Today, Caltrain is hardly recognizable with regular and punctual half-hour service all day, every day, using swift and comfortable trains that are the envy of any North American regional rail system. This unequivocal success sets the agency on the best possible trajectory out of the pandemic doldrums. Unfortunately, that won't be enough.
Most of what is still wrong with Caltrain is concentrated in Santa Clara County, where the wrong priorities are hurting Caltrain's finances. Revenue comes from a good product: a passenger experience of fast, frequent, and regular service. Here's what should be fixed in Santa Clara County:
Frequency is freedom
It is a well-established research finding that short and regular headways result in a faster-than-linear positive response in ridership and revenue. Unfortunately, Caltrain planners have decided that an end-to-end trip under the one-hour mark (known in the diesel era as the Baby Bullet) is worth sacrificing regular headways. Once an hour during the morning and evening peaks, a train will skip five stops to make this stunt possible: Santa Clara, Lawrence, San Antonio, California Avenue and Menlo Park, just across the county line. Fixing this error at a grand total of eight minutes of run time would unlock 15-minute clockface frequency throughout Silicon Valley, at zero added operational cost. The ridership induced by this tweak will dwarf the tiny number of long-distance riders who abandon Caltrain due to a longer trip, recalling that the average Caltrain ride is ~25 miles.
Re-imagine connecting shuttles
With 15-minute peak service throughout Silicon Valley, connecting services can be reconfigured so they no longer need to reach "major" stops (known in the diesel era as Baby Bullet stops). Silicon Valley always was a continuous employment and housing blob, and "major" stops were an artifact of diesel service patterns where the tradeoff between frequency and trip duration was far more pronounced than it is with swift EMUs. To reach "major" stops, shuttles spend precious minutes stuck in gridlocked traffic sewers that run parallel to Caltrain, such as El Camino Real and Central Expressway. Ditching this gridlock not only speeds each connecting trip, but allows the same number of shuttle drivers and vehicles to be redeployed towards more frequent trips; both effects will generate Caltrain ridership. The vast fleets of luxury coaches that ply highway 101 can be viewed as an indictment of Caltrain's service pattern; major employers will respond if Caltrain upgrades to a compelling 15-minute product.
Ditch diesel
Operating and maintaining a separate diesel fleet to provide infrequent part-time service to the small towns south of San Jose generates less than one percent of weekday Caltrain ridership (see chart). This astonishing under-performance persists even after the addition of a fourth daily round-trip to Gilroy in late 2023.While transit agencies aren't profit-seeking businesses and their purpose isn't always to maximize ridership, the Gilroy branch is one of those cases where the cost of providing the service is very far out of proportion with the public benefit. While Caltrain doesn't break out the cost of Gilroy service, the marginal cost of the fourth train is quoted as ~$3M, so we can extrapolate at least $12M plus the fixed operating and maintenance costs of separate tooling, training, parts, etc. associated with sustaining the diesel fleet. Caltrain would be better off spending this money on contracting with VTA for more frequent 568 rapid bus service. Between Gilroy and Blossom Hill, this south county bus is already much more frequent (~every half hour) and barely any slower (~8 minutes in peak traffic) than Caltrain.
Divesting of the remaining diesel fleet (9 locomotives and 41 cars) is a one-time source of income, but has some strings attached because the FTA funded its original purchase. Hanging on to diesels for "fleet resiliency" is becoming less critical as the electrified system demonstrates increasing reliability. The Trump administration is unlikely to care either way, and there are plenty of operators who might be interested, such as a potential new agency based in Monterey County.
Is this poking south county in the eye? No, because there's a much better plan. Read on.
Acquire UPRR's Coast Subdivision and electrify to Blossom Hill
As compensation for deleting service to Gilroy, Caltrain should extend electrification and frequent EMU service by six miles from Tamien to Blossom Hill. This portion of the corridor has high residential density to support significant new ridership if well-served, which it currently isn't. Caltrain likes to argue that a railroad has high fixed costs, and that cutting service can't save much money. The converse must also be true: adding more EMU service, using the existing fleet, can't cost all that much.
![]() | |
A stack train under the wires; it's really no big deal. (Samuel Walker photo) | |
If this sounds like a megaproject, it isn't. It does not require any new traction power facilities; no new paralleling station is needed at Blossom Hill if this short extension is initially built as basic 25 kV without feeders. It does not require environmental clearance, thanks to new laws (Alex Lee's AB2503). It does not require any new trains, as Caltrain's EMU fleet will soon swell to 23 trains, where today's service pattern only requires 14. It's about as basic as electrification projects come: string up 15 track-miles of wire.
To get VTA interested in helping to fund this capital project, you would call it the "South County BART Connector." Since San Jose Diridon station would then require two tracks and a single island platform to support all Caltrain service, there could be savings in postponing the gold-plated Diridon Integrated Station Concept, a megaproject that costs $3-$6 billion while providing no identifiable service benefit for Caltrain passengers.
Failure of Imagination
With pandemic-era federal subsidies expiring and a new transit-hostile federal administration, Caltrain needs to show more creativity and imagination in adjusting its offering. The success of the initial electrified service shows that the best prescription for financial health is to focus relentlessly on the product: fast, frequent and regular service. Anything that doesn't contribute to the product is a distraction.
Re frequency and stopping patterns: Elsewhere in the world it would be possible to run all-stopping trains every 15 minutes and limited stop trains every hour. And then the limited stop trains would not just skip a few stations, but skip a lot of them creating a way faster service. But also: It makes no sense skipping stations near the ends of the route as people near the end benefit the most from a faster train, and thus the skipped stations should be in around the middle of the route (from say 25% in to 75% in on the route length).
ReplyDeleteRe south of Tamien:
Not only Caltrain and Cali HSR have interests in acquiring the UP route.
There are talks about a Gilroy-Salinas service, and also a Santa Cruz-Pajaro service. In the latter case the public sector already owns the railway, but there are NIMBYs who don't want trains disturbing their digestive system... Santa Cruz - Pajaro would to some extent stand on it's own as a way for local commuters to bypass slow traffic, but it would obviously benefit from being connected to a greater "south of the bay area" system.
IMHO the state and all transit agencies and counties in the area should buy all major private railways in the bay area, all the way to Sacramento, Stockton and also buy the coastal route all the way to San Diego.
I get that the Gilroy service has a low ridership and that makes it hard to put money into it, but on the other hand you also correctly writes that a good service greatly increases ridership. Four round trips, especially focused on job commuting, isn't a good service. It would be better if the four round trips ran every four hours than the current schedule where trains run not that far apart. Like it's not worth running more than one train peak hours if you don't run any off-peak trains, as the peak hours trains are anyways so infrequent that they are only attractive to those that are world champions in being on time at the rail station, and for those it doesn't matter if there are one, two or three trains, they will ride the train that happens to exist either way. Meanwhile if you don't run off-peak trains anyone who don't intend to spend more or less exactly the equivalent of a work day in the bay area won't take the train from Gilroy, at least if they don't think the bus is good enough, and if they think the bus is good enough then they will take that anyways if it runs at a slightly more convenient time.
In other words, electrify all the way to Gilroy. And while at it, continue to Hollister and to Salinas.
It's worth remembering that Santa Clara county and thus VTA is rich! They have enough money to spend on a more expensive than it has to be BART extension to Diridon (I.E. stacked tracks/platforms in large tunnels than cheaper separate bore tunnels), and they are fine with running one of the least effective transit services in USA.
I think we should try to avoid austerity style discussions when talking about VTA and Santa Clara county.
For anyone like me who's not from the area, and unlike me can't be arsed to look at a map, Santa Clara county extents southwards to halfway between Gilroy and Hollister, and westwards from Gilroy it extends halfway to Pajaro.
Time separation is no longer necessary in the US. Caltrain’s EMUs are based around the exact same crash safety requirements as the rest of the DMU shuttles in the country.
Delete@Nick:
DeleteI'm thinking about using DMUs that aren't FRA crash worthiness compliant.
I.E. is it possible to use time separation to at some times classify a route as a railway, and at other times as a metro/tram?
There must be some rules that allows this in some way, as both non FRA crash wortiness compliant vehilces from the LIRR and FRA crash worthiness compliant Amtrak, MTA and NJT share parts of NYC Penn Station.
===================
In general re buying the route:
It would probably be a great idea if Caltrain, Cali HSR and Santa Clara county together did a serious enough study about what it would cost to either acquire land adjacent to the current right-of-way, I.E. just build new tracks next to the current line, but on land that UP don't own. And also what it would cost building a railway on one or two alternate routes, like for example following highways or whatnot. Even of all these options would in the end be kind of dumb, it would most likely result in a better bargaining situation between Caltrain, Cali HSR and Santa Clara County v.s. UP. It would for one be totally obvious that UP would never get more money than what a route adjacent to the existing right of way would cost. But there is also some room for negotiating the price even lower, as if there isn't an agreement on buying things from UP, UP would have to reduce the value for that thing in their book keeping. Like currently the Tamien-Gilroy route has a certain value in the form of being possible to sell to the public sector or maybe even rent/lease part of it, but if the public sector builds their own alternative route then the only remaining value is their ability to run a few freight trains on the route.
I don't know what parts of the public sector are legally allowed to do, but I would hope that it's legal for it to do something that at the moment might not be the most profitable option, but in the long run might be.
There are basically no DMUs which do not comply with US crash safety requirements which are also available for sale in the US. Anything EN15227 is substantially compliant and the (no longer for sale) Nippon Sharyo DMU is a standard compliant vehicle. I don’t know why anyone would go out of their way buy something not compliant when there are so many options which are.
Delete@Nick:
DeleteI don't know what's available for sale now, but the eBART, NJ Riverline, Austin (TX) Capital Metrorail and the A-Train (Denton, TX) uses the Stadler GTW that isn't FRA crash worthyness compliant.
It seems like that at least the NJ Riverline uses time separation to be able to run freight trains.
In other words, Caltrain could use the same type of cheap small DMUs that eBART uses to run the Gilroy line, assuming their top speed (87mph / 140km/h) is good enough.
https://en.wikipedia.org/wiki/Stadler_GTW#United_States
Side track: The eBART Wikipedia article states that a BRT solutuon would cost somewhere in the ballpark as much as the DMUs. Given that in the Gilroy case both the highway and the rail line already exists, I wonder if running DMUs could even be cheaper than running the existing bus service? (I assume that the bus would still be needed to serve intermediate stops and whatnot, but still?).
https://en.wikipedia.org/wiki/EBART#Finalized_plans
What are the requirements for non-FRA crash worthiness compliant rail vehicles otherwise? I assume that there are other regulations they have to comply with. How hard would it be to just use slightly modified versions of DMUs used elsewhere in the world? In Europe it's not uncommon to have DMUs with a max speed of 125mph / 200km/h, and those are available in lengths that are far shorter than the Caltrain EMUs, I.E. won't need that much dead weight.
The only downside of using a non FRA crash worthiness compliant DMU on the Gilroy service is that it would either need to just run to a separate platform at Tamien, or run on the freight line all the way to Diridon, and it would then have to be separated from all the other passenger trains (I assume?). Not impossible to solve, but needs to be considered.
Btw Gilroy has a population of about 59k, while Hollister has a population of about 41k. Extending the diesel services to Hollister makes sense. It's a straight route and afaik only a single daily freight train uses the line currently, so there is no lack of track capacity. The distance between San Diego and Gilroy is far longer than the distance Gilroy-Hollister. Adding Hollister increases the catchment population (for a lack of a better wording) for the service from 59k to 100k.
Well Stadler no longer sells the GTW, so that's not really much of consideration.
Delete“The success of the initial electrified service shows that the best prescription for financial health is to focus relentlessly on the product: fast, frequent, and regular service. Anything that doesn't contribute to the product is a distraction.”
ReplyDeleteWhile the service, as you mentioned, is unrecognizable compared to the diesel trains, ridership hasn’t recovered to pre-pandemic levels. I genuinely hope that the Caltrain is “rewarded” for its success, but I’m concerned that they aren’t.
I completely agree that prioritizing more frequent service is far more crucial than reducing the total travel time by a few minutes. The fewer riders have to adjust their lives around the Caltrain’s schedule, the better. Especially considering the unreliability of the connections, particularly on the MUNI.
On that note, one of the most significant improvements would likely be the Transbay extension. However, not only was that project already far behind schedule, but it has also been delayed again!
I would say that no matter if the Salesforce extension gets built, if the Link21 tunnel gets built, or if we are stuck with the current terminus, we need updated FRA regulations re what has to take place, or rather what doesn't need to be done, when turning around trains. In particular there needs to be an update to the regulations where a train can be left in an "on but no cabs activated" mode where the drivers can just switch cabs without needing to redo brake tests and whatnot. Also any testing procedures for the safety/signalling system has to be so fast that it doesn't hinder operations (I.E. happen while the driver adjusts the seat and whatnot).
DeleteThat change alone would increase the reversing capacity at terminus stations. Either adding stepping back (changing drivers so each driver drives the next train) or having staff do double duty (I.E. checking/selling tickets and whatnot in one direction, driving in the other direction) would increase the capacity even more.
With a more effective procedure, including improved regulations, the existing terminus is by no means the bottle neck for the current double tracked railway.
Turnaround can already be done extremely quickly (5 minutes) if there's no conflicting traffic and the dispatcher can line a route. Only the following is required:
Delete-Walk to the opposite cab (this is half the time)
-Class II brake test (simple application and release of brakes)
-Initialize PTC (departure test not required if it has already been done at that cab since the last train power up)
The idea the coach buses are driven by Caltrain frequency is cope. They’re driven by the fact that Caltrain is accessible to very few neighborhoods in SF and mostly mediocre neighborhoods at that. I bike from the mission to 22nd each day (so I can live in an interesting place) and my coworkers think I’m crazy.
ReplyDeleteI only need to come in 2 days a week (3 days one week of the month) (Thanks, COVID) But I drive about 10 minutes and park around the 22nd Street station (which costs about $3 on the street).
DeleteWhile I realize that one-seat rides are greatly preferable, with 101 traffic getting worse (even in the toll lane), and tech companies getting more cost-conscious with employee perks, I could see the buses being reorganized into Caltrain feeders at both ends of the trip... if the train comes every 15 minutes
DeleteOne of the other difficulties is that tech companies *really* don't want to share long-range shuttles even with other adjacent companies (e.g. Google and Intuit in Mountain View) for confidentiality reasons --- and if you can't work on the bus, it becomes much less attractive to would-be riders. It's not obvious whether that effect would be seen on trains, given plenty of people work on Caltrain already.
Delete"The idea the coach buses are driven by Caltrain frequency is cope."
DeleteMy dude, everything is cope!
Caltrain's never going to win the SF Richmond District to Palo Alto commute market.
But we can at least not abandon Mission District to Sunnyvale. Because that's what shitty headways do! Taking Caltrain is twice as slow as driving on a good driving day. but there are some upsides. "I just missed my train and have to wait half an hour for the next one" is not one of those upsides.
A normal person — I used to work with them in the awful decades I did soul-destroying 22nd—Palo Alto/California Avenue/San Antonio bike+shit-train+bike commutes — experiences this once, and, rationally, says "Caltrain; never again. My life isn't worth that." (An irrational person such as myself would miss a train on 3-hour headway (6am to 9am bike-on-train blackout period in the 1990s!, or hour hedways through the 2010s) and say "I must sacrifice myself for the planet!". Fucking loser. Sucker.)
But if missing a connecting train means being home 14 minutes late, well, maybe, sometimes, maybe not every day, maybe not every rider, but maybe!
And the ridership jump due to electrification continues. April 2025 numbers just out show a 53% increase in average weekday ridership (23,131 to 35,374) and a 60% increase in total ridership (577,883 to 924,999). I'm totally "on board" with abandoning the diesel trains that serve Gilroy. And also expanding the current electrified track down to Blossom Hill, provided that the solution doesn't preclude, or require any additional work, to allow for HSR use of the tracks. So it may be something more than just "stringing up the wire", but you get the idea. As always, money is the main impediment to anything being done.
ReplyDeleteThe weird thing in California (and the U.S. in general) is that there is plenty of money to build stuff, just not enough to operate it. There was never a problem finding money to replace bridges over Los Gatos Creek and the Guadalupe River.
DeleteSpeaking of the Guadalupe River Bridge(s) lengthening & replacement project funding … it has (surprise!) turned into quite the extended fuster cluck. Turns out Caltrain originally insisted on replacing the old wood MT1 bridge over UP objections. UP only wanted to pay for a retrofit/rehab. It went to arbitration and so UP ended up only having to pay a fixed price based on a now-old rehab estimate. Caltrain did the initial work inconsistent with their numerous environmental permits and has been stuck in multi-agency permitting hell ever since getting caught … and the new soon TBD price has soared and Caltrain is eating all the overrun, looking to cannibalize more of their already constrained captial budget.
