimum of four tracks are needed at the CBD station even if all of them are through-running, else regional trains will be constrained by HSR's longer dwell times, and there just doesn't appear to be any other site that can achieve this.

      Long-distance trains should enter the SF CBD from a transbay tunnel, as proposed by Onux and others here. For starters, even if the blended Peninsula plan had no problems, this is the most direct way for HSR to get from the Central Valley to the SF CBD: a direct alignment Altamont Pass->Dublin Canyon->Oakland->SF, with BART transfer at Bay Fair and ideally in Oakland (I have some more thoughts on doing this with minimal tunneling in Oakland). Yes, this would require re-gauging the BART Tri-Valley line, 10ish miles of tracks, but it would not require demolishing the thing and re-pouring a ton of concrete (if desired, a straighter and faster Dublin Canyon alignment with some viaduct and/or tunnel could be built as a future improvement, but it's not a requirement). I also submit that doing this would make the Tri-Valley line a much, much better regional rail corridor (at ~5 tph it would not be nearly as difficult to blend this corridor with HSR compared to the Peninsula). The northern San Joaquin Valley generates a huge and growing amount of commute traffic to the core of the Bay (Dublin/Pleasanton is the highest-traffic BART station outside of SF/Oakland/Berkeley/Daly City!), but this BART line is a meh option for commuters today because it takes a long time to get to Oakland/SF (45 min Dublin/Pleasanton to Embarcadero) due to the large number of intermediate stops and low top speed. The Valley Link program is trying to extend this corridor with a DMU line and a forced transfer a la eBART, but it will be limited in its effectiveness by the long travel time and the transfer. A 200kph standard-gauge electrified line with fewer stops could combine the BART Tri-Valley line and Valley Link into a highly effective regional corridor. West Dublin and Castro Valley (two of the weakest stations in all of BART, just a stone's throw from other stations) could be eliminated. After the Bay Fair transfer, the line could follow the existing Oakland Subdivision ROW (there's room for another two-track viaduct next to BART) without stopping again until downtown-ish Oakland (a new transfer station south of I-880 near the soon-to-grow Howard Terminal area and the soon-to-grow-if-they-demolish-980 area to the north?), then continuing into the new transbay tunnel. This would unlock the maximum value for high-speed rail (minimum travel time from SF/Oakland to Southern California) and for regional rail on Northern California's highest-growth commuter corridor.

    3. I disagree with Richard that there is nowhere for standard-gauge regional trains to go in the East Bay. As already stated, I think some should go via the Tri-Valley to the Central Valley, for starters. I also think the eastshore ROW (currently used by Capitol Corridor) presents some options. Capitol Corridor's long-term vision found that it is feasible to build passenger-only tracks in this ROW. For one thing, there are relatively dense residential neighborhoods in west Oakland, Emeryville, and west Berkeley that are far from BART and inadequately served by today's meager level of service (and where would riders even go today? Jack London, to transfer to nothing?). Beyond those areas, the service could branch and serve several areas of western Contra Costa and the North Bay which currently clog our skies with smog from their idling cars. There is clearly commuter demand in these areas: after Dublin/Pleasanton, El Cerrito del Norte (first station you come to on I-80) is the next highest-traffic BART station outside the core. The wBART proposal imagined extending BART to San Pablo and Hercules, and possibly then on to Vallejo. Like the Tri-Valley line, extending the Richmond line was found to make no sense because the long travel time (due mostly to the high number of stops as it meanders through El Cerrito, Berkeley, and Oakland) will limit ridership from the farther-out areas. However, if the western Contra Costa line feeds into a few-stop eastern shore line directly into a transbay tunnel, it would be much more appealing to forego the freeway. A Carquinez Strait bridge to Vallejo, then roughly following the freeway to Fairfield and Vacaville, could also be a better route to Sacramento than Capitol Corridor's vision of a horribly expensive Franklin Canyon tunnel followed by a Carquinez crossing at Benicia. Not only is Vallejo a larger market than Benicia, but the modern freeway bridge projects at both crossings have shown that Vallejo is far less geotechnically troublesome. Other than western Contra Costa/Vallejo/Sacramento, another branch could cross a new San Rafel bridge (the existing one is reaching the end of its life, and Marin politicians are already talking about rail on the replacement bridge). It probably should not just through-run SMART trains to SF because of their diminutive length and massive number of grade-crossings, but a station near downtown San Rafael could be a big improvement on the Golden Gate Ferry's 45-minute travel time, making transit more attractive for Sonoma/Marin commuters and reducing vehicle traffic to SF.

      With intercity rail entering SF from a transbay tunnel, it's possible to make the two problems of the existing plan (Salesfarce and the "blended" Peninsula corridor) go away. This is probably going to be the most controversial part of my plan, and I haven't seen it suggested elsewhere--possibly for the reason that I'm flat-out wrong and it's absolutely stupid, so feel free to say so. My plan for the SF CBD includes two parts: a six-track 400m terminal station for intercity trains beneath the low-rise portion of Embarcadero Center and Embarcadero Plaza (just north of Sacramento St, from Front St to Embracadero), and a through-running S-Bahn/RER/Crossrail-style regional rail link with 200m platforms which would take over the upper level of the Market Street Subway (currently Muni Metro) from Embarcadero to Powell, then curve south along 7th St to join the existing Peninsula ROW at King. The transbay tunnel would reach land beneath the Ferry Building, just north of the existing tunnel's transfer structure, and would branch between the terminal station and the crosstown tunnel. If the transbay tunnel has a total capacity of 24tph (leaving some slack due to merging regional and intercity trains, compared to the existing transbay tunnel's capacity of 30tph once current projects are completed), perhaps 16tph could be crosstown regional rail and 8tph could be terminal intercity rail, but the exact mixture requires more analysis.

    4. First, the terminal station. While I agree with Clem that it's not ideal to park trains in the CBD for longer than it takes to load/unload them, there isn't a better option that doesn't involve sending intercity trains down the two-track high-frequency regional corridor between SF and Bayshore. Mixing a lot of regional and intercity in the transbay tunnel is doable because there are no stations in the tunnel: all trains are going the same speed without stopping, so there is never any need to worry about passing at the right place at the right time. But sending intercity trains down the Peninsula line and trying not to get in the way of regional trains as they make several stops is a nightmare that would bring us back to permanently constraining the capacity of both systems. Freed from the need to serve through-running regional trains, a six-track station is perfectly capable of turning all the terminal trains that SF is ever likely to need. The Embarcadero Center site is a perfect fit with a Market Street landing for the tunnel. The low-rise portion of the site, plus Sacramento Street, gives us more width than the Salesfarce basement. With modern secant-pile or diaphragm-wall construction methods, the station walls could be built within a few feet of the buildings on either side, giving plenty of space for 3 island platforms. As for the low-rise buildings on the site today, it is a retail area which is completely dead whenever the office buildings aren't full. The developer would probably jump at a public-private partnership to redevelop the low-rise portion of Embarcadero Center as San Francisco's central rail station, driving far more traffic to the site. Like Europe's central stations, it would become a premier retail location in the CBD. Digging up the much-despised Embarcadero Plaza (ex-Justin Herman) would have few opponents. The surface of that area could become a far more appealing public space and entry to the station. One obvious concern is sea-level rise. However, the Port of SF's existing Embarcadero Seawall Program is going to be necessary in this area whether this station is ever built or not. The entire CBD is built on landfill, and the existing seawall must be rebuilt for both seismic and climate-change reasons, station or no station. Some parts of the world will be surrendered to the waters, but the CBD of San Francisco will not be one of them, period. As a final note on this station, BART connectivity to intercity rail would be better: at their nearest point (Drumm St), a pedestrian tunnel from Embarcadero station to the terminal station would be shorter than the proposed tunnel from Salesfarce to Montgomery.