DeleteAnd now staff is planning to “speed” construction by de-energizing the OCS on the MT2 bridge for something like 8 months coupled with a Tamien-Diridon replacement bus bridge that they said they’re quietly hoping so few will bother to ride (vs. getting to Diridon some other way on their own) that they can cut more costs by ending it ASAP after it starts. The board is NOT at all happy:
Update on Project Budget, Contract Capacity, and Funding Plan Increase for
Guadalupe River Bridge Replacement Project Construction Contract 22 -J -C -032
Stupid questions but:
Delete1: Why didn't they build the new bridge with two tracks in anticipation of this replacement?
2: If it really is an issue, truncate trains at Diridon, but run a basic shuttle from Tamien - Diridon.
3: Why not build an entirely new bridge just north of the existing old one, move traffic there, dismantle the old one?
If they are so inclined and planned a bit better they could reuse the right of way to get 3 or 4 tracks across the river. Future-proofing the corridor is always easier than going back and redoing it later.
Edit to #2 above - it should read "run a basic rail shuttle" meaning 1 track and 1 train going back and forth between Tamien & Diridon.
DeleteThey cannot run electric trains during construction because regulations require them to turn off the power to Track 2 while the Track 1 bridge is under construction. They evaluated a diesel rail shuttle and evidently decided against it in favor of a bus shuttle. Honestly Tamien has so little ridership I'm puzzled why any train other than Gilroy service stops there at all.
DeleteI admit that I don't know where the electric feed can be sectioned off, and I'm too lazy to try to find out, but:
DeleteIf it would be technically possible to have power on at Diridon and Tamien, but turned off at the bridge, they could just floor it at Tamien and Diridon and if a train would get stuck (i.e. something causes a signal to show stop or whatnot) just temporary stop the bridge works and turn power on.
(I assume that it might be possible to put up a large++ conductive net, grounded, between the two tracks. Think like the net that is used to avoid balls from certain type of sports places to weer of into the neighborhood).
Also: Would it be possible to have time-separation between FRA crashworthiness compliant and non-compliant vehicles on the track they aren't working on at the moment? If so maybe check if it's possible to "borrow" a vehicle from the Anitoch shuttle or whatnot?
Agree that it would probably be a good idea to reinstate removed tracks while at it. In particular build a new double track bridge for the missing track 3+4 and when the bridge for track 1 needs replacement just replace both track 1+2 with yet a new double track bridge. (Don't know the reason for the bridge needing replacement now, but in general a bridge that can't take rail traffic will stand without load for ages without needing much maintenance, so it wouldn't be much extra cost to just keep it for a while). As a bonus, expanding to quad track could be something that the Cali HSR project could partially fund.
"Future-proofing the corridor is always easier than going back and redoing it later."
DeleteThis commonly-held belief is actually wrong, and stems from a mistaken belief that the main stakeholders are riders or taxpayers. Redoing it later is actually quite profitable; why do something right when you can do it twice, and get paid for it?
Is it still true that VTA pays for the entire operating cost of Caltrain San Jose to Gilroy service?
ReplyDeleteNo. *Everybody* from SF to SJ is scewed by worthless Gilroy trains.
Deletehttps://www.caltrain.com/files/caltrain/Assets/Full%2BSpecial%2BCounsel%2Breport.pdf
"Since 2001, the annual operating costs have been treated as a mainline cost and
are paid by all member agencies"
It gets worse: "In 2018, the member agencies established average mid-week boardings, adjusted annually as the means of allocating operating costs amongst the member agencies"
This means that the *worse* service performs, the more *other people* pay for it.
But at present nobody pays for anything. It's *all* Other People's Money. Woo-hoo!
https://www.caltrain.com/media/34732
Annual Comprehensive Financial Report Fiscal Years Ended June 30, 2024 and 2023
"Member Agencies provided operating funding to the JPB prior to fiscal year 2023. Net operating and
administrative costs were apportioned on the basis of mutually agreed contribution rates, updated on an
annual basis prior to fiscal year 2023. In fiscal years 2024 and 2023, due to the funding from Measure RR
transaction and use tax, the JPB did not request member agencies contributions."
Going back to FY 2020's Annual Financial Report
https://www.caltrain.com/media/1455
the average weekday boarding-based per-county operating subsidy split was
San Francisco County: 27.0%
San Mateo County: 30.6%
Santa Clara County: 42.4%
If updated to most recent guesstimated Ridership (all-month, weekdays not broken out)
https://www.caltrain.com/about-caltrain/statistics-reports/ridership/fare-media-based
that would be
SF 26.5%
SM 32.9%
SC 40.6%, of which Gilroy "ridership" contributes 0.9% = 8343 of system-wide 924999 for April 2025.
So VTA would be paying than half of Gilroy extension operating costs, if it were paying anything, which it presently isn't.
Great success!
First, I agree that an electrified service should be extended south of Tamien. I'd pitch it as a way to do the following:
ReplyDelete-Improve speeds by removing CEMOF, and relocating to the Lick Quarry/flea market site, which also unlocks...
-Massive TOD opportunities around the Capitol site
-And, Blossom Hill station TOD
Alternatively, you could also extend the electrification to Bailey Avenue and 101, creating a terminus station there. That becomes the new CEMOF and a park & ride station. You can now truncate the service to this park and ride station, getting rid of the diesels.
Second, I agree that the Gilroy service as of now isn't worth what they're spending on it, but that's not the full picture. There are three big factors that people miss:
1. It will enable a bunch of key connections including CAHSR in the future, it's a key part of the route for long distance Amtraks, and it's a key connection to the Monterey Bay & Salinas areas: https://www.reddit.com/r/cahsr/comments/1kbq0cj/why_gilroy_is_the_right_choice_for_highspeed_rail/
2. It can be easily upgraded to shave as much as 15 to 20 minutes off of the total travel time of CAHSR. Instead of taking 45ish minutes to go SJ - Gilroy, it would take about 26 minutes, saving almost 20 minutes off of the total 2h40m travel time for much cheaper than Burbank - LA or SF - SJ upgrades. (Assumptions: average is 90 at first to go 68 miles, then 160 to go 68 miles).
3. When thinking in the long run, this SJ - Gilroy route would have a lot of other benefits for a regional statewide rail network. It allows for a Hollister connection, Monterey Bay/Salinas, and even down to the Paso/SLO/SB/Ventura area if we start investing in upgrades all along that corridor. It then becomes a nice secondary lower speed HSR line (think 110 to 125mph) that also serves that underserved population along the coast. That could mean travel times from San Francisco to Santa Barbara in just 4 hours, or San Francisco to Paso Robles in 2 hours. That's pretty significant when thinking longer range.
How would you get SF to Paso in 2 hours? SJ to Paso currently takes 3.5 hours, then add an hour for SF to SJ. That means doubling speeds between SJ and Paso. The slowest part is between Gilroy and Castroville. That segment is not very long and straightening it out looks to be expensive.
DeleteSame person as 12 May Anon: SJ to Paso - good question. For the Gilroy - Salinas segment, I would either upgrade the tracks via Chittenden a bit for $500M or so to speed up travel times with transfers at Pajaro to the Monterey line, or route a short ~2 mile tunnel just past San Juan Bautista roughly following San Juan Canyon/Old Stage Road. The issue is that if you have a fast connection via this tunnel, it makes the Monterey line much more difficult to transfer to. I would actually just throw $500M to upgrade the curves double track, and figure out Chittenden to get it to close enough to 79mph, and call it a day.
DeleteThe biggest time savings are from Salinas to Paso Robles, tbh. The tracks are straight, flat, the land is sparse and cheap, and a decently fast train (think 110 to 125mph) is actually pretty doable. With just one stop between Salinas and Paso Robles at King City and an average speed of 85mph gets you about an hour (assuming approximately 100 mi).
Not to crayonista too much but I think if a regional line was established that goes from LA - SLO and is gradually upgraded and realigned to get to 110-125mph, that would also be very popular. I would add new tunnels from SLO - Paso Robles, realign the line to go to Santa Maria instead of Guadalupe, plus a connection via 46 to get to the Central Valley HSR trunk line at Hanford. This makes Paso Robles a nice transfer station for the central coast.
The major issue for this segment IMO are in five general areas, some more expensive than others:
-Chittenden (either ugprade it or bypass it via a tunnel south of San Juan Bautista)
-Paso Robles to SLO (new tunnels/faster alignment are needed north of SLO)
-Ventura - SB (tracks are literally at sea level e.g., at Santa Claus Lane in Carpinteria)
-The Vandenberg routing needs to be deleted: I would add a tunnel at Gaviota to Solvang and Lompoc then to Santa Maria
-There's no mid-Central Coast connection to the HSR trunk line, hence my suggestion for a line from Paso Robles - Hanford via 46
You're looking at probably $5B to $20B for these upgrades, but everything else in between can be upgraded to 110-125mph operation. If you did a Brightline Florida approach, that'd be a great Phase 3 effort for CAHSR to help create a regional connection from the coast to the rest of the state.
A wild card re the Santa Cruz - Pajaro - Salinas - SLO -onwards line:
DeleteIf rail gains significant political/opinion traction in a not that distant future, it might be worth looking in to a tunnel connecting Santa Cruz to San Jose. If this is done in a good way, this could speed up travel times on the coastal route.
Re speed: If it would be possible to do 125mph then a non-stop train would take about 50 min Pajaro - Paso Robles. It would obviously take longer for a train that stops at intermediate stations, and it would obviously not be be possible to achieve 125mph for all of the route.
But if it's possible to close enough at grade crossings and do whatever is required for higher speeds, I think it's reasonable to aim for true HSR speeds for the section that is flat and straight, and then just live with the snail speed near SLO, and the relatively slow speed between Pajaro and Gilroy.
Also, note that it's "only" the middle of the route between Pajaro and Gilroy that is slow. On both ends it's fairly straight. Reroute it across the Aromas granite quarry and build about a mile of rail in tunnel, on a curve, and connect to the straight section of the existing railway towards Gilroy, and you have improved the speed quite a lot with a way shorter tunnel (and perhaps bridge across the quarry) than what a new tunnel Santa Cruz - San Jose would cost.
And yes, the junction where the Pajaro route connects to the straight Gilroy-Hollister route would have to be improved as it's a fairly sharp curve. Fortunately a straightened route would use farm land rather than built up land. Btw a new alignment would probably have a junction exactly at the current edge of Gilroy, about 2 miles south of the current station. That seems like a great place for an additional station for all-stop trains.
Btw now I'm imagining a future that probably won't happen, but looking at Hollister the railway used to continue along road 25. It looks like there is enough space to reinstate a single track line except next to the current end of the line. Road 25 might need to have a lane removed at places, some shopping areas might need to have their access on the adjacent local roads, and it would likely be a good idea to grade separate Sunset Dr. vs road 25 without having them connected. By doing this there could be a "Hollister south" station and south of it a depot for stabling trains and whatnot.
@Anonymous - That is an ambitious though feasible plan. I'd expect it would have to wait for the all phases of CAHSR (including Sacramento to San Diego) to complete before the state would consider such a big service upgrade to the less populated Central Coast.
Delete@MiaM - There's actually a tunnel already piercing the Santa Cruz mountains left behind from the old railway between Los Gatos and Santa Cruz. However development since closure would likely preclude reusing that structure. The ROW leading into the north portal of the tunnel is now under the waters of Lexington Reservoir for example.
@MiaM & @thielges — Connecting San Jose & Santa Cruz has been debated for years. IMO, the only viable option is a base tunnel: either from the Permanente quarry (via the Vasona branch or 85 ROW to Caltrain), or from Los Gatos via Vasona. I prefer Los Gatos, as an elevated stop near Main Street over Hwy 17 would serve downtown and enable local service to Diridon, improving regional access.
DeleteFrom there, a line could follow Hwy 17, entering a tunnel above the dam. An 8-mile tunnel to Scotts Valley or 7 miles to Boulder Creek is far shorter than a 13-mile tunnel from Permanente. Summit tunnels are a non-starter due to NIMBYs, property impacts, environmental issues, and unknown structural viability.
A long base tunnel avoids steep grades, curves, and major opposition. Still, at \~\$5B (like CAHSR’s Pacheco Pass), it's only worth it if SC/Monterey rail projects succeed and CAHSR reaches SJ.
As for Hollister, no need for a new line along Hwy 25—existing rail nearby just needs upgrades. A park & ride near 156/25 could work. At 14 miles, \~\$12–27M/mi = \$168–324M. 20-minute trips at 80mph beats 30-minute drives. Better ROI than \$171M spent on the 101/25 interchange.
Pajaro/Chittenden Pass: worth upgrading for 15–20 min trips to Gilroy. \~\$1B to fix curves/pass. South of Pajaro, the Elkhorn Slough is the only real hurdle—viaduct & double-track it for \$1–2B. 40 miles at avg. 65mph = 45–50 mins—competitive with driving.
For the Central Coast (SLO–Ventura), gradually upgrading segments to 110–125mph would build ridership. Over time, consider bigger projects (tunnels at Solvang/Lompoc, bypassing coast, etc.). With the Cross Valley Corridor online, extending to Coalinga and tunneling to San Ardo (or via 46 & Paso Robles) would link the interior to the coast effectively.
If you've ever seen that photo of the three-rail track bent into an S by the 1906 earthquake, that was in the Santa Cruz Mountains (as the South Pacific Coast division of Southern Pacific was just upgrading from 3-foot to standard gauge that year). The fault slices across the mid-to-upstream end of Lexington Reservoir. The railroad crossed the fault on the surface, but the tunnels were affected by the fault, too. Not something you'd probably want to cross underground; the reservoir and steep slopes all around makes it tricky to pick a good surface alternative. (Ben in SF, peasepress.com)
Delete@first anon:
DeleteRe Hwy 25 - I'm referring to reinstating the removed tracks south of the current end of the line in Hollister. That way it would be possible to have an additional station in the southern outskirts - only costs time for the transit agency but not for any other passengers as it would be the end of the line - and then right outside the edge of Hollister have a depot. That way you don't need to reverse trains at "Hollister Central" (as compared to having a depot at the northern outskirts of Hollister, or for that sake empty running from a depot at Gilroy).
re Santa Cruz - San Jose - it almost feels like the political struggle might end up inside San Jose. Sure, it might be possible to use the current freight line with all it's level crossings. adjacent to the VTA line, but that wouldn't be great, at least if the line in the future would be used by trains to Salinas and possible further south along the coast.
Re around Pajaro: Agree that the Elkhorn area would also be good to straighten out the route along. It's less severe than at the Aromas quarry, and also for trains that stop both at Pajaro and Catroville it might not have that much importance. Don't know if it would be reasonable to skip Castroville? If Salinas would be the terminus then I'd say let all trains stop at Castroville (and consider one or two additional stations in Salinas), but for future longer distance trans it might be worth skipping Castroville.
A philosophical problem, of sorts, is that on one hand if Gilroy-Pajaro is improved then a future service to Salinas and/or Santa Cruz would have a shorter running time and thus could become more popular. But on the other hand, if a service would be popular that increases the incentive for building a tunnel Santa Cruz - San Jose, which in turn makes any Pajaro-Gilroy improvements less useful. They would still be useful for connecting the coast to Gilroy, Hollister and HSR trains to the valley, but still.
@Second anon:
From a safety persective, afaik earthquakes are known in advance so no train would be inside a tunnel when an earthquake may happen. From a cost-benefit perspective, it seems like USA builds a lot of things that don't last that long, so a tunnel that might collapse every 50-100 years seems to not be worse than many other things. Sure, it's not great, but still.
Also: I assume that the 1906 earthquake affected tunnels that were just dug/bored through the mountains as is. I don't know what the plans are for the Cali HSR tunnels but I assume that they will be reinforced. I was about to bring up the LA metro tunnels with their thick concrete structures, but those are probably not relevant as they don't go through mountains.
@MiaM - The San Andreas fault slips horizontally. In the unfortunate case of that fault slipping right where a tunnel passes, no amount of reinforcement will prevent the tunnel from being at least partially severed. I imagine that engineers would try to avoid such a situation by routing the tracks at grade where it crosses a major strike-slip fault like the San Andreas. The at-grade rails would also be severed though a derailment at-grade isn't as dangerous as one underground or on an elevated viaduct.
DeleteBART crosses the Hayward Fault in the Berkeley Hills tunnel. If the train is operating at normal speed and you count to 10, that's when you'll feel the sudden sway where they've been gently easing the tracks to accommodate the creep in the fault. They have studied what to do about it and any permanent fix is VERY expensive. If you search, the report is online somewhere.
DeleteCHSRA has done their best to avoid crossing faults in tunnels. Their solution, should they need to cross a fault underground, is to excavate a "seismic chamber", which is a horizontal widening of the running tunnels to allow tracks to be realigned to compensate for any rupture or creep without exceeding their track geometry guidelines.
I'm sure, if you search, you will find Japan has their own solutions to deal with faults on high speed lines.
@Anonymous:
DeleteThe important question though is if modern earthquake detectors will detect a coming earthquake with a high enough reliability to avoid that a train is in the fault zone when an earthquake happens?
@Michael:
Do Cali HSR intend on manual intervention after every earthquake that to some extent changes the geometry of the tunnel, or do they use some sort of "bridge" inside the tunnel, with expansion joints, to in most cases not need any adjustment, just inspection?
Btw this reminds me of taking off with an airplane. For airplanes there is a decision speed where you shouldn't abort a take off if you have exceeded that speed, as you can't safely come to a stop without veering off outside the runway. Thus you kind of have to take off even if there are fairly major faults, like an engine failure or whatnot, and certainly for all minor faults that the plane can fly with but you are supposed to emergency land anyways if they happen. I.E. take off and immediately land again as soon as possible, after running through check lists.
For trains in earth quake affected areas I assume that you would have some automatic system that selects if the train should brake to a stop before the fault zone, or if it should speed as fast as possible through the zone. Like if you can't stop before the fault zone, there is a hard choice between either flooring it to get out of the zone as soon as possible, or to brake anyways to reduce the impact if you would run into a failed section of a tunnel.
@Mia- Earthquake "warning" systems simply transmit a warning immediately at the beginning of the quake. Therefore, the distance from the epicenter determines how advanced the waring is. A few seconds if nearby, maybe a minute if far away. HSR takes well over a minute to stop in a full emergency at top speed. To me, warning helps, but won't solve the problem if the tracks fail on a fault if a train is approaching and nearby at top speed. Broken track warnings are ancient technology that would signal trains to stop regardless of cause of break. All BART trains automatically stop if an earthquake of sufficient magnitude occurs and then slowly advance to the next station to inspect the track.
DeleteAs for the CHSRA tunnel designs, they are designed to allow enough room to realign track should it be needed across the fault. It doesn't assume the track would be suitable for operation immediately after a major quake.
I'm guessing you're from a place that doesn't deal with earthquakes. They will happen, and we will try to engineer structures to the best of our knowledge of quakes. But nothing can ever be certain with a quake.
@Michael, ten seconds is right on the money. I remember the reports about the Hayward Fault crossing; I don't think I was able to see much different - I must have felt a slight bump or shift in the track. In my mind the train passes through a seismic chamber, but it appears that is a figment of my imagination, and remembering the proposals in the report you mention. Never could see much out the offset front window (even when it was wide; I no doubt annoyed a lot of train operators as a teenager). The Hayward Fault was not nearly so well understood (as equal or greater danger of a big eruption than the San Andreas Fault) when BART was engineered and built. Now they have 40 years of measurements in the tunnel that crosses it. And maybe a tiny bit of inadequate retrofitting. If you watch Vincent Woo's Tunnel Vision video (https://www.youtube.com/watch?v=3-Jrp6it9Ss) his SFO-Pittsburg train enters the Berkeley Hills Tunnel at 54:00 and there is ALMOST nothing to see at 54:10, but right about 54:10 to 54:11 there are some extra gray channels or stripes along the tunnel walls/ceiling, then they're gone again (just the regular conduits). Maybe these are actual reinforcement, maybe sensors and monitoring devices. Possibly a slight bend in the rails and catwalk. Not much room to make big changes! (At the usual 60-70 MPH this would be about 700 to 1000 feet inside the west portal). Anyhow there is much less there than I thought there would be but interesting nonetheless. - Ben in SF
DeleteYeah - there are solutions to the fault lines. Not just monitoring and analysis tech but also building a large chamber at the fault line to compensate for the lateral shifts that may happen. The metro lines and Shinkansen in Japan run just fine right after earthquakes, fwiw, because it's designed for that
DeleteSo the TL;DR is that it's actually reasonable to build a tunnel through a fault line, from a safety perspective, kind of sort of.
DeleteBtw the literally nuclear option would be to reintroduce "project plow share" and nuke the mountain, to be able to build a flat surface route :D
Still far out but at least a reasonable thought experiment would be to estimate the cost to use conventional explosives to remove a mountain. First random google result says that the mountain is about 1100m tall, about 0.7 miles. That's a lot (and it might not be the height for where a railway would go), but I think it would be worth doing a rough estimate on what it would cost, just to know once and for all if it's at all feasible or if it's totally bonkers.
Well, yeah. You could use the resulting mountain scraps to build a causeway out over Silicon Valley, maybe to Apple Headquarters; take out a mile wide swath of suburbia but then sell homeowners embankmant-side view lots to relocate. Elevator down to ground level. Problem solved, kinda sort of. Yeah, watersheds, habitat, oak woodlands, redwood groves, mountain lions, whole towns and subdivisions... better to reintroduce grizzly bears.... Ben in SF
DeleteWhile Santa Clara is supposed to pay separately for the cost of Gilroy service, it has not in recent years, mostly due to the difficulty in apportioning the cost. Post-electrification this is no longer a problem as the cost of Diesel service is easy to break out. Expect how this service is paid for to become a key topic in the upcoming negotiations for how to close the looming Caltrain budget shortfall.
ReplyDeleteWhile I agree an electrified extension to Blossom Hill would be nice, I think most of the benefit could be achieved with an extension to a new station and yard at former Lick Quarry just north of the existing Capitol station. The location can be reached via a siding from CP Lick so does not need any huge payout or protracted negotiations with UP. Since the VTA 568 bus goes up Monterey Highway nearby, the bus connection to Morgan Hill and Gilroy is much better than having to go all the way to Diridon or Tamien. Most likely a 3rd track would be needed for at least part of the distance from Diridon to CP Lick since UP will likely refuse to allow MT1 to be electrified.
If the cost is so easy to break out, I sure haven't seen it done. I would really like to see it, as I view the Gilroy operation as a sort of canary in the coal mine... it will reveal if Caltrain really has the fiscal cliff they claim to have.
Delete@Clem There was a memo from VTA CEO Carolyn Gonot to a recent VTA committee meeting where she outlined her concern that Caltrain was now able to figure out how much they owe for Gilroy service and an expectation that this would be used by SF and SMC to try to extract more money from SCC in budget negotiations.
DeleteAgree that it now should be easy to break out. But it has to run for a while, with no diesels running north of San Jose, to build some stats.
DeleteAlso to really break out the cost you'd need to keep track of details that might not be on record now. Like it's obvious for how long how many staff work on the running trains, and it's obvious what vehicle type specific spare parts cost, and also the electricity vs diesel bill is certainly easy to break out.
But what about maintenance staff? Hard to exactly break out how many minutes was spent on each train.
Also, what about the diesel locos acting as a stand by recovery option if there would be a major electricity failure and the EMUs needs to be fetched to go home, in order to not leave them parked worst case across level crossings, or best case "only" requiring security people to avoid "unauthorized paint schemes" being applied and whatnot.
Also, since the diesel trains now terminate at San Jose, I assume that they are shortened to fit the demand Gilroy-San Jose rather than Gilroy - SF with more people using the trains between San Jose and SF? If so that should save some fuel, but it might also allow for a slight time table improvement as the locos still have the same power while the total train weight is reduced. (I assume that the extra drag is negligible).
Adding to this discussion, this is why I really think that it'd be ideal to see some sort of reutilization of the diesel fleet to do one or more of the following:
Delete1. Extend the diesel service to Hollister. It's just 14 miles, and brings in San Benito County to help with some costs. You could get away with a SMART style shuttle.
2. Add in some Monterey County service during the day for Salinas to San Jose.
3. Extend the diesel service north to Levi Stadium/Santa Clara. You could gradually extend this service (along the Coast Subdivision) in partnership with Capitol Corridor to Newark/Fremont, and then to Coliseum. This brings in Alameda county to the Caltrain system, and sets up Caltrain nicely in the future to do some level of ring-the-bay service.
The idea is that by using the diesel fleet more than just a few times a day, it maximizes the utility and benefit of the fleet, and brings in four different counties: Alameda, San Benito, Santa Cruz & Monterey counties.
@Anon:
DeleteAgree on all.
Would the existing diesel rolling stock be able to also run to Hollister, or would it require more trains? The cars for sure would be enough as I assume that there are way more train cars than needed. But is the amount of locos enough?
Would be nice to see some sort of summary of what equipment Caltrain currently have, after a lot of the old diesel trains were moved to South America.
Gonot strikes me as a schemer, seemingly always on the side opposing cost transparency.
Delete@MiaM, the fleet composition is on Caltrain's website, but it hasn't been updated in a little while. Locomotives 900-919 were retired, as were all the 38xx and 40xx gallery fleet. That leaves the diesel fleet with nine locomotives and 41 Bombardier octagons. They should all be sold.
@Clem, would your thoughts change if the diesel fleet was repositioned to serve one or more of the East Bay, Hollister, and/or Gilroy-Salinas, something akin to a Capitol Corridor type of service that's all day and serves more destinations?
DeleteI think the biggest issues here are in no particular order:
1. Poorly utilized diesel fleet for a token line, including poor assumptions (e.g., only limited unidirectional commuter hours, compared to Caltrain between SJ - SF which was able to massively increase weekend ridership by offering more service all day).
2. Such poor land use that even if the diesel fleet was run more frequently for SJ - Gilroy it wouldn't change much.
3. Doesn't serve potential ridership generating locations or destinations: primarily north of SJ (Santa Clara/Levi, Oakland); with some potential destinations south (Hollister, Salinas).
@clem: Thanks for the fleet composition link!
Delete@Anon and others:
Re "poor utilization":
When vehicles have reached a certain age, the cost is almost linear to mileage, if we ignore the cost of storage space (and I assume that Caltrain won't rent out space at CEMOF to someone else if they get rid of the diesel trains). Thus it doesn't matter much what the utilization level is, and also it's almost no cost at all, except the lost income from selling off old vehicles, to keep some extra vehicles. This assumes that the vehicles are used often enough to not be affected by problems that happens when they sit unused for a long time. I.E. just make sure to have all the vehicles in rotation.
In fact I would think that the cost per mile would actually go down slightly if the Gilroy trains would be extended to Hollister, as there would be fewer cold starts / heat cycles per mile for a longer route. Don't know if that would matter much though. (And also I don't know how cold starts are handled - do the locos use (electric or diesel powered) block heaters before starting?).
I agree that the unidirectional commuter thing is a major problem. But since CEMOF is in San Jose, the trains must run in the other direction too. Do they run empty? Or are they stabled at CEMOF daytime and somehow in Gilroy at night? At least change any empty runs to running in service (even if it means skipping in-between stations for those runs, and maybe even have a "fuzzy" schedule, i.e. just say that the train departures some time between 10.30 and 11.00 depending on other track usage. I think it would still be worth having the train run in service. Just lock all but one car if the passenger numbers are low if that would save some cost).
One look at Gilroy in google maps would tell you there is a storage yard there. Also: there are significant fixed costs to maintaining a diesel fleet. Spare parts, training, time-based inspections, etc. It's not as if the cost scales linearly with miles operated. y = mx + b where b is the fixed cost. With old rolling stock (the diesel fleet is pushing 25 years, with some approaching 30) the m increases too.
DeleteOh, yes, zooming in shows that it's actually a fenced in area. Sorry, missed that.
DeleteStill, with the current staff turn around rates, for how long could the diesel service run without training?
A point of running the diesel service is to keep the service in existence until it's replaced with some other train service.
But also, as a a comparison: S.M.A.R.T. ridership is 3400/day, with 18 stations that is an average of 188 per station. I tried to find data for individual stations but couldn't find anything.
https://en.wikipedia.org/wiki/Sonoma%E2%80%93Marin_Area_Rail_Transit
A very qualified guess is that trips between stations south of Tamien is negligible, to compare the S.M.A.R.T. ridership values we have to multiply the data for each station on the Gilroy diesel service by two, as I assume that the ridership stats refer to entries but not exits. In that comparison, Gilroy station is on par with S.M.A.R.T. average, while Morgan Hill is slightly above and Capitol, Blossom Hill and San Martin have rather bad stats. Data source:
https://www.reddit.com/r/transit/comments/1gdh0ag/i_updated_my_map_of_caltrain_2024_daily_ridership/#lightbox
I see percentages posted elsewhere in this thread, but TBH the actual numbers and not percentages are what's needed to judge what's reasonable to run or not.
Also: While the numbers are low, the stats shows 319 riders (in each direction) which is about is 6-7 non-bendy buses on average with no capacity for fluctuations.
Also re the google photos of Gilroy: It shows four trains, one with five cars and three with four cars. If the average ridership is 106 per train (319/3) then why are they running four car trains? Don't know the specific specs of the Caltrain cars, but Wikipedia in general says that the Bombardier bi-level cars holds 136-162 seated passengers. In other words, one or two cars would be enough capacity wise and that would reduce the mileage related wear and tear of the cars two or four times, and also save some fuel and loco wear and tear when accelerating.
My main issue with this discussion though is why would Santa Clara county save money by removing this service, when the county is fine with with running one of the most inefficient transit agencies in general within larger US cities?
I know that it's apples and oranges, but the VTA light rail had an average of 263 passengers per station (Q4 2024, source: wikipedia page about VTA light rail). On one hand the system is way shorter, but on the other hand VTA runs a vehicle every 15 minutes most of the day, while the Gilroy service runs three unidirectional trains per day.
Also: Very few people who are comfortably able to drive their own car rides a bus, but some of them takes the train. So if the Gilroy service were cancelled then the travelers wouldn't just switch to buses, but a chunk of them would start driving. Do we want that?
I suspect the overhead imagery in Google dates from before EMU service, when the trains originating in Gilroy ran through to San Francisco. Today with the terminus in San Jose, they run three-car trains.
DeleteThe latest official Caltrain ridership statistics are available here.
Re all-stops south of Redwood City (one of the essential elements of The One True Service Pattern now and forevermore, amen):
ReplyDeleteIt's always been a happy conincidence (which is no conincidence, as the physics of train acceleration and deceleration versus stop separation and the demographics of station ridership dispersal versus stop separation are not unrelated) that choosing to have limited-stop trains skip the closest-together stations pretty-much delivers the best bang for the buck on aggregate trip time saving and least hit on ridership.
A simple exercise is to compare Caltrain's ridership guesstimates (because guessing is where they've gone since 2019; and despite the fact that automatic passenger counters -- much like working ETCS/ERTMS train control -- come standard on Stadler KISS trains purchased for 60% of the cost by everybody else on the planet) with the present "2004 called and wants its Baby Bullet back" limited-stops service pattern and other, superior, choices.
Here are the April 2025 guesstimate total riders by station:
San Francisco: 205538 22.2%
22nd Street: 33912 3.7%
Bayshore: 5737 0.6%
South San Francisco: 20118 2.2%
San Bruno: 12475 1.3%
Millbrae: 43735 4.7%
Broadway: 954 0.1%
Burlingame: 18458 2.0%
San Mateo: 35572 3.8%
Hayward Park: 10991 1.2%
Hillsdale: 40961 4.4%
Belmont: 17715 1.9%
San Carlos: 17946 1.9%
Redwood City: 61016 6.6%
Menlo Park: 24575 2.7%
Palo Alto: 94953 10.3%
California Avenue: 25613 2.8%
San Antonio: 20750 2.2%
Mountain View: 61659 6.7%
Sunnyvale: 51411 5.6%
Lawrence: 18891 2.0%
Santa Clara: 26491 2.9%
College Park: 843 0.1%
Cahill Street: 60132 6.5%
Tamien: 6211 0.7%
Capitol: 1018 0.1%
Blossom Hill: 1581 0.2%
Morgan Hill: 2855 0.3%
San Martin: 482 0.1%
Gilroy: 2407 0.3%
Some things to think about with respect to Santa Clara County:
56.8% of ridership is north of and inclusive of Redwood City
72.5 of ridership is north of and including California Avenue, Palo Alto
86.9% of ridership is north of including Sunnyvale.
1.6% of ridership is south of and including SJ Tamien
0.9% is on the Gilroy tail-waggging-dog boat-anchor black-hole.
* San Bruno is far worse performing than I guessed.
* San Antonio somewhat better.
* Bayshore and Tamient are dogs and ought to be mothballed.
* And of course Broadway, Hayward Park and College Park closed permanently, starting decades ago.
...
...