    5. The toughest sell is obviously converting the upper-level Market Street Subway from Muni Metro light rail to regional heavy rail from Embarcadero to Powell, but I really believe it is the best option for several reasons.
      * The transbay capacity bottleneck, the place where new transbay capacity actually needs to go, is to the very center of the SF CBD, the area around Embarcadero and Montgomery. Every year these two stations *each* serve *twice as many* passengers as Powell or Civic Center, and laughably many times more passengers than any other stations in the system. A new heavy rail line which misses this core-of-the-core area is a waste of money. Yes there is growth in SoMa and Mission Bay, but it pales in comparison to the inner core. Proposals for a new transbay tunnel often show a connection to the Peninsula line south of I-80 (missing the inner core) or to a new Geary line via SoMa (also missing the inner core). These must be fought at all costs. A through-running line via Salesfarce is somewhat acceptable by this criterion, but it still barely grazes the inner core. Market Street between Embarcadero and Powell *is* the inner core.
      * Muni Metro is a pitifully weak utilization of a heavy rail line with 700-foot platforms, which is what the upper-level Market Street Subway is between Embarcadero and Civic Center. Muni trains are 75-150 feet long (225 feet in the absolute maximum case of shuttle trains that don't run on the streets). Their frequency is also deeply constrained by the need to weave several street-running lines together in the tunnel without a total meltdown. A heavy rail line through the highest-traffic core-of-the-core is just sitting right there (it's been sitting there for 50 years!) and it's being shockingly underutilized. It's the equivalent of using an entire transatlantic fiber cable to type out Morse code by hand, and nothing else. Even using the New Muni Metro proposal's "M Market" concept to separate Muni Metro from the street-running trams, the trains still could be no longer than 284 feet (or 425 feet if West Portal and Forest Hill are rebuilt, as 425 ft is the length of the stations on Market St west of Civic Center). The core-of-the-core needs maximum utilization, even if it means Muni streetcar passengers have to transfer. The transfer can be made less stupid than it currently is.
      * If we are going to build a parallel east-west tunnel through the CBD anyway, it would be much cheaper to do so with Muni-sized stations than with 200m heavy rail stations. A few ways to do this come to mind. The cheapest would obviously be to not build a parallel tunnel at all: terminate Muni Metro at Civic Center with a quick transfer to BART (no more walking all the way up to the mezzanine just to come all the way back down). If we don't want to force a transfer, Muni could turn south under 7th, then east under Howard or Folsom, with a transfer to Central Subway at 4th, then north to terminate at Embarcadero or Montgomery. Alternately, if the Unity Boulevard proposal (removal of the Central Freeway and new BART station along Otis St) is implemented, Muni could be dug cut-and-cover under the new boulevard, with BART transfer at Otis and Peninsula transfer around 7th/Brannan, then run through SoMa along Brannan and then north to Embarcadero or Montgomery. A variation of this idea would have Muni turn north from Brannan onto 4th St where it would *become* the Central Subway. Muni could then have one long light-metro subway without any street-running mixed in. Any of these concepts (except terminating at Civic Center obviously) would provide new service from western SF to growing mixed-use areas south of Market St using light rail, while heavy rail focuses on the core-of-the-core along Market St.

    6. West of Powell St station, the upper-level tubes would (gently) reverse-curve inward to the center of Market St, then dive down between the lower-level tubes. They would bend south to 7th St and could have one or two stations in SoMa as desired (7th and Folsom? 7th and I-80 with entrances on either end? 7th and Brannan or Townsend or even King to serve both southern SoMa and Mission Bay? Open to ideas). The ROW would be trenched south of Townsend rather than bored, in accordance with basic logic, then join the existing ROW down the Peninsula.

      One obvious drawback is that the standard-gauge line would not be able to use the Stadler KISS rolling stock. At 15.5 feet, it is far too tall. This is unfortunate, but I don't think it should be a deciding factor. One rolling-stock purchase should not define what the system becomes for the next 200 years. As has been discussed here before, Caltrain ideally should buy single-level rolling stock for dwell-time purposes anyway, and this is feasible in the tunnel. The Bombardier Aventra trains purchased for Crossrail are just 8 inches taller and 5 inches wider than Muni's LRVs (in fact there is about another foot of width to play with in the tunnel beyond that, to match the width of BART cars). The Aventra has been produced with top speeds up to 177kph (same as the Caltrain KISS) and Bombardier markets it up to 200kph. If it has to be sold off, the KISS rolling stock will find buyers in a post-FRA-overhaul America. For starters, Metrolink in LA really ought to electrify its highest-capacity lines.

      The biggest obstacle of all would probably be agency turf. Muni can hardly be expected to joyfully give up its prime real estate under Market St, even if it gets a shiny new parallel tunnel as compensation. This concept could only happen with massive pressure at the state and federal levels.

      I'm curious to know what people here think of this concept. This came out a lot longer than I had in mind for a comment, but I don't know anywhere else to bring this up with a lot of people who are more knowledgeable than I am about rail in the Bay Area.

    7. Prop 1A 2704.09 (4) San Jose-Los Angeles: two hours, 10 minutes.

    8. "Prop 1A 2704.09 (4) San Jose-Los Angeles: two hours, 10 minutes"
      Prop 1A: nobody has ever given a shit, except to cash the checks.
      Forget it, Jake, it's Chinatown.

    9. "nobody has ever given a shit" The Peninsula billionaires and their reps do, you know, the ones paying for it.

    10. Wow, Bryan, lots of fresh ideas to think about. I don’t have an immediate reaction but appreciate the effort that went into thinking this through.

    11. "The Peninsula billionaires and their reps do, you know, the ones paying for it."

      Infrastructure plebiscites around here are driven and have always been driven by the rent-seeking construction and design mafias who profit from it. Shitstain shills like the "Silicon Valley Leadership Group" greenwashing front are happy to give their endoresement to anything that increases public-private wealth transfer -- just as dimwit "environmentalists" will endose anything labelled "transit" no matter how insanely ineffective or costly -- and your SLVG type organizations are very very happy to greenwash money from any number of corporate sponsors, but if you think Musk or Zuckerberg or their ilk were behind Prop 1A or will pay a cent for CHSR, well, yeah, sure, man.

      Nobody except a handful of blog commenters even remembers Prop 1A, let alone pays a millisecond of attention to it. Once the money started flowing all bets are off, as they always are, and WSP and friends were off the the races, spending as quickly as possible on doing as little as possible. Same old story every time.

    12. Bryan, to me the first chunk of your extensive and nice set of comments honestly comes down to just "The "blended" Peninsula concept might have been workable when it was assumed that the whole corridor would be four-tracked, but now that we know it won't be, this concept would be detrimental to the future of both regional and intercity rail in SF and on the Peninsula."

      First, this is a choice that Caltrain's limitlessly stupid and unprofessional staff and consultants have made. It's a choice that is not undoable, not yet, not close even.

      To me your proposal honestly amounts to "well gee Burlingame is mean. I guess because Burlingame city council (representing literally dozens of squeaky wheels, some with quite murky motivations) we are FORCED, yes forced to spend $50 billion [wild underestimate!] replicating the BART Dublin and Fremont lines, building new HSR stations somewhere in Oakland and somewhere in SF that isn't Transbay, building a non-BART tube for the non-BART parallel tracks, and somehow connecting this non-BART Dublin/Fremont line to something in SF, as long as that isn't Transbay or the Caltrain line."

      Well OK then.

      That sort of "I guess we have no choice but to spend unlimited amounts of Other Peoples' Money on truly stupidly placed Concrete Before Organization because Atherton is mean and oh no please don't throw me in the briar patch" argument is one that plenty of truly horrible people profit from and promote. Are profiting from and are actively promoting.

      I know you've spent a lot of time thinking about this, but I've spent 30 years watching many of the ideas you might like and some of which might once have been desirable and possible die, almost entirely through active agency malice by "public" agencies.

      The locations of quadruple tracking between SF and Redwood City needed to accommodate a stunningly good 30+-year of growth level of service are known and simple to identify.
      You write as if anything involving mutiple stopping patterns (hint: FEWER IS BETTER, as people OUTSIDE Caltrain have been harping on for decades) and different equipment performance and constrained rights of way and limited budgets is something nobody has ever thought about.

      The locations of the quadruple track between Redwood City and San Jose needed to avoid congestion and extensive (because longer station spacing means higher speeds means much longer overtake sectionns) quadruple tracking is also know -- build the third and fourth tracks between Redwood City and Fremont.

      I guess you think you might not die before the fifteen round of consultant-enriching Vision Exercises driven by Stakeholder Input from SPUR and the Millbrae City Council are concluded. I'm pretty sure I'm going to die before Caltrain runs anywhere in SF besides Fourth and Townsend under this scenario. I the context of the not-at-all-figurative death of the ecosphere that's coming with the inconceivable levels of loss and of suffering of tens of billions of human and non-human animals this is absolutely nothing, but it still pisses me off, just a bit, you know.

    13. Bryan, re your second comment chunk re some sort of non-BART Altamont line (but not HSR, because Pacheco is sacrosant or something?) taking over BART's Dublin line.