DeleteAnyway, applying simple per-station ridership over total ridership percentage cuts, we crank out some simple service patterns:
* 3%: SF, 22nd, Millbrae, SM, Hillsdale, RWC, PA, MV, Sunnyvale, Cahill
* 2.5%: SF, 22nd, Millbrae, SM, Hillsdale, RWC, MP, PA, CalAve, MV, Sunnyvale, SClara, Cahill (adds: MP, CalAve, SClara)
* 2.0%: SF, 22nd, SSF, Millbrae, SM, Hillsdale, RWC, MP, PA, CalAve, SAntonio, MV, Sunnyvale, Lawrence, SClara, Cahill (adds: SSF, SAntonio, Lawrence)
(Aside: 3.5% cutoff has the same stops as 3%. 4% would lose 22nd, SMateo, SClara)
Comparing to Caltrain's two (2!) different peak-only limited-stops patterns
* SF, 22nd, SSF, Millbrae, SM, Hillsdale, RWC, PA, MV, Sunnyvale, Cahill
* SF, 22nd, SSF, Millbrae, SM, Hillsdale, RWC, MP, PA, CalAve, SAntonio, MV, Sunnyvale, Lawrence, SClara, Cahill
there's a good match with a 3% cutoff with the 1tph 2004 Baby Bullet Got Back limited stops, with the curious special-case-pleading exception of under-performing South San Francisco;
there's an exact match for 2% cutoff with 1tph "Silicon Valley Limited" One True Service Pattern making all stops south of RWC.
Anyway, just slightly the dial from "3.0" to "2.0" (and no it doesn't go up to 11) and we get the One True result we "should" get, and without pleading extenuating circumstances for SSF or any other particular station. Do it!
FYI I've long believed in stopping all trains at SSF, as Caltrain has started doing, and that it will realise higher potential in the short-medium term than we see today. And I KNOW from direct experience that stations like Menlo Park, CalAve, Santa Clara are woefully under-served today. Back before the 2004 "Baby Bullet" timetable, California Avenue's platforms could be more crowded than Palo Alto's with riders hoping to get on peak hour limited trains. Stanford Business Park is still there, but Baby Bullet killed off nearly all the connections.)
;SInce I had Caltrain's guesstimate numbers to hand: each station's percentage of total system ridership growth 2025-04 vs 2025-04
DeleteSan Francisco: +8.5% +78687
Palo Alto: +3.0% +27440
Sunnyvale: +2.8% +26158
Mountain View: +2.3% +21654
San Jose: +2.3% +21397
Redwood City: +2.3% +21000
California Avenue: +1.6% +14597
22nd Street: +1.5% +13716
San Mateo: +1.4% +12969
Hillsdale: +1.4% +12817
Millbrae: +1.4% +12530
Santa Clara: +1.2% +11213
San Antonio: +1.1% +10597
South San Francisco: +1.0% +9585
Menlo Park: +0.9% +8548
Lawrence: +0.8% +6948
Belmont: +0.7% +6784
Burlingame: +0.7% +6467
San Carlos: +0.7% +6242
Hayward Park: +0.6% +5471
San Bruno: +0.6% +5351
Bayshore: +0.3% +3111
Tamien: +0.2% +1469
South San Francisco quite disappointing, in light of biggest service upgrade of any station.
California Avenue greatly outperforming, despite "Baby Bullet from 2004" exclusions, which should be the surprise of nobody with any clue at all.
San Antonio outperforming.
Menlo Park and Lawrence, skipped by "Baby Bullet form 2004", doing compararively worse than likewise-underserved CalAve, SClara, SAntonio, to my mild surprise.
Richard it's also worth noting that if one looks at land use in Palo Alto, and considers TOD (heaven forbid), the area around CalAve, especially if you expand out to where the old Frys used to be, is a very logical place to site higher density high rise condos. Making a Bay Area version of a metrotown. It's taken forever, but Sunnyvale should be commended for building up density around Sunnyvale Station. It continues to boggle the mind that my fellow denizens of Santa Clara County mostly can't seem to wrap their head around the idea that much higher density needs to happen on the CalTrain, BART and ECR corridors and that long range land use and transit planning need to be have been better coordinated.
DeleteNo one asked, but I'm waiting for a politician with vision to suggest that a BART extension beyond Diridon go not towards Santa Clara (which is already served by CalTrain), but instead head West down San Carlos/Stevens Creek and terminate near DeAnza. Watching Cupertino City Council heads explode with the thought of an elevated guideway down the center of Stevens Creek for BART would be a lot of fun.
Kitsune - Big fan of your ideas here. Well designed density around stations is key to the vitality of both cities and the transit systems. Hopefully San Jose can figure this out with downtown, as it has so much potential for housing and for densification, and it's super fustrating. If they can add in 50K-100K residents in the downtown core, we can really see it become a third major city core like SF and Oakland.
DeleteAs for BART - yes. In my opinion, BART should go down Stevens Creek at minimum. I'd elevate it, build it down San Carlos and Stevens Creek from Diridon. Tunneling is not necessary, and expensive. However, at DeAnza, there's a rail line paralleling 85 that isn't used anymore since the Pernamente Quarry shut down. I'd sign that over to VTA or Caltrain for potential new service, but using some of the right of way to store BART trains as needed.
The second phase would go north roughly following 85 to get to El Camino Real. You could take the Santa Clara section of BART and route it up El Camino Real for a loop from Santa Clara BART - Santa Clara Koreatown - Sunnyvale - Mountain View, south to De Anza - Cupertino - Apple/Vallco - Kiely - Santana Row - Diridon.
This, tied in with an automated metro system like AirTrain for SFO from Diridon - SJC (at most 4 miles in length), gives us a seamless transit network for the San Jose/Silicon Valley region. Alternatively, you could do VTA light rail here that goes Diridon - SJC - North First & Skyport (for transfers) - the industrial area roughly around Gish Rd and 880 - Berryessa BART - Capitol & McKee - McKee & Toyon. This would be better for regional connectivity, but worse for SJC service, fwiw.
Add in some tail or stub tracks to go in various directions as needed over time, and now you have a nice integrated network for all of San Jose. I'd also look into converting some of the VTA lines to full grade separation to fill in the gaps in the BART and Caltrain regional backbone (e.g., speed up the VTA Green Line to Los Gatos past Vasona Junction). In this way, we can see BART and Caltrain operate as an S-Bahn or RER type system while VTA operates as a more local and U-Bahn type system.
@Kitsune and Anonymous:
DeleteGood idea to extend BART westwards on Stevens Creek. Would kind of sort of bring back the Peninsular Railway, or at least one of it's routes. The history that there used to be electric rail transit along this road probably also is a big part in why it has a higher density than it's surroundings.
https://www.saratogahistory.com/History/PIrailroad.htm
https://en.wikipedia.org/wiki/Peninsular_Railway_(California)
The railway that goes to the Permanente quarry seems to be owned by UP (source: regrid) and the disused railway northwestwards seems to partially be just the right-of-way of the public roads (Fremont Ave / Foothill Expy) up to Alta Mesa. Then the ROW seems to mostly be owned by Leland Stanford University, except that the last block next to the Caltrain route seems to belong to private companies in general. More importantly except for a few places like maybe Los Altos it seems to be more or less only low density sprawl along the route, so the viability of transit on this route seems low. Unfortunately the same seems to be true southwards from where it intersects with Stevens Creek all the way to down where an extension of the VTA green line light rail would end up. Don't know how suitable/unsuitable the ROW would be to store BART trains, but for extending BART (and/or the VTA green line) along this railway I think there is a need for a lot of increased density.
Btw it feels like VTA seem to have bad luck when making decisions in a lot of cases. Part of the light rail seems to have been built where it was possible to build, rather than where it makes sense. Like for example the southern end of the blue line should obviously connect to Blossom Hill station on the Caltrain route, and not end up at a parking lot that is too far away to walk. Given the low density sprawl, the parking lot is needed but having it at a place with both VTA light rail and Caltrain (extent the electrification to at least Blossom Hill) is way better for the passengers. Imagine riding the VTA blue line rather than Caltrain for that distance, just because the walk from the parking lot is too long and the Caltrain frequency is too low - super inefficient use of potential transit infrastructure.
Re connecting Diridon northwards - the Caltrain route is capable of a train every 5 minutes. If they want they could just buy a few say 2-3 car versions of the Caltrain EMUs, and shuttle them between Diridon and Santa Clara station, rather than spend a gazillion on a weird bend on the BART route.
Speaking of Diridon - Santa Clara - looking at Regrid it seems like UP owns quite a bit of the land that CEMOF is on, while the public owns the westernmost part of that area. I wonder if there are some gnarly parts of the contract with SP/UP that makes Caltrain unwilling to change up the CEMOF area?
Fun Fact- The distance from BART's planned Diridon Station to the end of line in Santa Clara is about 13,000'. If you go west from Diridon on Alameda, diagonally under the Hanchett neighborhood, and then out San Carlos and Stevens Creek Blvd, you reach Valley Fair/Santana Row in about 13,000'.
DeleteWorried about a yard under that scenario? Bulldoze the outdated commercial strip buildings along San Carlos and build a cut and covered yard, deck over it, and redevelop. Hudson Yards of Silicon Valley!!! Of course, none of this was investigated. It had to go to Santa Clara, so everyone can have a choice of FOUR rail options from Diridon to Santa Clara– Caltrain, BART, Amtrak, and ACE.
Is a separate stabling yard even necessary? I.E. can't the trains just be parked on the running line between the end and Diridon? Fill one track to the brim and the other as much as needed. Those who live near the end would have to suck it up in that the first morning trains and last evening trains don't go all the way to the end station. Or maybe stable trains on one track for a longer distance, allowing the first morning train and the last evening train to call at the end station, if that is desirable.
DeleteIt sounds to me that Caltrain could sell the fleet to Monterey County and allow them to run a train service that still stops in Gilroy and Morgan Hill.
ReplyDeleteThat's a great idea - subcontract with Caltrain to provide that South County service, with integrated ticketing, and the only difference would be that Monterey County takes on the costs, while Santa Clara County/VTA can reimburse Monterey for the Gilroy - San Jose segments. Both counties would save money and get a decent level of service.
DeleteWould Monterey County be that thrilled of the idea of subsidizing transit within Santa Clara county though?
DeleteAlso what about negotiations with UP re track rights?
I know that the freight companies threatened to run all freight via road if Cali would actually enforce zero emissions for freight railway. I think it's time to call this bluff and also offer the freight railways an exemption if they are willing to sell off the full route Sacramento-Oakland-San Jose-Los Angeles-San Diego, all routes that are used or ought to be used by Metrolink, the two San Joaquins routes north of Merced (and/or whichever route HSR phase 2 would take to Sacramento), and the ACE route, and any stubs from the coastal route that might be interesting for passenger services, like the one to Hollister, and also any other routes suitable/in use for passenger services.
Combine that with a toll on road freight at various places where it's easy to toll everyone, I.E. where you don't need that many toll stations to ensure there isn't a detour to avoid the tolls.
As I understand it, a lot of the freight on rail within California runs between the ports and various places in USA outside California. Given the political climate, for example related to Cali HSR, it might actually be a good idea to do something that potentially disrupts neighboring "red" states. The main goal with calling the freight companies bluffs is to either force them to electrify everything, or sell of the passenger routes, but it would seem like a good idea to also punish neighbors for voting for a president that is against Cali HSR.
MiaM, not all rail in the USA can be electrified. However, as has been said before, some of it can, the major rail routes in metropolitan areas and beyond them into not only what has been the outer commuter shed for so many years, but to a place where locomotives can be switched. The classic instance of this is in the L.A. area and it includes something you have noted:
Delete"As I understand it, a lot of the freight on rail within California runs between the ports and various places in USA outside California."
The Alameda Corridor, Alameda Corridor East to reach the warehouses that have come to the eastern (outer) Inland Empire, and Cajon Pass can be electrified, all the way to Barstow, where the Tehachapi route intersects it. Yard activity can include locomotive swapping, electric and Diesels, where the wires end there. The Tehachapi route could also be electrified, all the way from Bakersfield to Barstow.
That's one example.
MiaM, don't forget that major truck, and bus, through routes can be electrified, including the main arterials, any streets connecting them on truck routes and other especially busy connectors used by trucks, the freeways, and as with rail out past the metropolitan areas to where tractors can be changed. As with rail, too, energy savings come in no small part from electric transport in place of Diesel over major mountain passes.
DeleteIt's amusing to see not only Santa Clara County, and Monterey Bay and Salinas, access discussed, but continuing onward between Salinas and San Francisco.
ReplyDeleteFirst, maybe sometime soon Clem may do more census depictions of persons and jobs within given distances of Caltrain stations, similar to
this earlier example, before the latest tech bubble and the pandemic, a census process described here. If it's ever done again sometime soon, you'll see not only more growth in the southern Peninsula, but continued support for extending electrification south to Blossom Hill..
Second, regarding Union Pacific: Do people really believe UPRR generally will have a different philosophy and strategy toward HSR Stuff on or near UPRR's property than expressed here, in 2010, which was 25 years ago? Note the mistakes by the HSR project identified in that letter, indicating gross ignorance or problematic behavior associated with "mistakes," beyond mistaken assumptions to specific, concrete examples of things assumed without permission or agreement with UP, likely without previous communications, and which UPRR rejects. (Rejected 25 years ago)
ReplyDeleteAh, the Coast Daylight (Siemens Venture trainsets done like Brightline's or plainly as with other state trains) and improved service alongside US 101 for the kind of free advertising most only neglect to realize alongside I-5 tangents...
ReplyDeleteThird: There has been specific interest before in making improvements to the Salinas-San Luis Obispo route, the Coast Corridor as named in the study that counts first and foremost, with boring conventional stuff in it. Here is a page with links to others on it, the best being 2.0 Alternatives that describes the changes., with links on it to other report sections.
If you want fun browsing about this route and others in the inventory of the state, and looking at some others' early vision of improvements to those routes, in addition to high-speed service, you can look here, which was one of many publications from planners under the late Peter Hall. Diesels are obviously the fit on the Coast Corridor and greater Coast Route service runs, and the group examined 110 and 125 mph with some arbitrary details assigned to each. That document is of use as a reminder that what needs to be kept in mind is a complete state system.
110 mph: Using 4000 HP diesels, the one-stop express time between Los Angeles and San Francisco would be 6 hours 27 minutes, with an average speed of 74 mph. Local service would take 7 hours 23 minutes, with an average speed of 64 mph. Skip-stop trains would take 6 hours 35 minutes, with an average speed of 72 mph. Use of tilting trainsets could substantially reduce travel times for this alternative. Tilt-trains would reduce express service by 72 minutes (19 percent), local service by 61 minutes (14 percent), skip-stop service by 73 minutes (18 percent).
125 mph: One-stop express travel between Los Angeles and San Francisco would take 5 hours 27 minutes, averaging 86 mph. Local and skip-stop services would take 6 hours 20 minutes and 5 hours 34 minutes, respectively. As with the 110-mph alternative, the use of tilting trainsets could greatly reduce travel times. Tilt-train service would reduce the express service time by almost 16 percent, the local service by 3 percent, and skip-stop services by 16 percent.
True modernizing would involve the Santa Margarita-San Luis Obispo base tunnel under Cuesta and also would abandon the lonely coastal section for inland, better service of the overall population south of Santa Maria in particular, with a Buellton station serving this function and using the Nojoqui gap to avoid a tunnel farther east, Capitan-Santa Ynez.
Note: Do you realize what was possible with the state and the entire Coast Route in the early 1990s, too? Yes, it was possible, it really was.
Clem, a new jobs-and-residents census and accompanying Caltrain diagram like the earlier one would be interesting sometime, before or after any development including large employers in the Diridon Station Area, which remains speculative and seems unlikely. Even without it the jobs boost on the southern end of the line through the southern Peninsula would make it more like a dumbbell, especially if jobs on-site, not remote, were included, possibly, and the hit to San Francisco would be significant at the same time.
ReplyDelete"Acquire UPRR's Coast Subdivision and electrify to Blossom Hill"
ReplyDeleteAll I can really say is "Is this a hostage video?"
Is your family being held in the basement at VTA HQ? Blink twice if so, Clem!
"As compensation for ..."
San Jose/VTA does not accept your tribute.
SJ accepts no tribute.
YOUR CHILDREN MUST DIE SO THAT MINE MAY THRIVE.
"... there is no reason for Caltrain or VTA to fight alone. The state should ... "
Caltrain and VTA, indubitably and long-ago proven to be the very worst negotiators for the public interest on the planet, are clearly not bad enough! The full force of the budget of the State of California, the seventh largest economy of the planet, should be bought the bear on the vital issue of enriching Union Pacific Railroad, enriching it behind the dreams of avarice.
Caltrain to Tamien a near-zero-ridership dog that sucks immense capital and operating resources and delivers nothing? The problem is that there isn't enough of this! Do more!
VTA "light rail" through South SJ (and everywhere else in SJ, and everywhere else in Santa Clara County) a proven failure? We didn't spent enough on rail in the hideous isotropic freeway sprawl hell of south SJ! One more line bro, just one more line.
BART extensions of nowhere exurbs a proven failure? Extend more!