      The time to have done this (ie to not have built a BART-technology freeway-median sinkhole of cost and inefficiency) was 20 years ago. Plenty of us were all about that at then, but here we are.

      There was an excellent window of opportunity post Loma Prieta earthquake to reconsider whether the shaky elevated multiple-freight-railroad-ROW-paralleling BART Fremont line ought to be expensively reconstructed as-is or might have been advantageously converted to "conventional rail", perhaps marketed at the Greater San Jose Co-Prosperity Sphere S-Bahn and cross-platform interfacing to the tired old non-dotcom SF-centric BART some place South Oakland-ish, but that ship sailed decades ago.

      (Actually, that ship was actively and ruthlessly torpedoed, as these things are, by the usual people who do this sort of thing and are still in charge of ensuring the worst possible outcomes for all of us, ever time, forever.)

      Re Dublin line conversion: freeway medians are not -- under nearly every real-world situation, and in particular not here -- good or acceptable high-speed rail alignments. And if you're not running much higher speeds than BART's 130kmh in the median of Highway 580 then why are you bothering? This sort of thing always, everywhere in the world that anything is actually built by people who aren't morons, turns into a HS rail line off to one side sort-of parallel to a freeway, more of less, some of the time. Regauging the BART tracks (the tiredest local railfan fantasy there is) is the least of it.

      Anyway the proposal seems to be to turn the BART Dublin line into non-BART "conventional rail", extend it over the Altamont Pass (but not not no no no not destined for LA or Sacramento, just Tracy or something), duplicate the BART tracks from Bayfair to Oakland to SF with non-BART "conventional rail", and then somehow do something with those trains in SF, avoiding Transbay.

      The basic benefit of all of this that BART Dublin trains are removed from the critically congested West Oakland-Daly City segment of BART. That's a benefit for sure! Maybe if you could find room for a BART/non-BART transfer station ($$$) somewhere before Oakland then maybe some passengers would prefer this to direct SJ-Fremont-SF rides, and maybe you can remove a couple extra TPH from the BART core.

      Also super super super important! -- it's "conventional rail" so that's A Good Thing, in some sense, somhow.

      Other "benefits" include HSR via Los Banos to Fourth and Townsend in SF, and Caltrain maybe doing something 30 to 50 years from now in SF that doesn't involve Transbay.

      I can agree with you and I can see -- in fact I saw this three decades ago, it isn't visionary rocket science after all -- that there were potential non-BART ROWs and potential non-BART rail routes in the SF-Fremont, Tracy-Livermore-Fremont, Fremont-San Leandro, San Leandro-Oakland-SF corridors. But given the actual undeniable and very concrete facts on the ground of BART to Dublin, BART to Warm Springs, BART to the San Jose Flea Market (opening any year soon, promise!) what's left is a huge metastatic pile of Too Much BART feeding into Too Little Transbay Tube, and nothing left (rights of way, routes) that BART hasn't assimilated.

      AT THIS POINT in the horrible history of the decline of the planet I don't see how the resolution to the problem of "Too Much BART feeding into Too Little BART Transbay Capacity" is "We will make there be less BART so we can have Some Other Sort of Transbay, hooray". It's just too late for that, and it just doesn't make sense, no matter how much I personally may despise third rail power supply and weird track guages. "Too Much BART" is unalterable, so "More BART Transbay" is what we're going to get, sometime.

    14. @Byran I tend to agree with Richard that the benefits of a standard rail line duplicating/replacing BART in the East Bay are probably not worth the huge costs.

      I did note that your plan includes the interesting option of kicking Muni Metro output of the Embarcadero/Montgomery stations. I agree that Muni does not utilize these stations well, with many LRTs crawling through that segment during peak hours due to slow boarding and overcrowding. It might make more sense for Muni to go back to the surface east of Powell and turn back on Drumm St., for example.

      However, instead of using the Muni platforms for standard rail, I think they should use them as originally intended for a second BART line up Geary (at least to Masonic), with the existing transbay tube feeding both levels. Combining a new 4-track West Oakland station with a 4-track 2-level Embarcadero station, since both branch afterwards, would allow for much greater transbay tube capacity, maybe removing the need for a new crossing at all. As a bonus, the Geary corridor is one of the few in the bay area with sufficient density to justify a new subway line.

      Interestingly, MTC / AECOM studied a variant of this concept way back in 2012, but I guess decided it did not involve pouring enough new concrete... See page 34 of the following PDF presentation:


      This, to me, seems the most sensible direction to go in if transbay capacity is really the issue here.

    15. There is a lot to unpack in Bryan’s posts, so I will reply to the major points in several independent posts.

      First, HSR is entirely possible on the Peninsula. To begin, if you read all of Clem’s many excellent posts, you would know that there is sufficient room for four tracks along almost the entire corridor (the ROW is mostly 100’ wide). In fact, at the costs we are talking about it should be possible to buy enough land to add an *additional* four tracks to Caltrain: 46.7 mi to Diridon by 100 ft wide is 566 acres; if an acre $4M (assume 4 houses at Bay Area median of $1M ea) then the total cost would be $2.264B – I don’t argue this *should* be done but with the money being thrown around it *could*.

      Four tracks is plenty to accommodate HSR and Caltrain. Again, look through the blog and you will see how Clem has timetabled out service patterns where an HSR leaves just before an Caltrain express then leaves the corridor just before it catches the express ahead of it (in some cases these patterns need only passing tracks not full four tracking). Indeed, your example of Berlin’s Stadtbahn proves this. The Stadtbahn uses two tracks for S-Bahn, and two tracks for ICE (HSR) and RB/RE regional services. But this is actually three speed classes since the RB/RE make stops that ICE does not. In fact, the S-Bahn only goes about 30-35km/20mi from Berlin, about the distance to Redwood City, while the RB/RE goes 60-100km/35-65mi – compare to Blossom Hill at 80km/50mi from SF (all distances are straight line not track milage). In other words, the two “fast” tracks on the Stadtbahn carry the equivalent of HSR plus what would be an “express” service for Caltrain.

      Finally, since HSR can turn to Altamont at RWC, it is possible to only four track from there to SF, instead of the whole corridor. The local would go to RWC, while the express would be local on two tracks from RWC to SJ, while making fewer stops from RWC to SF to better match HSR speeds and stopping patterns.

      HSR on the Peninsula is entirely possible without hobbling itself or Caltrain.

    16. Transbay is not a horribly placed station. The ideal station would be Montgomery St BART at 2nd and Market (the busiest BART station; dead center for jobs in SF is Market between Spear and Battery). Transbay is 1st between Mission and Howard. This is less than a quarter mile from the ideal, just two blocks away. Your proposed site (centered about Sacramento and Drumm) is almost twice as far as the ideal centerpoint for an SF CBD station. Given that the site for Transbay was already owned by the government and didn’t require any takings or demolition beyond an existing transit facility that had to go anyway, I think that there was never nor will there be ever any reason to look beyond Transbay as the site for SF’s downtown train station. (Note that while undoubtably the right location, this doesn’t mean there are not other problem’s with Transbay’s design as built.)

    17. I have not argued directly for a transbay tunnel for HSR, although I would have no problem with HSR going to SF from Oakland, mainly because the East Bay and downtown Oakland are such larger concentrations of jobs and people than the other routes (SJ, or Fremont-RWC via Dumbarton) – a transbay crossing would also allow SF-Sac service using medium speed trains in about the same time as HSR going the long way around through the valley. In fact, given the Bay’s unusual geography, there is an argument that from Altamont HSR should branch to SJ and go to SF via both Oak and the Peninsula, so that all major areas of the Bay (DT SF, DT Oak, DT SJ, Silicon Valley) would receive intercity service. I will say going via Dublin Canyon is an interesting idea (I always assumed Niles Canyon due to legacy route) but you would not be able to re-use the BART ROW for HSR. The BART tracks appear to have curve radii of about 600m, which is only going to get you about 100kph, which is not sufficient for an HSR train to LA or beyond. Just as with Niles, you would have to bore a new tunnel through the hills for tracks that serve HSR.

    18. You are entirely correct that there is more than enough ROW for Caltrain East and other services (regional to Sac, Stockton, etc.). South of Oakland there are *THREE* separate ROWs (Oakland, Niles, and Coast Subdivisions), and the Niles and Coast both go to SJ. They have three to five extant tracks between them, and all have room to be double tracked (even, I believe, the Oakland Sub after what BART took).

      The Martinez Subdivision north from Oakland through Berkeley has been four tracked before (you can see the evidence in aerial images where it crosses streets like Gilman). From Richmond north there are once again two ROWs (the Martinez and Stockton Subdivisions) although they mostly cover the same corridor.

      Going east there are the Stockton and Tracy Subs through Pittsburg/Antioch while the Oakland Sub turns to go through Pleasanton/Livermore (plus the Niles Canyon RW, which dies, but it does double track most of the way to Pleasanton absent an HSR tunnel for Niles). Again, three separate ROWs on two corridors.

      In short there is plenty of ROW to support freight, commuter, regional and intercity service in the East Bay, in fact more than the Peninsula has or will ever have. There are four existing and easily eight future tracks that can feed a tunnel Oak-SF. Including branches there are seven existing and up to twelve future tracks without doing things like a new Carquinez Strait bridge or re-gauging BART through Concord and Dublin. For comparison, the two tracks coming from the North into Oakland carry 36 trains per day via Amtrak right now, while Caltrain was running 66 per day in 1992. As Bryan correctly notes the main thing needed is a short grade separated route through downtown Oakland.

      Richard’s assertion that BART has “taken all of the ROW” in the East Bay is unsupported by the evidence, as is his unspoken assumption that the travel needs of approx. 4M people in Alameda, Contra Costa, Solano and San Joaquin counties will be fully served by BART if it just gets two more tracks across the Bay. This is like saying the Long Island Railroad is unnecessary because the 7, E and F already cross the East River. (In fact, Alameda and C. Costa counties are almost the same size as Nassau and Suffolk counties, and are closer to the SF CBD to boot).

    19. Converting the Market Subway from Muni Metro is an absolutely terrible idea. The Market St Subway sees some 72k boardings per day over 3 mi. (at least, this data is from the 15 year old TEP). By comparison Caltrain gets about 65k over 77 mi. Do not let the cart guide the horse. Shorter more frequent trips always overwhelm longer trips, at every scale. Just as intercity trains/HSR should not drive commuter rail (for instance, Caltrain’s 65k/day, while dwarfed by not just Muni in SF but also AC Transit, is about double the total number of people who fly between NorCal and SoCal each day. Just as harming service on the Peninsula to fit HSR is not a good idea (HSR should be fit in where it works best for Caltrain, not the other way around) taking away a major urban subway for commuter rail isn’t either. Commuter rail gets the secondary route.

      You incorrectly say the Market St Subway is underutilized because it has shorter trains, but it is actually *better* utilized because it sees more passengers now. Utilization is not the same as capacity. The goal should be to increase both capacity and utilization of that route by making it fully grade separated through Park Merced, increasing frequency and train length, building a connecting network to feed passengers (Geary subway, Van Ness-Mission subway), etc. Moving it to commuter trail would increase theoretical capacity but kill actual utilization. If we assume standees for Muni it might not even increase theoretical capacity if we run Muni frequently enough.

      Addressing this by stopping Muni Metro at Civic Center for a transfer doesn’t help. Transfers are important for anywhere to anywhere travel in a well designed network (ride N from home in Inner Sunset, transfer to T to dinner in Chinatown) but should not be forced for travel to the largest destination (CBD). If at all possible all routes, of all modes, should through-run the CBD. This is also why Richard’s idea for a 7th & Townsend station with a stub loop of Muni is a bad idea; there is a Muni line (the T) running right past 4th and King that allows access to both Mission Bay/neighborhoods south of downtown and someday all major destinations north of it (Union Sq, Chinatown, N. Beach, Fisherman’s Wharf). Why force Muni to run inefficient turnback service or keep people from using Caltrain to get to Mission Bay (assuming his turnback loop only takes Muni from the north).

      In a perfect world the Market Subway would have four tracks/two platform (BART and Muni on the same level with cross platform transfers) and we could talk about Caltrain going underneath them for a more-perfect RER/S-Bahn design, but that didn’t happen. Even more boldly we would talk about yet another tube to bring the Muni portion of the Market St Subway across the Bay to provide the inner East Bay (with high ridership potential) a proper subway network (there are much higher priority corridors in SF first, of course). For all of the complaints of BART not having express tracks, it is important to remember that in distance and design, for most of the world BART *IS* the express tracks, taking riders from about 20km out using 2-3km spaced stations. What is missing in the Bay Area is the true urban core subway, i.e. one with stops on Mission at 20th & 30th, not just 16th and 24th, or one that goes up Telegraph and has a station under UC Berkeley, instead of following a freeway median to downtown Berkeley.

    20. "Richard’s assertion that BART has “taken all of the ROW” in the East Bay is unsupported by the evidence, as is his unspoken assumption that the travel needs of approx. 4M people in Alameda, Contra Costa, Solano and San Joaquin counties will be fully served by BART if it just gets two more tracks across the Bay. This is like saying [... always the NYC analogies, because that's the only thing most anybody in the US can imagine or has seen ...]"

      Dude, I'm about the biggest Oaklandish booster there could be, so quit with the "needs of the four million" bullshit aspersions.

      Anyway, the fact is that the good urban and suburban rail rights of way in the East (and now South) Bay have been assumed by BART. Sure, one could idiotically choose to build exactly parallel to BART largely (but never entirely = $$$) along BART's tracks in places where BART ridership and BART traffic is comparatively low, but really, cui bono? Not the downtrodden 4 million ... more like the rent-seeking 1%.

      Where new tracks are needed are where traffic density is highest. That's not San Leandro-Warm Springs and it's not Bayfair-Dublin. (Hint: It's MacArthur-SF.) Just because something looks like it could be physically constructed or because you're in love with the idea of non-stupid non-BART rail doesn't mean it's useful to build wrong things in the wrong places.

      We're had 50 years of stupid things being built in the wrong places, to the immense profit of sleazebags and the immense cost to the planet. The situation's utterly shit, but no need to dig deeper to make it worse.

      Link21 is a consultant's endless wankfest. The time to have done this (not "studied" it, again, again, again, again) was 30 years ago. Some of us tried. We failed. We're fucked. Let's not keep shovelling money at the same fuckers to get more fucked.

    21. "Transbay is not a horribly placed station".


      "Converting the Market Subway from Muni Metro is an absolutely terrible idea."


    22. "7th & Townsend station with a stub loop of Muni is a bad idea; there is a Muni line (the T) running right past 4th and King ... Why force Muni to run inefficient turnback service or keep people from using Caltrain to get to Mission Bay (assuming his turnback loop only takes Muni from the north)."

      It's a fantasy crayon track, connecting to nothing. The more fantasy tracks one draws, the more train people like. 𝄡 🎶 I am human and I need to be looooo-uh-oved, just like anybody else does.

      Anyway if it connected to anything, it would be to the Muni tracks that run down the middle of the Embarcadero and presently terminate in quasi-freeway median hellscape in the middle of King "Street".

      Maybe they run King-Fourth-Townsend instead. Maybe King-Fifth-Townsend (restore Fifth Street, you cowards!) Whatevs. Remember that three entire city blocks bounded by Seventh/Townsend/Fourth/King are free to put Muni tracks anywhere you like. Go nuts! Pencil in a spiral viaduct over a miniature golf course and a roundhouse -- it's a free country.

      Everybody on Caltrain is going to and from Transbay anyway -- always has been the case, always would have been the case (and probably never will be the case, because we need to build out a brand new ab initio some East Bay S-Bahn somewhere, somehow, for some reason, first.)

  19. One of the idiosyncrasies of BART is that it has no passing tracks, and therefore has no express service (it's "all stops, all the time"). BART admits that it can run more trains through the existing Transbay Tunnel, but needs more electrical power, more facilities in Hayward, more rolling stock, and better train control. These would seem to be inexpensive compared to new tunnels. Adding passing tracks to offer express service would solve the problem mentioned above about long commute times from Dublin/Pleasanton, and should be cheaper than many other upgrades that need underground solutions (and would pair nicely with running more trains through the existing tunnel).

    1. No, the idiosyncrasy of BART is that it is a metro with cancerous offshoots metastasised into cow fields and cow towns where it should never have run and has now completely consumed all resources and all rights of way.

      You also don't understand how express tracks actually work. Bypassing West Dublin or Orinda (or all of San Jose) does basically nothing except to reduce HEADWAYS (which actual people, beside the morons who "plan" Caltrain "service", actually care about) because BART dwells are short and BART train performance is good that you lose more than you gain, and because the far-flung out places where throughless people think they can bypass (on express tracks or by not stopping) are places where the trains have low ridership (again inappropriate metastasised metro service) and stations few and far between.