Muni Central Subway proven to to be a total fraud? Extend to Fishermans Wharf!
Seriously, screw failure. Screw perpetuating failure and fraud through More of the Same.
"Freight trackage rights would be preserved, and ..."
Why should UPRR agree to this? Name one reason, one reason that is not followed by a dollar amount involving eight of more zeros (which even VTA/PCJPB/California aren't going to sign off on, because BART already ate all of their eight-or-more-zeros lunch money), why?
"If this sounds like a megaproject, it isn't."
Caltrain can't even built ONE SIMPLE TURNOUT AND 1000 FEET OF TRACK WITHOUT ELECTRIFICATON, inside ITS OWN FULLY-OWNED FULLY-CONTROLLED RIGHT OF WAY, for under 10 million dollars.
ANYTHING involving ANYTHING AT ALL on Caltrain is now a fucking megaproject. Because profit.
Building one bog-simple two-track rail bridge across one simple road (Broadway in Burlingame) where Caltrain completely owns and controls the right of way? $700 million. 700 million OR MORE! MUCH MUCH MUCH MUCH MUCH MUCH MORE!
1000 feet of non-electrified non-revenue track in San Mateo? $10 million.
This is insane.
Caltrain ridership south of Sunnyvale of negligble.
Running (empty!) trains to Blossom Hill or anywhere other terminal point in South SJ sprawl is going to involve new platform construction (they'll demand at least two platform faces), turnback tracks (they always want turnback tracks), mid-day and overnight storage tracks (ditto). And that's all ignoring the skys-the-limit knee-capping demands that UPRR can and willl demand.
Caltrain ridership south of "downtown" San Jose (= Cahill Street, not Tamien) is a rounding error. Always will be.
Just stop it! All Caltrain service should terminate at SJ Cahill. The single-track once-an-hour Tamien extension, like the discontinued VTA single track one-station Chynoweth–Almaden light rail joke, needs to be terminated, not extended. There;s no there there. There's never going to be any there there to serve. SJ made its bed, now we get to sleep in it. But not run empty trains to it. End of story.
Re south SJ sprawl census tract population, as a proxy for ridership ... the sprawl is sprawl, cul-de-sacs designed to be actively incompatible with bus service or walking, all bsected by wide high-volume freeways and median-divided expressways. Even the housing that is somewhat close to the UPRR freight line "turns it back" on the combination of four-late 140-feet-wide Monterey highway and ~100-feet wide UPRR freight line. (Bonus at existing Blossom Hlll station site: another 140-foot-wde divided highway stroad on the other side! Bonus at existing Capital station site: an actuall quarry on the other side.)
DeleteSuch office park employment sites as exist within striking distance of the UPRR tracks also actively turn their back on it, or are sited on the other side of a freeway from it. I mean, that's precisely why these places were built -- cheap, subsidized land, freeway access, no transit.
Also, the UPRR freight line is pretty much at the eastern edge of the metastatized sprawl. In positive contrast, VTA's zero-ridership light rail line along a freeway median to the parking lot known as Santa Therea station, kind of bisects the sprawl, is "connected" to the sprawl by divided-median expressways, runs at unsuppoortable headways all day, and still has zero ridership.
Caltrain lumbering back and forth alongside UPRR tracks along Monterey Road is going to do much much worse.
If there were rail ridership potential from South SJ to points north of "downtown" SJ, there would, today, be non-negligable ridership on that VTA light rail line, and there would be non-zero passenger transfers at Tamien.
There isn't.
EVERYTHING about the suburban sprawl form says "drive". That's what people do. They take the hint. They're not stupid. They don't rde VTA. They're not going to go miles out of their way to drive to Caltrain when they can just drive directly where they're going, just like SJ wants them to do.
There's no there there. There is nowhere south of SJ Cahill on the UPRR tracks you could point at on a map and say "here's a good spot for a train station:". Not one! That's not how ths part of the world was subdivided! There's no ridership potential. (Even grasping-at-any-straws-at-all CHSRA saw zero GIlroy-South SJ potential and deprioritized any spending there.). It's hopeless! JUST LOOK AT ANY AERIAL MAP.
Re: "It does not require any new trains, as Caltrain's EMU fleet will soon swell to 23 trains, where today's service pattern only requires 14."
DeleteIf there's one thing this blog knows about, and has been on about for two decades, it's marginal operating cost. Also known today as "Fiscal Cliff! Fiscal Cliff! The Sky is Falling! Fiscal Cliff!"
And Clem even comments later above "The weird thing in California (and the U.S. in general) is that there is plenty of money to build stuff, just not enough to operate it."
So let's talk about the operating cost overhead of shuttlng empty (EMPTY!) trains from SJ Cahill to SJ Blossom Hill.
FIrstly, if this happens at all (it shouldn't and musn't!), it's going to be at peak commute hours, when trains in service and crews drawing salaries are at their peak level.
The ~8.2 miles (~13.3km) one-way trip south of SJ Cahill Street, creeping (35mph limit, through presently 15mph!) to stop at SJ Tamien, then somewhat faster past the quarry to SJ Capitol and terminating at the marker on a map that is the SJ Blossom Hill station is going to take on the order of 12-15 minutes each way. (Not very meaningfully different from today ... electric zoom power doesn't really buy anything here, and of course never-level-boarding-never Caltrain dwell times are barely affected.)
So to dead-head service through SJ Cahill Street station -- the southern-most end-point of any actual passenger demand on the corrdor -- to Blossom Hill simply n order to reverse will cost, at peak
* 3 trains and 3 crews at 15 minute headway (or if Caltrain went to BART-copy 20 minute headways)
* 2 trains and crews at 30 minutes
* 1 train and crew for hour "don't even bother" headway.
* (4 trains 4 crews at never-happening 6tph)
I mean. sure, we know that Caltrain's sub-moronic world-worst rolling stock procurement consultants and staff purchased too many far too large and hopelessly inflexibly mis-configured trains from Stadler (motto: "We'll built anything, no matter how stupid, if you're paying, but OMG EPIC EYEROLL") and that there are a bunch of really big and quite shiny KISS trains just lying around, but ... who has the year-after-year OPERATING BUDGET to run these, and run them empty?
At 4tph, we're looking at a premium (ie blow-out of ever-scarce operating budget) of 25% for extra crews and trains, to run empty to reverse in a nowhere by the side of a highway. (12 trains needed for 4tph SF-Cahill, 3 more needed for Cahill-Blossom Hill.)
At never-happening 6tph it's 22% (4 for South SJ on top of 18 for where there are actual riders.)
These sorts of numbers can't possibly be justified, even if some idiots did buy too many trains and bought them at a 60% price premium (and hoo boy that US consultant cost-plus pile-on was BEFORE Trump Tarrifs...) Caltrain could own 70 sets of KISSes, but there's nobody who has the operational budget cash to pay to send them, along with with their wage-drawing "enginner", "conductor", "assistant conductor", "assistant conductor's factotum" "boatswain", quartermaster" and "cabin boy", back and forth into the hopeless sprawl of South SJ.
And, yes, you can be 100% certain that the City of SJ isn't ever going to be the one paying for such a purely intra-SJ boondoggle.
Demand terminates at SJ Cahill Street. Caltrain terminates at SJ Cahill Street. The end.
I would treat any separate diesel shuttle just to serve Blossom Hill and whatnot as a proof-of-concept, in order to judge if an all-day 15/30 min service would increase ridership or not, and if so go ahead with electrification.
DeleteBtw, luke warm take: Build the new HSR line Tamien - Gilroy elevated above Monterey Road. That would result in a 100% grade separated route while not having to deal with UP at all, or at least minimally. I would like to see the faces of the people at UP who thought they would earn lots of money on allowing a HSR route to be built on their land.
For a possible interim electric service to Blossom Hill, just remove a lane or rather the median of Moterey Road. Again it avoids having do deal with UP.
People aren't made to drive. They prefer to drive. Nothing beats door-to-door, your route, your schedule. The problems arise once too many people want to drive to and from the same places, and along the same routes to and from these places, as traffic will be channeled in some way -- and the cars come with them, which must be parked nearby when not in use; or, with magical automated cars, they will be circulating on the roadway and congesting it when not in use or when deadheading, or when circling around the places with most demand or big businesses, or parked nearby. (The same problems happen with more-magical automated aircraft for personal use or air taxi use.) Downtown San Francisco is a typical breaking point example. For many who can take it, Caltrain and BART are a solution to the problem to get in and out of downtown with ability to range farther from downtown before returning.
DeleteThe Santa Clara Valley's good, hearty, healthy, wholesome sprawl with modern or contemporary roadway dimensions (with many mistakes in development like houses along four-lane streets, either arterials or too-wide collectors) is like most of Orange County's that is, other than the earliest built. It's nothing new or any surprise to (real) Californians.
It will be better to spend money (capitol / operation) on core SJ-SF business. For Safety, all station need grade separation, especially Redwood City. Mountain View and Sunnyvale. Redwood City need 3 or 4-tracks elevated station allows SF-Redwood City local train turn around and express train connection. With this facility, most of SJ-SF train become express between Redwood City and SF.
ReplyDeleteI updated the example model scenarios of Ye Olde Taktulator (Clem's bog-simple train performance data from 2011 still rocks! I've only made some UI changes to the crusty but still valid code since then) to the minor timetable revision Caltrain made in January (compare September 2024).
ReplyDeleteShort story: dwells got longer; now modelled as 50 seconds (would be 20 with level boarding on a normal non-Caltrain KISS, sigh) on the limited-stop services, and an a slightly extended dwell for schedule recovery has been added at Palo Alto (an OK idea in general.)
Nice change: the schedule is now almost completely regular northbound vs southbound, AM peak vs PM peak!
The diesel Gilroy trains are the obvious outlier, and there's still the utterly inexcusable (do the Jesuits have pee tape kompromat on Caltrain?) Bellarmine boys school private stop aka "College Park", but slowly slowly some is indeed positive change.
Off-peak (mid day, at least with decreased regularity early morning and evenings) service is the best one might expect: take the peak service, remove the limited-stop trains, and the remaining all-local schedules and stop times are identical peak vs non-peak.
And, as we can see, the Taktulator (with terminal turnback/recovery times modelled at a very sluggish 15 minutes minimum), accurately reports a required fleet size of14 trains and 14 over-sized crews (8 off-peak); the same number that Caltrain staff reports.
So, since we're on the topics of South San Jose Caltrain and of service cost and FISCAL CLIFF FISCAL CLIFF WE'RE ALL GOING TO DIE ...
ReplyDeleteLet's look at various minor but effective one might make to today's (effective 2025-01-27) Caltrain timetable (the title of the model says "AM Peak" but recall that PM is, now, hooray, finally the same service pattern as AM aside from one-way Gilroy branch diesel connection irrelevancies.)
First, we can, as Clem suggests above, turn off the atavistic 2004 "Baby Bullet" service pattern which skips station stops south of Redwood City and replace it with another "Silicon Valley Limited" on even 30 minute headway. (Also turning off Gilroy diesels to avoid AM/PM distraction.) Result: No change in fleet/crew size (still 14); very slight increase in 2011 Taktulator timetable quality score due to even 15-minute Santa Clara County headways outweighing negligible one-per-hour "Baby Bullet" trip time "savings".
SCIENCE says: Seems legit, man! Do it.
Next step, we can turn off today's 1tph extension to and from SJ Tamien.
DeleteAs we know from Caltrain's official monthly ridership guesstimates, ridership to and from Tamien represents 0.7% of total Caltrain corridor total ridership (6211 of 925000 for April 2025)
But ... holey moley! WE HAVE SAVED AN ENTIRE TRAIN AND CREW, a 7% improvement in naîve operating costs (yes, Caltrain has many, far far far too many fixed operating costs on top of this, and there are operating crews and trains in reserve beyond the nominal minimum to run the service, theoretically), but still ... but still ... saving anywhere near 7% of your peak crew wage burden while shedding 0.7% of your ridership is a pretty damned great trade-off.
To nobody's surprise, it turns out running empty trains beyond SJ Cahill Street to a near-zero-ridership turn back location in the freeway hell of South SJ has real costs, today. (My surprise is that axing 1tph round trip-to one station, two mouse clicks in the Taktulator, without tweaking a single other thing paid off so simply and immediately.)
So yes, there are costs versus benefits to South SJ and South-of-SJ service that nobody should even consider paying. It's nuts! Advocating running even more trains even further south into the featureless transit-hostile-by-design sprawl where nobody rides, rather than turning the trains back into revenue service at the spot where actual ridership and revnue ends is ... worse than nuts.
SCIENCE says: You're insane not to do it. FISCAL CLIFF ARRRRRRRRRGHHHHHHH
(Neglected to mention: axing Tamien reduces minimal off-peak train/crew requirement from 8 to 7. Not as bottom-line important as minimizing weekday peak, but still ... Woo hoo! Especially for weekend shifts)
DeleteBut we're not done!
DeleteThere's this ONE WEIRD TRICK Caltrain could also do, today, to reduce peak train/crew requirements at peak and almost without changing anything! Money for nothing, my friends! Step right up!
If one looks at the existing timetable (or the fundamentally similar "Caltrain 2025 sans Tamien" scenario above that already saved 1 of 14 peak trains/crews) one can see in the helpful Taktulator output that trains/crews fritter away a ton of time, at peak hours, when they're most needed, just being parked, not carrying passengers, at the SF Mission Bay terminal. Turnback times are either 27 minutes or 40 minutes -- way more than the already-glacial 15 minute minimum we assume in light of Caltrain's slovenly timekeeping and chaotic no-level-boarding-no-never dwell time stochasticity.
We're literally burning train crew wage and cash and watching our billion dollar train fleet sit around depreciating by parking them out of use when they're most in demand.
But there's this One Weird Trick one can pull in the Taktulator to make the real world cheaper. At at no cost! Just change a couple numbers and click "Generate new timetable"!
Simply change the "NB minutes" (SJ Cahill departure time) of the "Santa Clara Limited" train from 43 to 29 (ie from :13/:43 past the hour to :29/:59) and change the ""Local" northbound departure time from -32 to -46 (ie from :28/:58 to :14/:44) This is just swapping the order of the northbound limited and local trains, so no real change to anything. It's that simple.
Result: peak hour train arrivals and departures are more evenly spaced out in SF (while remaining unchanged in SJ), meaning ... 17 minute (more than Caltrain-sloth assumed 15 minute minimum) turnbacks in SF at peak, not 27 or 40 minutes of slack, meaning ... ANOTHER PEAK TRAIN AND CREW SAVED. Down to 14 (today, with "Baby Bullet" and Tamien), or 14 ("Baby Bullet" put to bed after 20 years), or 13 (Tamien mothballed until circumstances radically change), to 12 (no Tamien, swap northbound local/express order.)
Money for nothing! 12 is a number smaller than 14. Same or better level or service, done more efficiently.
We've gone from
* Fleet size: 14 trains (not including spares)
* Fleet utilization: 68.5% (265 non-revenue minutes of 840 train-minutes/hour; revenue distance 614 train-km/hour, 44 km/hour/train)
to
* Fleet size: 12 trains (not including spares)
* Fleet utilization: 80.4% (142 non-revenue minutes of 720 train-minutes/hour; revenue distance 609 train-km/hour, 51 km/hour/train)
and the only is mothballing one hopelessly under-performing station that serves an average of only 207 riders/day.
Pretty sweet.
FISCAL CLIFF AIIIIIIEEEEEEEEEEE!
You make excellent points! South San Jose is indeed configured as transit-hostile, circulation-hostile car-oriented sprawl, and population density by itself is a necessary but not sufficient condition for high ridership. It's hard to argue against the terrible ridership of the VTA light rail to Santa Teresa.
DeleteMinor counter-point: if Caltrain EMU service were extended south, there would obviously be track improvements to enable faster movements through SJ and Tamien without the absurdly low speed limits currently in effect.
In your "focus on the core" scenario with just 12 trains, with nearly half the EMU fleet (eventually 23) sitting idle and depreciating, I wonder if there would be a positive change in preemptively going from a 30-minute base takt to 20 minutes? This would conceivably use ~18 trains in the peak if the phasing you came up with works at that interval. The revenue bump from more frequent service and the attendant ridership increase might outweigh the increased operating cost, since the "large fixed cost of operating a railroad" they constantly yammer about becomes amortized over more service.
Bonus: 1:1 connection to BART at Millbrae since both systems would have 20 minute takt, although that connection might now dictate the phasing of Caltrain... turning it into the BART peninsula line all but in name.
Clem, I'm wondering if your thoughts on going farther south than the Cahill (now Diridon) station to beautiful, wondrous Blossom Hill have changed since you wrote about it earlier. That's real thoughts, not just activist-speak about cars and sprawl and stuff, with the Santa Clara Valley just being newer, like other than the oldest Orange County areas, for example, compared to older, smaller-scale (often substandard now) places. Is the southern part more like the line's San Jose Appendix now?