      Where express service is needed is where stations are close together and where express service adds service both in terms of frequency of service and of crowding.

      That's inner cities (real cities with any sort of density and street life, like SF and Oakland) and it's the inner urbanizing suburbs near cities with short station spacing and good ridership.

      That's why real metropolitan regions have complementary metros and S-Bahns, both serving the CBD cores with tightly-spaced high-ridership stations, but with the latter expressing through parts of the cities and innermost suburbs the metro serves better.

      But BART fucked the SF Bay Area over totally. Except on the Peninsula (= Caltrain) THERE ARE NO RIGHTS OF WAY FOR A NEW SF-OAKLAND S-BAHN TO SERVE. BART's taken them. They're gone.

      Is it worth building a new transbay connection for BART's far-flung exurban extension together with many miles of inner-suburban BART bypass tracks, probably extensively in tunnels (because there are no rights of way there either)? People argue that. Most of them are either morons or getting fat swilling with the transit-industrial complex, but some others argue it.

      Is it worth building a fantasy new transbay connection so Caltrain can run to Oakland and ... terminate because there are no rights of way to destinations that justify or even allow through-running? Some argue that, including our smart and charming host Clem.

      Anyway, where you want to run express trains you need to go insane on new tunnelled rights of way in the inner East Bay. You don't need to or want to run express trains where stops are few and ridership is lighter. BART doesn't need expresses and isn't going to run expresses.

      (A particular -- which is saying a LOT -- sub-moronic BART director pushed through limited-stop trains shortly after the insane SFO extension was opened, on the theory that whisking dozens of Imporant PG&E executive airport travellers a week from Orinda to SFO was a vital priority, but BART Operations canned that stupid experiment in short order: all downside, no upside.)

      The only place in the Bay Area to run express trains is between Redwood City and San Francisco. Fortunately, we (mostly, despite Caltrain staff's limitless stupidity and corruption) have the necessary rights of way for those extra tracks. Unfortunately, we have the sort of people who run Caltrain running Caltrain.

    2. I'm struggling to understand the case that there are no good rights of way for a medium-distance S-Bahn or intercity rail in the east bay, or that tons of tunneling is needed. I think it can be done with just 0.7 miles of tunnel in Oakland, though with a fair amount of viaduct.

      The Oakland Subdivision ROW has room for a second two-track viaduct from Bay Fair until just east of Fruitvale (already used by the BART viaduct, freight track is near-entirely disused, Capitol Corridor has thought about buying it in the past). West of the point where the Oakland Sub branches from the Niles Sub, a new line could still follow the BART tracks all the way to the East Oakland Yard if it's feasible to acquire several car-parts lots and rebuild the BART parking deck at Fruitvale. Alternately, the line could switch from the Oakland Sub to the straighter Niles Sub near San Leandro BART (would require acquiring one office building and one parking deck). The Niles Sub has room for two additional tracks all the way to East Oakland Yard as well, except possibly around a single-storey mini-storage facility near Fruitvale. Either way: westbound from the East Oakland Yard, after crossing the creek at surface level, enter a roughly 0.7 mile tunnel from Fallon St to Clay St, then surface just west of where BART surfaces. Elevated 400m transfer station on the straight segment between Jefferson St and Market St. Viaduct following the BART viaduct until entering new transbay tube.

      An S-Bahn is exactly what we need for farther-out areas like the Tri-Valley and near-Central Valley areas like Tracy, Stockton, and Modesto. The commuter flow from those areas to the inner Bay is huge and dumps barbaric amounts of carbon into the sky every day, and BART's franken-U-S-Bahn isn't a good option at that distance. I agree that BART fucked up by trying to have the same system be both a metro and an S-Bahn. I just don't agree that it's unsalvageable. The ROW is there and does not require miles and miles of tunnel.

      Closing West Dublin and Castro Valley isn't the main point of re-gauging the Tri-Valley line (though with 125mph trains, losing two stops does become a bigger deal), it's just easy savings because (in spite of being far-out where stations ought to be far apart) they are very close to other stations and thus have extra-low ridership. The bigger gain is from precisely what you said: express service in the inner suburbs where stations are closer together, with no stopping between Bay Fair and the new Howard Terminal transfer station. This would meet the criteria of adding service in terms of frequency and crowding: SF/Oakland-bound riders from the East Bay south of Bay Fair would switch to the express at Bay Fair, reducing crowding on the BART green line into SF/Oakland. The ability to transfer to an express at Bay Fair would also increase overall ridership and make the BART line more attractive to southern East Bay riders by shortening total travel time to the center.

  20. I've been updating http://www.pobox.com/users/mly/TTT-Howard-202110/ (Caltrain 22nd Street to Transbay via Howard Street) with more compulsive-level detail over the last couple weeks.

    It's frankly kind of terrifying how not completely and laughably infeasible this all appears to be.

    (That's not much or hurdle to clear, given that "not infeasible" doesn't describe anything any Caltrain, TJPA, MTC or CHSRA have proposed at any time in the last 30 years. Still, a girl can dream!)

    I'm personally very dubious of the cost/benefit of Mission Bay (Seventh and Townsend) station, but if somebody wants to pay somebody to dig it using somebody's money, well, it doesn't seem completely and laughably infeasible.

    Now if only anybody anywhere cared about lemonade from lemons, or, for that matter, cared about anything at all. If only!

    1. Another plus of 7th& townsend station is that you have an easy connection to the muni using the existing caltrain tracks and boom! no lost transfer due to shifting from 4th and king. It's not clear to me from your diagram where the tracks go below grade relative to the existing 22nd st station (as I doubt you'd entertain the Pennsylvania avenue tunnel). In the event of the Pennsylvania tunnel Muni gets to make it to existing 22nd st station. I can't take credit for this idea, I swear I read it somewhere.

    2. Can you add the envelope of the "Fucking Central Subway" and the Moscone Center to the vertical profile .svg as you have with the streets, overpasses, etc? Would help to see exactly where those fit in and why it is necessary to go so blessedly deep.

    3. Owen, nope, because I don't know that precisely (I was too depressed about the whole disaster to even grab real bid drawings of the Fucking Central Subway when they were grabbable) other than vague "sixty feet below grade" references in EIRs. There's little point to including fake or guessed information, so I generally don't.

      But it seems Roland LeBrun of San Jose did grab some drawings. Eyeballing this it seems that the FCS Top of Rail is about elevation -53 feet. But I still don't know how much deeper the bottom of the FCS bore lining lies. (Plenty of people do, and they should be free to chime in, and also to point me to a full set of civil and structural drawings for the FCS.) Anyway, I guesstimated that 40 feet vertical was an absolute totally tricky barest squeakiest minimum between elevations of Caltrain TOR and Muni FCS TOR, so I aimed for an elevation of about -100 feet under Fourth and Howard and tried to see what it would that to make that. Now with steeper rail grades than in the sketch I uploaded one can of course go slightly deeper, eg going from -2.86% to -3.15% (which is 3.50% "compensated" for the evil 198m curves getting into the evil TTT box) buys an extra 6.4 feet of depth. Is this enough? What is enough? Fucked it I know. But I gave it an honest try ... which is more than you'll ever get from TJPA or SFCTA on any matter, ever!

      The very most amazing thing to me was that anything can be salvaged from absolute disastrously unprofessional, incompentent, stupid and ignorant Caltrain/TJPA/CHSRA/PTG "concensus track alignment" and structural column placement. (Really, no joke, launch all of those people involved in any way at any of those "public" agencies or scraped-way-past-the-bottom-of-the-barrel consultancies on a rocket to the heart of the sun to boost Earth's total IQ. Don't bother supplying oxygen.) That the really really really really stupidly placed curves heading down Second can be twisted -- without torturing them! -- into an as-good-as-one-could-ever-hope-to-salvage layout of curves and turnouts heading down Howard is just dumb luck. With the emphasis on dumb.

      Re Anonymous "It's not clear to me from your diagram where the tracks go below grade relative to the existing 22nd st station". The whole point is to use the existing tunnels and tracks right up to the north end of Tunnel 1 (lying just south of Mariposa Street), starting realignment and trenching from that point. A fairly steep dive (I use 2.80% = curve-compensated 3.0%) is necessary get the tracks under Sixteeth Street which is the entire point of the exercise. Raising the road surface of 16th by a couple of feet maximum may be necessary, but that's all. The next constraint is getting under the concrete box sewer at the end of the Mission Creek channel -- the base of that is at elevation +0.34 feet and isn't a challenge given the headstart depth already achieved to get under Sixteenth. (SF Water Department has plans to change this, but any remotely competent public agencies would do so co-operatively. Right? Right.) I just go deep enough (TOR sketched in at -8.5m = -28ft) to get under various sewers and utilities, while allowing a trenched, NO-MEZZANINE, simple-access, hugely fire-safer = hugely cheaper open-air station. (And again, I don't really buy the cost-benefit of a station here. Nice to have, not nice to pay for, just move on.) Anyway, see my vertical profile sketch for deets.

      PS Yes the Seventh and Howard fantasy drawing includes bonus Muni tracks running along Townsend from Fourth to a loopy turnaround stop at Seventh. Nicer than a freeway median, for sure. So much crayonage! So many winnings!

  21. Can we please finally put the insane legal challenges to HSR to rest?
    Court of Appeal Opinion

    1. Well, the project is dead anyway. The Governor said there's no path to complete SF to LA, much less Phase 2. I haven't seen anything to suggest the federal government is going to bail CAHSRA out either (nor should they). It was a stupid idea from day 1, and the next Republican Governor will do everything they can to kill it entirely, as well they should.

    2. You're on the wrong blog mate

  22. In case you didn't see this, today's (Dec 7) update from trains.com --->

    Electrification of Caltrain’s line between San Francisco and San Jose, Calif. Is now estimated to cost $2.44 billion, some $462 million over the initial estimate and $129 million more than a Federal Transit Administration estimate from earlier this year, the commuter operator said in a news release.

    The 47-mile electrification project, which began construction in 2017, remains on schedule for completion in 2024.

  23. @Reedman: here's the electrification project cost overrun staff slideshow presented to the Caltrain Board at their special meeting to decide how to proceed. (The staff recommendation to settle with and continue with their current contractor Balfour Beatty was approved.)

    1. Your link is broken. Not that anybody needs to be incensed by seeing Powerpoints of Caltrain staff and perma-temp rent-seeking consultant mafias incinerating hundreds of millions of public dollars without consequence.

  24. News story about yesterday's Caltrain board action on the cost overrun (behind easily-bypassed cookie-based paywall):

    Caltrain board approves electrification cost hike

  25. As of half an hour ago, Steve Heminger has added to his considerable legacy of catastrophically incompetent, absolutely disastrous, unprofessional, stupid and ignorant (borrowing some adjectives from Richard, king of acerbic modifiers) public transit decisions in the bay area. His proposal to return the J church to the subway has been passed by the SFMTA board of directors.

    There were some legitimate accessibility concerns, which muni proposed to inadequately address with a street-level transfer to the N judah. It's unclear to me how much this will impact subway reliability given how apathetic headways are currently. Muni incorrectly claimed in their original blog post announcing and justifying the J church modification that J and N trains had to cross tracks of all outbound trains. There is a flying junction at Duboce portal. But returning the J to the subway will certainly impact reliability once the L taraval is brought back and headways are reduced to the deserved levels.

    1. This is insanely off-topic, but Muni's inept losers are lying when the claim that the Market Street subway can't accommodate the J, and they're particularly lying when they claim that today's "lack of congestion" in the subway is due to the removal of the J, instead of the fact that they're not running much service at all, for anybody, at any time and yet they're still running it truly horribly. There's just no cause and effect here at all, but these people can say whatever they like and they get away with it, always. There are no professional consequences, because there's zero professionalism to be found anywhere.

      Anyhoo, Muni's Market Street problems are due to:

      * Utterly batshit insane high/low platforms (convert everything, everywhere in the city to low-floor level boarding everywhere, low the stair gadgetry, improve dwells and accessability and timekeeping everywhere.

      * Utterly batshit insane coupled shorty "steetcars" from the 1970s with too many driver cabs consuming too much space and too much money.

      * Handfuls of motorists prioritized over thousands of transit riders, everywhere, at all times, throughout the city.

      * Not turning trains back right at Embarcadero when possible, avoiding huge deadheading wastes of service when trains run empty to and from "tail tracks". (Recall that the two stub platforms and simple scissors crossovers accommodated all of Muni's much-higher levels of service before the $200m million service-quality-reducing Muni Metro Turnback opened in 1998.

      * Shitty flaky badly-maintained power systems. (They shut down the system for thousands of hours more than once a decade to replace their overhead wire, for just one example, for God's sake. CONDUCTOR RAIL, people, rigid overhead conductor rail. Same goes for "emergency communication systems". Same goes for everything. It's eternal, it's crap.)

      * Shitty flaky badly-maintained unnecessary CBTC signalling systems. Hey, it wasn't needed, it doesn't work, give us a few hundred million more to replace it with something equally bad which isn't needed and won't work! And repeat every 15 to 20 years.

      * Simply neanderthal operation practices, systemwide. So little revenue service per payroll hour and per vehicle hour. So little. So little.

      Oh, look, here we are back in 1998 again!

    2. Sorry Richard, no can do. We can't just give transit priority at intersections, not without a multi-year long visioning and R&D exercise to develop a unique in the world system to replace beg buttons with LIDAR for the very special and unique needs of San Francisco. And we can't just buy new low floor trams, they'd be too long for the J which would be inconvenient, we'd have to replace a bunch of single point switches, and besides we just blew our capital budget on the Very Useful and Important central subway.

      Apologies for the off-topic thread. I don't know where the cool train kids hang out these days.

  26. @Clem - love your blog, very thoughtful. Have noticed a number of broken links - you may want to consider doing a periodic scan with a free tool, and keeping links current. I recommend you only look at "not found" errors, e.g. "404 not found" or "-1 Not Found server name..."

  27. Here is one example of a free link checker - can scan by page or whole blog: https://www.deadlinkchecker.com/website-dead-link-checker.asp

  28. Rolen LeBrun is being held in Santa Clara County jail, on suspicion of making criminal threats at a VTA meeting:


    1. Wow. I knew he was a little weird, but seriously? BTW, as of 12/23/21 he's still in custody, and his bail is now $350k, up from the $100k originally reported by the Mercury News.

  29. Speaking of tubes and grotesque levels of regional corruption, just a small reminder -- not that one is needed to anybody with a single functioning brain cell that is not involved in profiteering -- of how un-possible and expensive and risky and a Dumbarton trans-bay tunnel connecting to Altamont HSR isn't: https://www.kqed.org/science/10108/inside-the-new-tunnel-100-feet-below-san-francisco-bay

    1. I still don't get why a Dumbarton tunnel is supposedly so much better of an idea than just rebuilding the bridge in place.

    2. I mean - I get the arguments: The geology is known! Wildlife refuge! Environmental review! Toxic sediment at the bottom of the bay! NIMBYs! Plus, some SNCF study perhaps 15 or 20 years ago recommended it!

      But I don't get why the standard concerns of NFPA 130 bloat like excessive ventilation, emergency egress, "only one train per ventilation zone", and just sheer cost of tunneling in the Bay Area, don't make the bridge nevertheless the overwhelmingly clear favorite?

      It seems there's an unproven assumption that environmental review to rebuild the bridge will be held to the same standard of a new-build corridor. It won't. They'll probably be able to rebuild and restore it with an EA/FONSI, or perhaps not even that - since the corridor was never even officially abandoned.

  30. Owen,

    You are correct, there is no reason to build a tunnel for the Dumbarton crossing. See the comments to this post for details: https://caltrain-hsr.blogspot.com/2019/03/build-dumbarton-rail-tunnel.html

    The key points:
    -Building at grade or a bridge is always less expensive per meter than building a tunnel, everywhere in the world, barring obscene land acquisition costs or deep water, neither of which exist at Dumbarton.
    -Tunnels incur higher operating and maintenance costs than at grade track or bridges, forever (running lights and fans, confined space work vs access to cranes, etc.)
    -Tunnels have rail operations penalties that can be offset, but only by even higher capital costs
    -There is no issue with the wildlife refuge, since the Dumbarton tracks are not in the wildlife refuge, a simple check of plat maps shows that the rail ROW is a totally separate tract, and as you note, one that is still an active rail corridor as far as the FRA is concerned.
    -The tunnel actually has greater impact on the refuge since it would require excavation in the refuge for ventilation and evacuation sites.
    -The Dumbarton tunnel argument seems to boil down to "The Bay Bridge Eastern Span was WAY overbudget but the Dumbarton water tunnel was simple and cheap". This ignores the fact that a tunnel to carry a water pipe is a world of difference in cost and complexity from a passenger rail tunnel. It also cherrypicks and ignores examples of tunnels that were/will be way over cost (Central Subway, DTX, SJ BART) and examples of bridges that were not (2nd Carquinez Bridge in 2003 and - ahem - the current Dumbarton road bridge in 1982).