DeleteDemographically, the southern half of San Jose is a rich but poorly tapped source of commuter ridership, with dense residential neighborhoods surrounding the corridor. More than 100,000 people live within two miles of the Tamien and Capitol stops, and 75,000 people live within two miles of the Blossom Hill stop. Census data argues strongly for locating the Caltrain terminus at Blossom Hill, with an electrified train storage yard / layover facility in this large vacant space [UPDATE: that large vacant space seems to be spoken for, so look for other unbuilt spaces in map at left], a far better place to park out of service trains than in the middle of San Jose Diridon. ACE and Amtrak trains could be turned at the existing Tamien layover facility.
Turning all Caltrain service at Blossom Hill would improve service for hundreds of thousands of San Jose residents and workers, at some increase in capital cost (to electrify) and operating cost (12 minute longer runs). On the other hand, it would greatly reduce Caltrain's requirement for tracks and platforms at Diridon station. Caltrain would operate through the San Jose Diridon station much like it does at the Palo Alto University Avenue station, using just two tracks and two platform faces. If that seems hard to imagine, remember that Palo Alto has almost 60% more ridership than San Jose Diridon; any perceived need for all those tracks and platforms at Diridon, and the profoundly mistaken notion of a "South Terminal", arises from existing jurisdictional boundaries and Caltrain's unhealthy habit of parking trains in the worst possible place to park trains.
https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgcgBkdiHYtpIgoZqpupkgNCLEWn2EZWAv3AXNIf8nEqcfnp8e14-EIdZRgXRQ2k_VxPk8P6vSB9zmdF-HULYEX-WSDCKW5Sq3TPUD-t2VU-HdYJiAwicUVRXtDKk3HuPHMhlw6aGIU1LCv/s1600/san_jose_census_data.png
https://caltrain-hsr.blogspot.com/2017/01/san-jose-done-right.html
So much for relying on the census data link, too. It's on the page at the last link, anyway, like the text.
DeleteThere will be a great opportunity to eliminate Tamien after the summer shutdown for bridge replacement. It's a lot harder politically to cancel service than it is to just never restart service after a temporary shutdown. See: Antioch.
DeleteAssume Caltrain has money to operate fleet size of 14 EMUs 5 days a week. How about allocate fleet size of 2 equivalent to weekend service? Weekend rider pay more chance of "Full" fare.
DeleteWith this resource, weekend schedule can run Limited train (SF - Millbrae - Redwood City /local/SJ) and Local train SF - RWC both 30 min frequency? This scinario will need 5 train set each of express and local with 15~25 min of turn around in each terminal. Limited stop train stops only Millbrae before arrive San Francisco. Traveling time of SJ - SF 64~65 min should be very attractive to leisure customers and SFO traveller. Only the question is Turn around local train at Redwood Junction.
Bridge replacement -- no doubt with ped-bike access included somewhere
DeleteA 12 foot or wider roadway with bridges and tunnels can serve maintenance and repair functions, can help with evacuations and with police and fire, etc. in emergencies, and can be open to non-motor vehicle use (won't exclude EVs easily) otherwise.
That's what's missing from the new Hudson rail tunnel project. Dolts
If anyone has seen what evolved from Bergen's light-rail tunnel's separate escape tunnel parallel to and next to it, well, nobody can expect another, but the point is it's nice for bikes and peds to be able to use such obstacle passages. Such a thing to cross the Hudson would be the biggest favor for cycling in the USA that could be imagined. (It would be about 1 1/4 miles, not the several miles that make it recreational to one-time and rarer by far in the Bay Area.
Anonymous (but everybody recognises the poster by his style) attempts snark "Bridge replacement -- no doubt with ped-bike access included somewhere" ... but no, sorry to tell you that's not part of this badly-planned and disastrously-executed project. The bridges (one, single track only, for Caltrain, one dedicated UPRR, so nice of us to pay) have just standard not-unreasonable 2 foot wide emergency walkways lying 8.5 feet from the track centreline. Nothing fun or nice for the all-powerful secret bicycle power cabal and its backer George Soros just this one time.
DeleteLong-time informed blog commenter "Reality Check mentions above just a couple things that scratch the surface of what's gone wrong with this project (on a piece of real estate that Caltrain nominally and possibly regrettably owns, but on which it is UPRR's bitch in every way, and on which, as I've noted with backing data above, Caltrain should in no way be operating) but bike/ped access isn't one of them.
And as for neck-breaking off-topic pivot to New Jersey and New York ...
Yes, the land use in San Jose is not very conducive to transit, but it's really not much worse than the rest of Santa Clara County, or San Mateo county for that matter.
DeleteThe real reason ridership at Tamien is low is that it gets less service than Diridon and is not any easier to access from the south. When driving to the station, it is quicker to get to Diridon than to get to Tamien from most places. The only reason to use Tamien is the VTA light rail transfer which is not a popular option. Parking at Tamien used to be easier but at current ridership levels parking at Diridon is no problem. They also converted part of the parking lot at Tamien to TOD.
I think if Caltrain extended electrified service as far as Capitol (or just north of there around CP Lick) and terminated all trains there there could be significantly more demand. There are lots of people who work at Stanford living in South San Jose that would be happy to drive to Capitol and park. The new developments on and around Communications Hill are another source of ridership.
The reality is that Santa Clara County provides over 50% of the funding for Caltrain now. Giving up on San Jose and South County, which together make up 60% of the population is not a good idea if you are relying on future ballot measures to keep operating. Santa Clara voters have been shown to support funding transit, even if they don't actually ride it much. It's important to remember that at 25% farebox recovery you need to run a service the taxpayers like, even if it's not necessarily the service the riders demand.
John jpk122s: "the land use in San Jose is not very conducive to transit, but it's really not much worse than the rest of Santa Clara County, or San Mateo county for that matter."
DeleteDude, in fact it fucking is. Stop making shit up! For the simple reason that there were actual towns clustered around stops on the historical rail line, also geographical things such as Stanford University. South of San Jose, there were orchards in recent living memory (and before that (BUY THIS WONDERFUL BOOK!)) , until rapacious assholes decided within the few several decades that featureless isotropic sprawlburbs were more profitable than the orchards.
Made-up shit like "Foster City" or "Blossom Hill" is hopeless; less-recently made-up shit like "Mayfield" or "San Carlos" or "Santa Clara" ssomewhat less so.
That's how things are. WIshing doesn't make it otherwise. Proclaiming that "There are lots of people who work at Stanford living in South San Jose that would be happy to drive to Capitol and park" doesn't make them take 85 to 280, at one third the time. They (all hypothetical fifteen of them) don't do that today, and there's no way they're going to do that in the future after driving the wrong way to a Caltrain parking lot to take a train to a connecting bus to their office park employment.
Recent "Communications Hill" subdivision isn't a "source of ridership": it's just more of the same, and it, explicitly having nothing to do with urbanity in any way or public transportation in particular, and everything to do with turning low-grade post-agricultural post-extractive land into low-grade low-density residential freeholds.
Nothing I wrote is neck-breaking, nor should any of it be over-challenging to anyone. But "should" is not the same as "is," I suppose. Many are aware of badly done projects in California, and in other states as well, in addition to missed transportation as well as otherwise functional opportunities.
DeleteInteresting sub thread. Sorry for being a bit late to the party.
DeleteI obviously 100% agree re reducing dwell times in SF.
Does the minimum time always apply or is there leeway for shortening it sometimes when there are delays? I.E. would it be possible to negotiate with the union (and/or "OSHA"?) to on a good day have 17 minutes but on average have 15 minutes, and allow even shorter sometimes when there are delays?
I think that there are two great methods to NOT make the union and whatnot angry:
A: Instead of talking about reducing staff and trains in service, talk about increased service without needing more staff and trains in service. I get that it would be worse to have an uneven time table, and if my "napkin style" maths are correct, you'd end up with a train every 13 minutes rather than every 15 minutes and you'd end up with a timetable that repeats every 180 minutes rather than every 60 minutes. But on the other hand a train every 13 minutes is closer to turn-up-and-go than every 15 minutes. With that increase it might be worth arguing for increasing the service to every 10 minutes peak hours.
B: Instead of discussing either keeping the status quo or reducing break times, change things up in having a train set parked all day in SF and have the staff change which train they operate every time they arrive at SF with the current schedule. I.E. unless there are delays let each train just have 17 minutes dwell time while the staff have their 27/40 minute break. That will likely create an incentive among the staff for actually working on the same train all day even with "only" a 17 minute break, rather than having their 27/40 minute break while also having to work on a train that they haven't seen in weeks and possibly where the staff compartment might be dirty/smelly from previous staff not being as clean and/or whatnot.
==============
Re fiscal cliff: Is there a fiscal cliff if Santa Clara county is forced to pay for it's share of what it costs to operate the Caltrain service? I.E. let that county pay for the Gilroy service, and either have the gilroy train shunt Diridon-Tamien in order to get the savings on the EMU operations you mention.
Also re fiscal cliff: Is Caltrain from Tamien and southwards really worse than the rest of VTA?
jpk122s mentions something that I think have important implications: They built some TOD at Tamien.
DeleteNo matter how bad the ridership is at Tamien station, a strong argument against cancelling that service is that if the service is cancelled, it sets a president that a train station with TOD isn't a guarantee that there will be any trains running in the future, and if you move in there (to live there and/or to run a business there) there is always the risk that you get screwed over.
Such president might end up with any future TOD depending on contracts where the transit agency has to pay hefty fines if they don't keep up a certain service level, or otherwise no developer would be interested in building any TOD and/or no one would be interested in moving in (except as a temporary measure if everything is rental rather than owner occupied).
Luke warn take:
Are Caltrain allowed to run the Gilroy trains on the UP tracks Tamien-Diridon, or do they have to use their own track?
If they are allowed to run the Gilroy trains on the UP track to Diridon, then do that and skip Tamien.
If they aren't allowed to do that, install a buffer at the northern end of one of the platform tracks at Tamien and terminate the Gilroy service there.
Either way, install buffers at the southern end of the other track at Tamien, and at the northern end of one track at Diridon, and rent a DMU from eBART or S.M.A.R.T. or whatnot to shuttle between Tamien and Diridon, to reduce the cost of running the service.
Or how about do the USA classic of tricking a manufacturer in Europe that they might get to sell a lot of trains, and have them send over a demonstration 25kV AC + regular 750V DC dual mode tram, and just connect the track to Tamien to the VTA light rail, and shuttle that demonstration unit between a VTA light rail platform at Diridon and Tamien.
Re Miam 0825 21:04, last item: Just for connecting Tamien and Diridon stations, VTA already HAS light rail platforms connected by light rail; there's even a wye at the junction (where the Blue and Green lines meet), but both lines go toward downtown San Jose's fabulous 5 mph pedestrian mall. They could run a shuttle from Curtner via Tamien Diridon much more than hourly, if there was potential for real demand (both stations have turnback tracks). For whatever reason (ignorance on Google's part or terrible frequency on VTA's part), Google directions say it's a 6 minute trip by Caltrain, a 12-minute drive, a 24 to 36 minute transit trip, or a 45-minute walk. Using light rail (tram/light rail preference) they send you up the Blue Line to Convention Center (then a 20-minute walk paralleling the Green Line to Diridon station's front door); or a few more stops into downtown with a bus connection to Diridon. Or a bus connection due west to Fruitdale Station and the Green Line to its platform at the west end of Diridon station. There are also 2 buses that run semi-directly. I fought with the map a few minutes trying to get it to show me a direct rail-rail connection and it wouldn't budge. The Green Line link is not very fast; it has a semi-private right of way and its own traffic signal cycle, but does not appear to have signal priority. It stops several times, zigzags, waits, turns, waits, then speeds into a tunnel under the station with what feels like a 10-15 mph reverse curve before arriving at the Diridon platform. Frequency is also limited to every 15 minutes by the mixed singe and double track route southwest toward Winchester (though if this were a dedicated shuttle, it is all double-track) -Ben in SF
DeleteGood point, for some reason I mixed up Blossom Hill with Tamien re the light rail.
DeleteThe travel time for a new line Diridon and onwards on the southern leg of the blue line would probably be decently fast.
I took a look at google maps and also regrid. It seems like the ROW might even be wide enough to both allow a quad track railway (assuming that any UP trains share tracks with hypotehtical future fast+slow Caltrain + HSR trains and whatnot) and ALSO a double track or at least a partially double tracked light rail. Since VTA seems keen on the express route thing for the blue line, this would kind of fit them.
Also Google seem to own a bunch of land, including a bit of land that trains run on at Diridon :O
A question: is the freight route that runs parallel to the southwestern leg of the green line in use? If so then the crossing might be an issue. I can't see how it would be a bigger issue than any other rail-road level crossing though.
But either way, as you say it would be really easy to run a shuttle Tamien-Diridon using the existing VTA light rail route. That would likely be a good replacement for the existing service, especially if if is more frequent than the existing service.
I was about to write an extensive "hobby study" about pedestrian connectivity along the southern leg of the VTA light rail blue line, but suffice to say there are lots of relatively cheap things that can be done to increase walkability.
DeleteAdd pedestrian connections to both ends of more or less each station. Add pedestrian+bike overpasses across nearby roads with heavy traffic. Use some mechanism to force private road holders to connect them via ped+bike paths to form shortcuts. Use the Caltrain ROW that is already publicly owned to form a ped+bike path on it's eastern side from Curtner to Capitol and Tamien - Almaden Rd. Use the already publicly owned Tamien Park to connect both Tamien stations to Pepitone Ave and Mclellan Ave.
Use eminent domain to put in easements for pedestians+bikes to use commercial parking lots as shortcuts, like for example to connect Captiol to Heppner Ln.
Actually pave obvious desire paths. Example Glenbury Way to where Santa Teresa Blvd connects to SR85.
Force the San Jose Unified School District to allow pedestrians+bikes to take a shortcut across it's parking lot and driveway to connect Guandabert Ln to the Ohlone-Chynoweth VTA light rail station.
Build a bike+ped path all the way between Ohlone-Chynoweth and Blossom Hill VTA light rail stations, connecting to the cul-de-sacs and whatnot along the route.
Also the end at Santa Teresa is really lackluster. In hindsight it might had been better if the route would have followed Blossom Hill Rd to the Caltrain Blossom Hill station, perhaps?
I don't know what can be done to make private land owners improve things. I doubt that it would ever be possible but how about taxing the difference between as-the-crow-flies distance and walking distance to adjacent transit stations? I think that would really bite, it might even bite so much that properties near a transit station join up and buy one of the properties that would be needed to put in an easement for a bike+ped path across someones yard.
Clem mentions: "In your "focus on the core" scenario with just 12 trains, with nearly half the EMU fleet (eventually 23) sitting idle and depreciating, ..."
ReplyDeleteNot my circus, not my clowns.
I'm not the one who skimmed off the staff and consultant and middleman overhead (at least 60%, all told!) through dumb (or super crafty, depending on which side of the ledger you live) procurement. I just know that somebody made bank. MOAR EARMARKS. TRANSIT NEEDS MOAR EARMARKS!
I'm not the clown staffing trains designed for one person operation with an engineer, a conductor, a navigator, a busboy, a fire control officer, a purser, a fireman, a flight engineer, and a cabin boy.
I'm not the one who outright refused to contemplate level boarding, let alone plan for, let alone implement, let alone have fully implemented 15 years ago -- as was both obvious and readily possible.
I'm not the one who prioritizes only those capital projects that have zero or negative effect on service efficiency, and even then can't build a single stupid one-road one-rail-bridge motorist-benefiting grade separation for less than a half billion dollars.
The fact is that nobody can or will afford to pay Caltrain to operate 20-ish trains at peak or at any time of day any decade soon. Even with single person train operation (supplemented of course by economically justified levels of spot fare inspection) nobody's going to pay for this. BART's in a similar mess, even though it's less dismally operated (if not constructed.) They money is not there, and it's not coming.
FISCAL CLIFF!
Everybody's just going to pretend that nobody could have forseen any operating cost problem at Caltrain, ever, that it's all the fault of COVID, or or 9/11, or of the 1990s dot-com crash, and, well, we're just going to have to park a lot of trains a lot of the time, sorry about that, but hey, anyhoo ... look over there Return to Office!, Google Transit Village!, Squirrel!, something.
Caltrain's shit consultants and shit staff specified and purchased too many too-large too-low-capacity-per-dollar trains, and that's just how things are.
Anyway, California HSR might be in the market some time maybe in the next 20 years or so for some small number of trains, any trains, to run empty on its nowhere-to-nowhere Central Valley "initial segment". (Maybe, but more likely they'll never complete anything, no matter how useless.)