    Clem was a fan of the Dumbarton tunnel, as evidenced by the post I reference above, but I believe his most recent comment on the subject referred to a bridge at Dumbarton. Richard M. has myopia and remains committed to the idea of a Dumbarton tunnel even though in any other situation he would blame his position on the incompetence of the transit-industrial complex ("They are proposing an tunnel to drive up the cost even though there is already ROW with a previous bridge in the exact same spot!!!")

    Note that one cannot simply repair the existing bridge. The 1910 trestle is old and hasn't been maintained in decades. For modern rail operations, including electrification and potential HSR, one would need a new concrete viaduct bridge and new track infrastructure on either side.

    1. Bridge is not always less expensive. The CHSRA's 2008 study found tube option was less expensive than high bridge.

    2. @Drunk Engineer,

      Yes, bridges are always less expensive, absent extraordinary conditions (a hypothetical bridge to span the English Channel could have been more expensive than the Chunnel if it preserved current shipping conditions, for example). At some point water is deep enough to make bridge foundations and towers more expensive. None of these conditions exist at Dumbarton: the water is shallow, the span length moderate, the soil conditions known, and several bridges have been built before.

      Even in an absurd situation where a tunnel was somehow less expensive per unit length, a Dumbarton rail bridge would still be cheaper because the tunnel has to be longer than the bridge. The bridge can end at the *existing rail ROW at each end* while the tunnel has to continue to come up to surface level past that.

      The 2008 CHSRA study is junk, completely sandbagged to favor the Pacheco alignment. The cost of the Dumbarton Bridge *plus its seismic retrofit* would be about $365M in current dollars - it is twice as wide as what a two track train bridge would need to be. If CAHSR was estimating *four times* as much (even accounting for track and electrification) that is more a measure of the organization's incompetence than of bridging costs.

      I can't find the study anymore (CAHSR website is giving 404) but I'm not even sure what high bridge vs low bridge is supposed to be. The Dumbarton bridge has a clearance of only 85' which is low in anyone's book. No rail bridge needs to be higher and you could probably get the Coast Guard to buy off on a lower bridge since there is literally nothing to sail to south of Dumbarton so as long as the occasional barge can make it navigation is maintained. If CAHSR was assuming $1.5B for a bridge entirely at viaduct height (a dozen or so feet above water) and then ANOTHER $1.2B to raise it to 85' across a few hundred feet of channel with a few thousand feet of approaches at rail grades then they are even more incompetent than my earlier statement. No bridge of this specification IN THE WORLD has ever cost *$1B per mile* (my approx calculation of a 2.7 mi bridge with same main span as Dumbarton and 1% grades - far more conservative than needed).

    3. The Coast Guard will absolutely require the shipping channel be maintained, so a low viaduct is not possible. The 2017 Dumbarton Corridor study says the cost of the bridge at $2 billion. That study was for a limited service -- HSR probably would have required more.


    4. "The bridge can end at the *existing rail ROW at each end* while the tunnel has to continue to come up to surface level past that."

      Oh whoa, look, lookee here, here we are, once again, hallucinating useable -- practically useable, for real service at real headways at real operating speeds used by real trains or any use to real people with real origins and destinations on real timescales -- surface rail rights of way pretty much anywhere hereabouts other than those already occupied by BART (or by shit-tastic Caltrain.)

      They key verb being "hallucinating". Hallucinating.

      Maybe I'll grant you elevated rail from Redwood Junction to some location either west of 101 or west of University. OK, sure. But east of there? Newark, Fremont at grade? Whence? At, what? 20mph? 1tph? Going where? Serving whom?

      (OK. Serving rent-seeking consultants and "public" agencies who will gladly "study" anything, forever, as long as somebody's paying, which they will, forever.)

      You don't even have to bike around to see the lie of the terrain. Just look at a map! Online! For best results enable "Satellite [of Love] View"!

    5. It's not a shipping channel. The channel peters out to 15' deep a mile or so south of the bridge, and to 10' or less before reaching anywhere else. 2/3 of the "Bay" south of the bridge is mudflat with a technical depth of 0' according to NOAA. Looking around Google maps it appears there is not a single dock, pier or anything for any boat to go to south of Dumbarton except for the Palo Alto boat dock - but it has a note that it is not useable around low tide and is apparently only used by windsurfers and canoes. It seems there used to be boats moving around down there (one park is named "Cooley's Landing" and the SPRR saw it necessary, or was required, to but a swing bridge over Newark Slough as well as the Bay) but there is no reason I can see to maintain sailboat clearance at Dumbarton. The number of sailors using it in a decade would probably be exceeded by a week's worth of train passengers (or less). They did build Dumbarton road bridge at 85' clearance, so maybe Cargill/Morton Salt is (was?) bringing something down there, so I could be wrong.

      Even if you need 85' clearance, cost of a billion plus for a bridge of this size, scope, and location is absurd, even by Bay Area contracting standards. The Al Zampa bridge at Carquinez is much more complex (150' clearance, main span seven times as long, suspension bridge with towers as tall as skyscrapers) and only cost $240M in 2003 for about half the total length. Viaducts are cheaper.

      And even if planners did make the bridge cost that much, the projected cost for the second transbay tube is $29B for 6km - so Clem/Richard's 10km Dumbarton tunnel plan from the post above would be $48B, or about $12B to replace the 2.5 km of current rail bridge and approaches.

      Once again, everywhere in the world constructing a concrete viaduct over shallow water is cheaper than a TBM bored tunnel of same length, no exceptions. You can't have it both ways and pick an absurdly high bridge cost then assume really low tunneling costs, even if CAHSR and Samtrans is putting out trash studies that say so. If a developer told you that building a three story 2,000 sqft house was going to be more expensive than a 10 story apartment building, would you say "Gee, I guess building with 2x4's is inherently more expensive that steel high-rises with foundation piles" or would you say the developer is full of it?

    6. Here's the thing: a TBM operating in perfectly understood and clear geology is a *factory*.

      There's a concrete casting factory doing the same thing over and over (with tiny pre-determined CAD-controlled variations for individually placed segments), there's a workface factory for turning geology (known geology!) into spoil, there's a conveyor belt connected to a factory that disposes of spoil.

      Special Friends of Willie Brown don't get to Enrich the Community.

      Special Friends of Steve Heminger don't get to pad the budget by billions and pad the schedule by decades with Unique Local Considerations.

      There isn't a Children's Art component.

      There aren't fifth-rate USA companies that make tenth-rate TBMs! You *have* to use machinery and experts who know the fuck what they're doing, because Buy American isn't a thing there yet, thank God.

      You don't even have the *opportunity* -- unlike everything at fucking Caltrain, unlike all the fucking cast-in-place make-work of cash-incinerating dumpster-fire CHSRA, to use 19th century construction practices.

      Once your *factory* is set up and the TBM is pointed in the right direction the worst you can manage is to feather bed the dump truck driving or something.

    7. > No bridge of this specification IN THE WORLD has ever cost *$1B per mile*

      The portal bridge replacement will far exceed this, current estimates are a whopping $10B/mile.

      I'm not sure the schmucks at CAHSRA can match the incompetence of Amtrak on that one

    8. Cargill appears to have an active spur, and no quay.

      You can see the vestiges of a shipping channel in the mountain view slough where the transmission lines rise over the channel.

      I couldn't figure out what used to be there last time I looked it up, but it's definitely not there anymore.

    9. As for where to go after crossing the bay, I suggest the best thing to do could be a mile-long viaduct following the water right of way, from Willow Street to Sycamore Street, and then smoothly curving onto the UP route through Fremont. This transition looks like it could be relatively simple to build with a roughly 1/2 mile, or 800m, radius. As far as my googling can discern, should be good for somewhere in the 90-110mph range. This would not be a "high speed" route by any reasonable definition of the term, but much faster than the slow reverse curve required to follow today's route, and certainly fast enough for suburban service on the Dumbarton corridor.