It's a pity that Caltrain has too many wrong-sized trains, and that Caltrain incinerated hundreds of millions of our tax dollars, but we can't do anything about that it now. (And, hey, look at me, trying so so hard to tone down the "up against the wall" hyperbole.)
What we can try to do now is not actively plan for bonfires of objectively scarce operating cash on bad services, in an environment where expensive stupid service is only going to result in a spiral of cutbacks to even worse service -- worse plus even more expensive per passenger. DOOM LOOP!
Being efficient in real world economics sometimes means idling excess capital equipment. Caltrain's not special.
FISCAL CLIFF!
Re the large amount of almost comically oversized trains Caltrain bought:
DeleteIf there were any efficiency enthusiasts at state level then a great solution would be to
A) solve the Cali HSR platform height issue by buying new single level ADA compliant cars that match the Cali HSR floor level
B) Buy a lot of various other cars and the work to reconfigure the existing trains to say about 60 half length trains
C) Electrify and double track the Metrolink network in the LA area, and also the route to San Diego, and transfer a lot of the trains to Metrolink.
I know that this blog is about Caltrain and whatnot, but it's worth noting that the transit agencies in the Metrolink area owns all infrastructure of two of the current lines, own the inner half of a line that is shared with UP, owns the tracks south from the Y fork in Orange County all the way to San Diego (LOSSAN/Surfliner) and the tracks northwest and northeast of this are owned by BNSF and not UP. Thus almost all of the network can be electrified without having to negotiate with UP.
And yes, I know that Metrolink owns a bunch of trains currently, but the service levels are abysmal with about as many trains per day as the most frequent metro systems run per hours (more or less). and if if we would aim for a 10-15 min frequency rather than almost a train an hour, the current fleet is tiny compared to what is required, and thus there isn't much sunk cost in replacing the current fleet with high floor platforms/trains. And in particular there is one Metrolink route that fully runs on an UP line and that would likely need to run the existing low floor trains for the foreseeable future, and for that sake there surely must be a used market for those trains elsewhere.
Hoisting reply to this note from Clem into a new reply thread: "... I wonder if there would be a positive change in preemptively going from a 30-minute base takt to 20 minutes?"
ReplyDeleteI think there might be! But precisely not in the way you're suggesting, which is 6tph service (20 minute headway local and limited-stops interleaved, 10 minute "show and go" headways south of Redwood City) requiring a fleet of 19 trains (using the Taktulator model of today's Caltrain long dwells and turnbacks.) To me frankly, that's an unsustainable and unsupportable amount of service on a fairly undeveloped corridor.
I've used the phrase "don't overthink it!" on this subject a number of times here over the last year: maybe the right thing right now, given the realised performance of the new Caltrain fleet, is just to stop hassling with the diminishing returns of timetabling limited-stop trains on what Caltrain has decided is two-track-only infrastructure, probably forever (because Caltrain can no long execute construction of anything at all, on top of never being able to design anything of any positive value), and just go all-in BART, all-stops, all-hours, all-day, every day. Just BART it!
Let's break out the diminishing returns of possibly overthinking it in steps:
First, let's start with the optimised version (2tph local, 2tph "Silicon Valley Limited", no1tph "Baby Bullet", no Tamien) of today's Caltrain timetable:
Let's name this base scenario 2/2 Cahill (2tph local, 2tph limited, no Tamien) Taktulator says this requires a fleet of 12 trains and scores 113 (per the crusty and over-simple but stable since 2011 quality metric algorithm.)
[Let's see if I can post something without stupid HTML errors and a hundred typos...]
I looked in the spam trap for the sequel, but nothing there... are you working towards 20-minute all-stops BART-like peninsula service? It's not a crazy omelet to make with these eggs.
DeleteYeah, blogspot.com (Google abandonware) rejected everything "Failed to publish comment. Please try again later."'
DeleteI did, about 50 times.
https://support.google.com/blogger/thread/253139831/failed-to-publish-comment-please-try-again-later?hl=en
(hah hah hah HAH HAH HAH "google" and "support" in the same URL!)
To start us on our voyage of incremental discovery, here we have
Delete"2/0 Cahill" (30 minute headway locals, no limiteds = today's Caltrain base headway): Fleet 6 trains(!); Score 99
Taktulator algorithm believes just 2tph electric local is as good as all bad old insane 2004-2019 6tph diesel timetables. I won't argue much with that.
and
"4/0 Cahill" (15 minute locals = pre-Covid BART base headway): Fleet 12 trains; Score 114
Both have 84% in-revenue (vs turnback dead time) fleet utilisation, which is high (by design; that's why they pay me the big bucks for this stuff) and possibly unrealistic (but shouldn't be!) per present-day Caltrain slovenliness as I've dialled down the SJ turnback to 13 minutes from standard 15 to make it juusssst work. Could evenly do 14 in SF and 14 in SJ instead.
[And lately janky Google abandonware blogspot has come over all "Failed to publish comment. Please try again later.", at random, for hours at a time. for no reason.]
Note above that fleet size for "4/0 Cahill" and "2/2 Cahill" are both 12 trains and the timetable metric score is nearly identical: the algorithm (which has its faults, but let's leave well enough alone) -- frequency is indeed dominant, the scores picked up from more service to lesser stations in San Mateo County balance out marginal (8 minute) limited-stop trip time savings on longer SF—Santa Clara County trips.
DeleteI don't know if this is a perfect model of reality, but I'm not unwilling to countenance it. The new train performance is such that all-local service doesn't feel as much of a horror drag as it used to in the bad old days. Maybe ... "don't overthink it!"?
Maybe limited stop service is something we might need to grow into, sometime not , but maybe not right 100 years away, but perhaps not right this moment?
Honestly the best thing Caltrain could do to reduce the perceived overhead of station stops may not be to run limited stop trains, but to QUIETEN DOWN THE HECTORING ANNOUNCEMENT VOLUME, kill all the bullshit repetitive safety and whatever spiels, and GAG THE FUCK OUT OF THE FUCKING DOOR OPENING AND CLOSING ALARMS (BART isn't great but it is not this bad, so this can't be for regulatory reasons) — it's 10x as nice to ride upstairs as anywhere near the nanny-state beeping of the downstairs doors.
The trains are pretty fast, and while the ride quality is depressingly sad compared to a KISS running Switzerland — US track maintenance standards are a freight joke straight out of the 19th century, and the US turnouts chooses to install are simply a ride-quality nightmare — it's still night and day compared to what went before, and the train interior noise level is ok ASIDE FROM THE BLARING PA AND BLEEPING DOORS.
So, yeah, we can dick around with whether South San Francisco should or shouldn't be part of a limited-stops service pattern, but maybe it might be better not to overthink it, and simply not to think about limited-stops service patterns at all, just right now.
[Could post that if I borrow my next door neighbour's wifi, changing nothing else. Idiot Google countermeasures]
DeleteContinuing our not-overthinking-it examination of too many simple service plan choices, let's go all in with today's BART-tastic 20-minute base headway.
DeleteFirst, here's "3/0 Cahill"— a Caltrain every 20 minutes, every service hour of every day, kind of nice actually — all-stops local service (BART-tastic!), no limited stop services. This requires a mere 10 trains/crews (compare to 15 minute headway's ("2/2" or "4/0") fleet of 12, which was pushing it utilization-wise) and scores 107 (compare 114.) Not too shabby!
I did make some effort to in this and what follows to optimize Caltrain's northbound and southbound timings around BART's at Millbrae (bearing in mind Millbrae's inexcusably hostile and slow up-and-over passenger route, and poor Caltrain timekeeping) rather than around possibly more useful "takt symmetry" stops that could be made elsewhere (Redwood, Palo Alto, Mountain View, Sunnyvale, etc ...) but such detail doesn't much affect simple scores we're looking at here.
And personally a base 3tph — miss a train at night and the wait is 19 minutes, not half an hour — feels kind of acceptable, not a humiliating affront; something one can put up with when things run over, something that won't force you to think you're an idiot for not just driving screw it.
Moreover, 4tph base (at 8pm or at 6am) just doe not seem anything Caltrain can possibly afford to offer, even without multiple sub-assistant-co-intern-sub-altern-conductors billowing out the cost of actually running each train, any time in the next decades. But 3tph? Maybe not inconceivable. Maybe!
Anyway, to continue, what happens if we think that a big ol' 7-car 183m-long double-decker train coming every 20 minutes all day just isn't enough, or we think we need limited-stops services for some reason?
....
@Richard: FWIW, I’ve been able to get an inexplicably rejected comment accepted by switching to a different web browser (e.g. from Chrome to Safari or Duck Duck Go). Also, I think I once recall switching from kerned quote marks to straight ones in my HTML href URL tags seems to have allowed a rejected comment to go through. Weird.
DeleteAnyway, to try to wrap this up, here's what happens if we go all BART-tastic, base 20 minute headway.
DeleteIn this tried to sync up same-direction BART and Caltrain at Millbrae, within the confines that Millbrae station being a total pile of hostile shit designed to make connections as bad a physically possible southbound, and that Caltrain timekeeping remains so poor that one ought not to risk an exactly timed arrival even in the nominally "cross platform" northbound direction; I went with 6(!) minute transfers in both directions, but it doesn't affect much to jiggle this. I've used the same Caltrain dwell times that so nicely model (thanks, Clem in 2011!) today's real-world timetable.
3/0 -Tamien +Hayward Park
First thing to note is (aide from "THAT'S A NICE LEVEL (and corridor-appropriate) OF SERVICE ALL DAY EVERY DAY") is: fleet size 10 (!!!).
Utilization is (76.8% good but not spectacular, with 30 and 15 minute turnbacks), score 107.
Compare that to today's Caltrain peak (2/1+1) without Tamien: Fleet 13 trains; Score 109.
We're nearly at parity per Taktulator score (I know, not the real world, but we're all living in the worst timeline dystopian simulation anyway!) by not overthinking it, and meanwhile we've saved 28% of our peak crews and trains just by the magic power of doing dumb stuff acceptably well.
One is very much lead to speculate ... with 20 minute all-day service, and with great honking inflexible 183m 662-seat 1668-capacity trains, is there really a need for extra peak hour service at all? (On-board bike capacity likely to max out, but that aside.) And if there is to be extra peak service, does running limited-stops trains really buy much at all over just slotting in an extra fast-enough all-stops train or two or three, on a ten minute headway?
...
[Ugh, as always, I failed to adequately proof my HTML tags -- enjoy the specious italics, y'all]
DeleteLet's add one all-stops local per hour at peaks on top of that 3tph base (quasi-parity with today's Caltrain "2/1+1" 4tph peak servivce), slotting it right into the middle of one of the three 20 minute service gaps per hour.
"3+1/0 -Tamien +Hayward Park" (Fleet 14 trains; Score 110)
Compare to today's Caltrain (sans Tamien)'s fleet 13 score 109; or to the above "4/0" 15 minute all-local scenario fleet 12 trains; Score 114. Taktulator scoring show signs of preferring closer-headways over dimishing-returns limited-stop services. And as Clem regularly reminds us, Caltrain's average passenger journey is around 25 miles, not 50. Also I've fixed these schedules around BART at Millbrae rather than attempting to optimize for fleet size.
...
I should have been less stupid with the "+1"'s northbound vs southbound deparure times immediately above: moving northbound 20 minutes earlier is obvious to save a train, so:
Delete"3+1/0 -Tamien +Hayward Park" (Fleet 13 trains; Score 110)
The basic deal is that addiing an incremental 1tph at 20 minute headway on Caltrain is going to cost you 3 trains+crews:
3tph = 10 trains (+1 becaue we're tied to BART; 9tph just possible otherwise with 13 minute turnbacks)
4tph = 13 trains
5tph= 16 trains
6tph = 19 trains
To continue to experiment towards the bitter end of over-thinking, let's run 1tph "Silicon Valley Express" instead of that bonus 1tph local.
Delete"3/1 -Tamien +Hayward Park" (Fleet 14 trains; Score 113)
There's the problem that a limited stops train running a nice even 10 minutes separation in Santa Clara County catches up to the preceding local around Bayshore—22nd, and since for now we're working with constraints from this, the worst possible dystopian timeline, in which Caltrain believes it is impossible to overtake or short-turn trains.
Choices are to ditch one local stop north of Redwood City (the answer has been obvious for 30 years: Hayward Park has to go!; but sure nuke depressing Bayshore instead if you prefer to stick it to Satanic San Francisco), or to not run on even headways. To be apples to apples, I'm keeping Hayward Park and running the limited 2 minutes later = 12/8 minutes between trains in Santa Clara County.
Are real people really going to time their lives around a 1tph train that saves 9 minutes of trip time, or just show up and take whatever train comes next? The Taktulator thinks so, but I have serious doubts at 3+tph. (The Taktulator also assigns ridiculously high non-real-world scores to Broadway and Hayward Park, but it is what it is and I'm not tampering with it.)
...
We can continue this game, but I'm going to stop linking to Taktulor URLs because of blogspot's retro comment length limit. If you look at the above scenario links, you'll find they include a pile of hidden "(Placeholder)" trains, which one can turn on an off in a somewhat-straightforward way to come up with all the combinations:
Delete3tph:
"3/0 -Tamien +Hayward Park: (Fleet 10 trains; Score 107)
4tph:
"3+1/0 -Tamien +Hayward Park" (Fleet 13 trains; Score 110)
"3/1 -Tamien +Hayward Park (Fleet 13 trains; Score 113)
recall at 15 minute headway all-local 4tph don't-overthink-it "4/0 -Tamien +Hayward" (Fleet 12 trains; Score 114)
while my optimized version of today's 4tph schedule is "2/2 -Tamien +Hayward Park" (Fleet 12 trains; Score 113)
5tph:
"3+2/0 -Tamien +Hayward Park" (Fleet 16 trains; Score 112)
"3+1/1 -Tamien +Hayward Park" (Fleet 16 trains; Score 115)
"3/2 -Tamien +Hayward Park" (Fleet 16 trains; Score 115)
6tph:
"6/0 -Tamien -Hayward Park" ie 6tph 10 minute show-and-go all-local just like BART (Fleet 19 trains; Score 118)
"3+2/1 -Tamien +Hayward Park" (Fleet 19 trains; Score 119)
"3+1/2 -Tamien +Hayward Park" (Fleet 19 trains; Score 119)
"3/3 -Tamien +Hayward Park" (Fleet 19 trains; Score 119)
My take: 3tph base all day every day (10 trains) NOICE!
Taktulator scores don't suggest big returns from adding peak trains on top of this; each additional hourly peak run costs THREE trains and crews (+30% over base service each time), while the Taktularor scores only inch up slowly.
I can't see demand for more than 4tph south of Redwood City any decade soon (6 or 8tph Redwood City—SF with (Belmont—)San Carlos—Redwood City quadruplication and turnback, which Caltrain scum are actively torpedoing as we speak.)
Maybe Caltrain could justify 4tph "3/1" scenario with 1tph express (30% peak train+crew premium) today. This has the same peak fleet/crew requirements as today with same Taktulator score.
Meanwhile base every hour every day would go from today's saddening 30 minute headway 7 trains score 99 (can be optimized to 6 trains) up to BART-tastic 20 minute base headway 10 trains score 107. Do it!
Note also that at 6tph "show and go" 10 minute headway level, there is zero score premium awarded for running any limited-stops trains.
@Richard:
DeleteA bit of civil obedience might be worth considering?
A) show up at Caltrain related meetings and use a megaphone to state that the on board loudspeakers are too loud, and the announcements are too comprehensive. If no-one seems to want to listen, just inform everyone where the emergency exit from the meeting room are every now and then.
B) get pieces of the sound deadening material used in cars, and use gaffer tape to tape it over the speakers in the trains, to quiet them down.
Tamien used to support a large number of rush hour riders. Several years ago the last weekday morning baby bullet (#329) was mostly full (and sometimes SRO) when it departed Tamien. Today's Tamien is a much more lonely place. Previously both main parking lots were full by mid-morning. Now the parking lot at 9am is barely 10% full.
ReplyDeleteIt is clear that something has changed. The most obvious explanation is the pandemic and WFH, but that alone can't account for the dramatic loss of ridership.
It's 2 things IMO:
Delete1 - WFH and less daily commuters than before.
2 - No more park and ride. Tamien was a popular park and ride destination right off of 85 and Almaden Expressway, so lots of folks would drive in from South San Jose, and park at Tamien. I would argue that Clem's ideas about extending electrification south with a ton of TOD at stations and redesigns would induce a similar amount of ridership if it's convenient to access by car from south San Jose, Almaden, etc
Tamien ridership (boardings) 1995-2014:
Delete1995 382
1996 468
1997 492
1998 531
1999 526
2000 676
2001 821
2002 634
2003 520
2004 480
2005 343
2006 446
2007 532
2008 610
2009 613
2010 544
2011 577
2012 653
2013 807
2014 970
2015 1102
2016 1283
2017 1264
2018 1286
2019 1422
[above from annual February ridership counts https://www.caltrain.com/about-caltrain/statistics-reports/ridership/annual-count ]
2024 1009.0
2025 1228.5
[above from guesstimates https://www.caltrain.com/about-caltrain/statistics-reports/ridership/fare-media-based ]
Tamien, today, is nearly as busy as it has ever been. By which I mean STICK A FORK IN IT, IT IS HOPELESS BASKET CASE, INCINERATING DUMPSTERS FULL OF CASH. (Just like everything else rail-related at VTA.)