      If the above is not feasible, then just stick to the existing serpentine route, but make lemonade out of lemons by adding a stop there for a transfer to the Capitol Corridor, which will be relocated to the Coast Subdivision at some point. In fact, this may make sense no matter what.

      Beyond that, The UP right-of-way appears to be an even 100 feet wide through Newark and Fremont, which is easily enough for another pair of tracks. Through Fremont, I count a total of six grade crossings, and four of them (Cherry, Cedar, Blacow, and Dusterberry) are highly suburban in context, to the point where a solution shouldn't be very difficult. Maple and Fremont will be a little trickier, but still child's play compared with some of the stickier situations on the Caltrain corridor.

      As for where to go after Fremont, Niles Canyon should be double tracked, whether by adding another track parallel to the existing one, or taking over the ex-SP Niles Canyon Railway route on the other side of the creek (and, conceivably, straightening it, with a new tunnel about 2/3 of a mile long.)

      Something like this:

      Is all of this easy, no, but it's sure to cost less than the SNCF Dumbarton-Water Right-Of-Way tunnel plan which is bound to run into eleven figures.

    10. @Owen Thanks for sharing your ideas and the google map. I agree with you that a bridge makes a lot more sense than a tunnel across the bay, but actually this is one of the easier parts of an Altamont HSR route as getting through Fremont is much more challenging as your map shows. A viaduct so close to residential properties is going to be difficult to get approved and UP may not be so inclined to share their ROW. Unlike SJ-Gilroy, this part of UPs network sees quite a lot of freight traffic. I'm not sure how feasible the SNCF water route was either but SF might be more flexible about sharing their ROW than UP. It would have been nice to have these ideas explored in more detail in a non-sandbagged alternatives analysis.

      Regarding a transfer station to Amtrak on the Coast subdivision, it is hard to see how to fit that here. Current Capital Corridor plan is for a station further north at Ardenwood.

    11. @Richard, there are number of flaws in your argument:

      -A TBM is a repetitive factory, but the actual boring is not a major component of the cost of a passenger tunnel. Traction power, power for services, ventilation, emergency power, fire/life safety systems, vertical boring for ventilation and escape, etc. make up most of the cost (as they do in, for instance, a high rise where the electrical and mechanical subcontracts will be several times the value of the concrete and steel). Even if you remove the cost variability *of the tunnel lining* in known geology, you still have many, many systems that can result in cost blowouts, budget padding, and "enrichment".
      -Every single tunnel project in the Bay Area has suffered from the problem of cost padding you are saying tunnels do not suffer from. Central Subway, the short, unnecessary Lake Elizabeth tunnel, the projected costs for a second BART tube or BART to San Jose have all been far higher than international cost norms. How you can with a straight face type that tunnels don't allow "special friends" or "unique local conditions" when every single tunnel has exactly that is beyond me.
      -Construction of a prestressed beam bridge is also a *factory* with precast elements poured over and over into the same molds, identical piles driven identical distances apart, and traveling overhead cranes dropping box bridge sections before crawling forward to drop the next section.
      -Featherbedding is quite possible with tunnels and even TBMs in particular. Streetsblog has been detailed how the TBMs on the Second Avenue Subway were staffed with two to three times as many people as identical machines in Germany. The New York Times reported that East Side Access had 200 people on the payroll for jobs that didn't exist until an audit discovered them. Given your incredible knowledge on so many things relating to rail construction, I am surprised that you are ignorant of these examples and would claim that TBMs and tunnels are immune from this problem when it has been clearly documented that they are not.

    12. @Robert Moses

      "The portal bridge replacement will far exceed this, current estimates are a whopping $10B/mile."

      This is astounding, and somehow far more absurd that the cost estimates for a Dumbarton rail bridge others have referenced.

      I don't want to try to defend the cost, but given that this project is preliminary work for Gateway, is the $1.56B cost just for the 961' bridge, or is it for approx. 5km of new track from Sawtooth to Seacaucus to prepare for a second set of Hudson tunnels. If that later it is still way too high of a cost, but if the former, I don't even know what to say.

    13. Hi Owen,

      Thanks for actually making a concrete proposal and thinking things through.

      Aside from the lunacy of trying to deal with the local city governments and their ever-encroaching sprawl and do anything at all anywhere evening remotely near the rights of way owned by the extortion-is-our-business-model assholes of UPRR, what you've sketched through Fremont seems feasible -- in any other country in the world, that is -- if low-speed, and the additional takings for connecting rights-of-way are modest and considered.

      But I guess my question with any of this is ... what's the point? You spend a few billion dollars to get from Redwood City (a billion dollars of work even before you leave the Caltrain line, given our appalling transit-industrial mafia cost multipliers) to thread through Fremont alongside some freight tracks and ... you end up running at 50mph or something through endless curves along the super-constrained river banks through Niles Canyon, ending up along low-speed UPRR-owned tracks south of Pleasanton.

      Who's going to ride that and why and at what cost? Recall that (incredibly bad and stupid) BART extensions exist and are being constructed and aren't going away.

      Decades ago there used to be a nice, feasible, well-connected route as part of a well-throught-out network through Fremont to Livermore to Tracy to LA/Sacramento, but no longer.

      As for putting my crayon where my mouth is, I suggest the power line ROW just north of Auto Mall Parkway ("Auto Mall Parkway"!) before going into a big bore popping up somewhere close to and south of 680 near Mission Pass and then parallel-ish to 680 then 84 through Vallecitos Canyon then ... umm umm umm ... something cross-country skirting the endlessly-extending terrifying Pleasanton sprawl, then south of Lawrence Livermore and over the pass to Tracy, at real open-line HSR speeds.

      All the once-possible good connections (to northbound BART in Fremont, to non-BART southbound San Jose rail in Fremont, to BART in Livermore) are gone. What might be salvaged is a fast and direct HSR connection to the Central Valley, one that minimizes the massive construction costs and intolerable Caltrain service impacts from Redwood City to south SJ.

  31. Any Bay mariner knows that the Bay is not considered navigable south of the Port of Redwood City channel for anything with more than 4' of draught. Even older nav charts form the 80's/90's don't indicate that the channel through the Dumbarton Bridge is dredged or marked. Loaded ships that would need 85' clearance draught upwards of 30'. I can't imagine a universe where anyone would sign off on that scale of dredging and stirring up gold rush era heavy metals in order to get container ships another 5 miles south (to dock where???)

  32. Decades ago I think the Navy used to receive fuel at Moffitt Field via barge. Not sure how NASA gets fuel there now.

  33. @Anonymous: I don't think the issue is loaded ships needing 85' clearance, but sailboats. A 50-60' boat could easily have a 75-80' mast. Yet this doesn't change the end result much. A boat that size will have 7-8' draft. There is very little of the Bay south of Dumbarton with the 9-10' depth for that, and nowhere to sail to or dock, so I don't see any reason that Bay couldn't just be closed to big boat sailing.

    @Ben in SF: Barges shouldn't be an issue. They are only 12' tall, with only a few feet above the water line. The very largest towboats are only about 50' above water, with 20-30' more common, and you can get towboats with retractible wheelhouses to lower their height when operating on canals with reduced clearance. Even accounting for 6' sea level rise over the next century it should be very easy to provide a Dumbarton rail bridge with this kind of clearance.

  34. Years ago I worked for Sun Microsystems overlooking the Bay south of the bridge. Aside from the daily U-2 or F-104 launch (either would be above 20K feet before passing the end of the runway), the most interesting sight was the occasional barge with a large shrouded object on board, heading to or from what was then the Lockheed Missiles and Space plant adjacent to Moffett field. Satellites are a bit smaller now than they were in the space shuttle days, so cargo planes can do most of the heavy lifting (a huge AN-225 still flies in and out of Moffett occasionally). I suspect that one reason to keep it a navigable waterway is there may still be a possible need to move "national security" payloads that are too large for air or road.

  35. Caltrain announced today that they finally completed all 3,092 foundations. How many years late is this?

    1. See charts & graphs here. The electrification substantial completion milestone has slipped by 2.5 years. While foundations are done, pole installation is proceeding very slowly in segment 2.

      Amazingly they are STILL doing the study with PG&E to approve the load imbalance of hooking up to a single phase of a 3-phase grid, and the lack of closure is delaying the milestone for energizing segment 4. This is a design issue that should have been closed years ago and here we are after a nine-figure amount spent on PG&E substation upgrades with no approval to flip the switch.

      Makes me wonder if US rail substations should have 3-phase grid feeds with single-phase inverters on-site, another expensive way to avoid this bullshit.