Weekday boarding of Tamien (2025) is 200~250, even lower than 1995. Traffic of 87, 280 and 101 North are not yet slower than pre-Pandemic.
DeleteArgh Anonymous 03 June, 2025 16:36 is correct; I made a stupid arithmetical mistake with the 2004 and 2005 numbers -- I took Caltrain total monthly guesstimate data, dividided by the number of weekdays in that month in that year (charitably assigned "all"the weekend "ridershiip" to weekdays) but then inexplicably stupidly mutiplied by 5 to get WEEKLY ridership, not average WEEKDAY ridership, and then didn't do basic eyeball validation. Idiot!
DeleteI apologize.
Corrected Tamien:
1993 332
1994 359
1995 382
1996 468
1997 492
1998 531
1999 526
2000 676
2001 821
2002 634
2003 520
2004 480
2005 343
2006 446
2007 532
2008 610
2009 613
2010 544
2011 577
2012 653
2013 807
2014 970
2015 1102
2016 1283
2017 1264
2018 1286
2019 1422
2024 202
2025 246
As for why South SJ ridership is, always has been, and always will be a dog, it's because Santa Clara Valley Transportation Authority EXPLICTLY WANTS EVERYBODY TO DRIVE. (Same story with San Mateo County Transportation Authority, of course.)
See for example page 47 of Caltrain's 2019 ridership report>.
Hey, let's widen 101!
What about widening 101 again? For SAFETY.
Seems like we might need some new interchanges on 101. FOR CONGESTION RELIEF.
Wow, 101 isn't wide enough. We'll something something greenhouse gases if TRAFFIC FLOWS MORE FREELY.
Need to widen 101 with carpool lanes. Because carpool lanes are tight!
Breakdown lanes! For Great SAFETY!
Hey [hot dog guy meme] can anybody find the person responsible for zero transit ridership?
As a bonus, in penance, here's screen-scraped corresponding total Capitol—Gilroy average weekday "ridership":
1996 514
1997 772
1998 1106
1999 1074
2000 1311
2001 1556
2002 1143
2003 988
2004 668
2005 635
2006 470
2007 441
2008 450
2009 393
2010 323
2011 349
2012 365
2013 421
2014 462
2015 559
2016 629
2017 590
2018 800
2019 752
2024 284
2025 332
Anonymous @ 3 June, 2025 16:36 identified the key factor: the freeways are still in a pandemic era free flowing state. Once the 87, 280, and 101 start jamming up to 2019's rush hour levels, Tamien ridership will likely rebound.
DeleteSo long as commuters have access to generously capacious freeways they will usually drive. But when freeways under-perform, Caltrain starts picking up riders. In 2019 the rush hour baby bullet trip from Tamien to Mt. View was something like 16 minutes versus a 30-40 minute drive. Now there's hardly a difference.
Tamien Station Electric Train Service Temporarily Suspended
ReplyDeleteFree Replacement Bus Service to San Jose Diridon Beginning June 16, 2025
Beginning Monday, June 16th, 2025, and continuing for approximately eight months, electric train service to and from Tamien Station will be temporarily suspended seven days a week to accommodate construction for the Guadalupe Bridge Replacement Project.
A normal PUBLIC agency in a normal country would advertize that XX months (typically 2 or fewer, not 8) and expenditure of XX million (typically less than 10, not THE SKY IS THE LIMIT) of CURRENCY will result n XX (typically 4 or more) trains per hour carrying more passengers on the crowded (NOT AN EMPTY PARKING LOT NEXT TO A FREEWAY) line.
DeleteBut Caltrain? VTA?
Spend anything.
Deliver nothing.
The latest (May 28) Guadalupe River Bridge Replacement Project Update slide deck says (among many other interesting things) on page 21 that there are currently only ~210 daily weekday and ~60 daily weekend Tamien riders.
DeleteOriginal budget: $63,698,593
New budget: $171,389,598
So only a $107,691,005 or 169% increase or 2.7x the original!
@Clem Hot off the presses, tomorrows VTA board meeting includes a presentation on negotiations with Caltrain regarding paying for Gilroy service. Here is a link to the PDF presentation. There is a very interesting table on page 6 which reveals:
ReplyDeleteThe total operating cost of Gilroy service is $15M with revenue of $2M for a net operating cost of $13M that Caltrain wants VTA to pay for in full.
VTA is saying if they have to pay for it alone they will want more control over service levels. They are also taking the position that the same rule should apply for future "extensions to the north" by which I assume they mean that SFMTA should be responsible for the net operating costs of DTX if that ever happens. Spicy!
In other news, the Caltrain Board will consider a proposal from staff to completely remove any provision for future capacity increases beyond the "Moderate Growth" plan in their future planning and ROW use. This would conveniently allow a property developer to build on the Belmont Station parking lots, blocking any future 4-track overtake in that location. It would also conveniently allow Prologis to redevelop 4th+King without providing enough replacement train storage to allow for future service levels beyond the Moderate Growth level. This all seems incredibly shortsighted.
ReplyDeleteSeems like they want to repeat the disaster of San Carlos station development at Belmont station too. Permanently reducing the potential capacity of the corridor for some short term development opportunities.
I hope there is some reconsideration of this before the August Board meeting when this proposal is scheduled to be approved. Given the revisions to the PDF slides perhaps the pushback has already started. One can hope.
This is just mind-boggling Trump-level total corruption bad.
DeleteThey're going to do it, aren't they?
There's no hope, is there? There never was.
Shit.
Yep, Caltrain is like other transit agencies, and can't resist the real estate "plays" at station sites, willing to debase itself to enable more of it while it too exploits the ignorant and emotional supporters as Useful Idiots.
DeleteDid Cali HSR force Caltrain to sign any agreements with Cali HSR when Cali HSR paid for parts of the electrification? If so can those agreements be used to repeal any selling off of valuable ROW for future quad tracking?
Delete@MiaM,
DeleteThere have been multiple agreements between the two parties for various reasons, and while I don't know about any arm-twisting or string-tugging with those funds, there was at least one agreement related to the electrification funding. I say "at least one" because documents can disappear, as I saw with Cal HSR some years ago, that weren't due necessarily to a computer system change at that time. The agreement does express intent to convey a property interest to Cal HSR in the corridor, but that doesn't preclude selling some of the right-of-way necessarily, in particular when the same agreement refers to "blended" operation and principally a two-track system.
https://www.caltrain.com/media/811/
Any MOUs between Caltrain and CHSRA are online. You can search for them and read them. It’s not like they’re buried in bound minutes on a shelf in a library.
DeleteMichael,
DeleteYou're writing this as if everybody is a policy wonk who knows the *PRECISE* search terms to turn up this stuff, and on whose web site.
And as if some "GRAPHIC DESIGN IS MY PASSION" assholes didn't redo the web sites of all of the agencies every couple years, breaking all the URLs and breaking all the search terms.
And as if any of the agencies involved WANTED anything to be discoverable.
And as if any of the agreements were signed were worth any of the electons with which you eventually, after hours of wasted effort, finally discovered some of them, or at least those discoverable today. Tomorrow: [shrug emoji IDK IDGAF WTF]?
@Anon:
DeleteThanks, that document is at least somewhat interesting.
Here's some details that I found interesting:
It refers a lot to a 2012 Caltrain/CHSRA Blended Operations Analysis document.
Page 6 Article I A: It seems like the contract says that whatever isn't already part of the contract and whatnot will have to be paid for by CAHSR.
Page 9 Article III A: This is a juicy thing! It says that "if PCJPB obtains Intercity Passenger rights in the Corridor currently held by Union Pacific Railroad" they have to give that track capacity to Cali HSR for free. This kind of says that Caltrain will never do something about this route unless a new agreement is made. Also all the capital letters makes it look like a native German speaker wrote that sentence.
Page 10 Article III B kind of just says that both parties have to solve any disagreement upon things. Super vague.
Page 11 Article IV is interesting. It says that the parties have to negotiate additional agreements about a list of things, including "passenger boarding, shared platforms and platform height". (line 9-10 for anyone too lazy to even read the full paragraph). It also says that if the parties can't agree upon things, the last resort after escalating within each organization, is that the Cali Secretary of Transportation nominate one or more mediators.
The main question here is any future agreements were made re "passenger boarding, shared platforms and platform height"? If so, what does it/they say?
If not it seems like Caltrain made a mistake in ordering their current trains without such a document in place.
I doubt that Cali HSR would drag Caltrain to some legal process to force Caltrain to switch to trains that are compatible with Cali HSR platform height, but still.
@Michael: An issue is how to know what documents to search for?
Meanwhile, the state high-speed rail project got skewered by the Trump administration, hardly the first to skewer it, comparing it unfavorably to Caltrain when it comes to electrification.
ReplyDeletehttps://www.transportation.gov/sites/dot.gov/files/2025-06/FRA%20Letter%20%20Enclosure%20to%20Mr.%20Ian%20Choudri%206.4.25.pdf
Shock news!
ReplyDeleteVTA eyes terminating costly SJ BART extension tunnel contract
Agency struggles to find over $1.2 billion in project cost savings
[…]
KST is expected to continue its pre-construction work, which includes building of the massive tunnel boring machine. In 2023, VTA purchased the machine, which has been likened to a mechanical earthworm, for $76 million. It’s being custom made in Germany and will be tested there before it’s disassembled and shipped in parts to Santa Clara County.
As part of the cost savings review, the agency is looking at designing a tunnel with a smaller diameter, but Maguire said that the decision to end the contract with KST is independent of that discussion. If they do change the tunnel size, he doesn’t expect the tunnel boring machine will go to waste.
“There’s a wide range of options that we’re exploring right now,” he said.
Someone watched Alan Fishers video...
DeleteThere's also this news, which I suspect many already know but was interesting to post here from the following source because it includes everyone's wonderful master "transportation" bureaucrat himself, Steve Heminger. (Maybe he shares the view of others, too, in wanting "transportation" persons and dollars spent instead on bankrolling housing projects and actual housing by other agencies.)
ReplyDeletehttps://padailypost.com/2025/06/06/caltrain-will-reduce-its-plan-to-increase-the-number-of-trains-it-runs-each-hour/
Here are the staff slides on the proposed “long range service vision” reduction agenda item. It seems Caltrain/staff is under pressure from cities struggling to afford grade seps and new bike/ped underpasses to stop insisting projects accommodate potential future quad tracking for up to 12 Caltrains + 4 HSR per hour per direction.
DeleteThis is simply mind-boggling corrupt and evil. "Stupid" and "short-sighted" if you prefer that sort of mealy-mouthed "mistakes were made" "who could have predicted" bullshit, but real individuals are actively advocating real and *irreparable* harm to the public interest, and there' no other way to describe their actions other than what they are: corrupt, and evil.
DeleteIt's like these sub-humans looked at how utterly *fucked* Caltrain is, predictably, by San Carlos "TOD", and said "hold my beer ... we can put the final nail in the coffin of Caltrain corridor service, FOREVER. Not only do we promise not to pland for or operate good service, but we are here to guarantee that NOBODY ELSE CAN, EVER." Utterly despicable.
Seriously, if you work for a "public" (hah hah hah) agency or for a "professional" (as if) consultong agency and you're directed to produce corrupt and evil lies, you should QUIT. Or kill yourself if you somehow find yourself on the wrong side of basic morals and ability to think or plan. The world doesn't need you to do this. And you're not going to be rewarded, either, suckers, the real payola isn't for the likes of you.
The staff presentation and board discussion begin at the 2:44:55 mark of the June 5 Caltrain board meeting video.
DeleteThe rationale seems to be there will never really need to be quad tracking beyond the already-envisioned short segments (e.g. new RWC station) for the moderate growth service vision because of WFH and because increased Caltrain mode share, HSR, population growth, and more trains from Dumbarton Rail & Link21 currently all seem permanently stalled or unrealistic “fictions” … and because preserving ROW has costly and/or inconvenient implications for both staff and “partner agencies” seeking to build grade seps and TODs like SF Rail Yards.
Totally at random, I happened upon this recent YouTube video of surplus Caltrain gallery cars en route from Schellville to Lima, via Port of Stockton. https://www.youtube.com/watch?v=rbXN9oC2eUo. -Ben in SF
ReplyDeleteHere’s another video entitled Off to Peru! …
DeleteRe-activating 22-mile Santa Cruz - Pajaro line would cost $4+ billion, according to SCCRTC:
ReplyDeletehttps://www.sccrtc.org/wp-content/uploads/2025/06/Att1_ZEPRT-Executive-Summary-DRAFT.pdf
$1 billion to overhaul/replace 33 bridges
$1 billion contingency
Operating cost for 30-minute headways would be 34-41 million (2-man operation diesel/battery/hydrogen). $17-21 million for 60-minute headway.
Although not likely, let's hope that they go for electrification and more or less show that relatively short mainline railway style EMUs are a thing!
DeleteIn correspondence for the 6/5 Caltrain board meeting, Gilroy city council member Hilton called for a 2-year pause to Caltrain South County Connector service. He cites the much higher bus ridership: "As of February 2025 the VTA’s Frequent 68 (San Jose to Gilroy every day) runs every 15 minutes averages 4,219 daily boardings and the Rapid 568 (San Jose to Gilroy Monday-Friday) runs every 30 minutes averages 932 daily boardings," in contrast with Caltrain's "Gilroy 100 riders/day, San Martin 22 riders/day, Morgan Hill 120 riders/day, Blossom Hill 54 riders/day, Capitol 42 riders/day."
ReplyDeletehttps://www.caltrain.com/media/35438/download
That's an astonishing statistic and argues strongly for VTA spending whatever Gilroy money they have on more buses, not Caltrain. By these numbers, there are 15 times more riders on SJ-Gilroy public busses (not counting private employer shuttles) than on trains!
DeleteKeeping Gilroy rail service makes a travesty of any "fiscal cliff"
I don't know for how many hours per day those buses run, and I can't be arsed to check, but if we guesstimate that they run 16h/day, the 4219 daily boardings for route 68 is 65 boardings per round trip, or 33 per direction per trip.
DeleteFor the Rapid 568 the 932 daily boardings corresponds to 29 boardings per round trip, or 15 per direction per trip.
Meanwhile the sum for the Caltrain ridership is 338, which divided by three northbound trains is 113 passengers per train.
Please correct me if anyone feel inclined to actually check the time table for how many trips the buses runs. (I just assumed 64 round trips for the 68, and 32 for the Rapid 568).
Since no data is mentioned for how many passengers are exiting the trains south of Tamien, I just assume that everyone who boards the train are going northbound. There might be a few people who travel between stations south of Tamien, but that is likely a negligible amount.
I'm in no way saying that the Caltrain ridership is great in any shape or form, but especially for the Rapid 568 bus route the ridership doesn't look great either. The only real differences are that buses are cheaper to run but also the give a cheaper feel to ride, i.e. generally have a worse ride comfort.
VTA route 68 and route Rapid 568 schedules.
DeleteDuring 2-year Pause, VTA is still responsible for cost of diesel fleet maintenance. Gilroy city council member Hilton seems not calculated this cost.
Delete"During 2-year Pause, VTA is still responsible for cost of diesel fleet maintenance ..."
DeleteIt's mothballed non-revenue equipment, of no relevance to Caltrain or to Caltrain service.
Somebody, anybody, who isn't Caltrain should be maintaining or disposing of non-Caltrain-relevant stuff.
As for "... seems not calculated this cost" -- that's minuscule given the piles of cash set on fire when one actually runs these trains, one round trip per day per train, while paying, for each of these trains, for under two revenue hours per train per day, a train captain, a train fireman, an ensign, a cabin boy, a navigator, and a chief petty officer to pilot each train, very nearly empty, from and to Gilroy.
Orders of magnitude, people. Orders of magnitude.
Regardless, VTA, as it happens, does indeed have tens of billions of dollars of Other People's Cash in a dumpster, one that is on fire and polluting all of California even as we're speaking. If VTA wishes to power some diesel trains to and from Gilroy using the toxic fumes coming off this fire, that's up to VTA. (Sorry, Palo Alto, the actual economic engine of Santa Clara County. You're fucked. Just like Calfornia is by Trumpland. Bad choice of county line! Secede!)
If VTA wants to buy the diesel locomotives and passenger cars and find a place to store them and secure them, for one day or two years or for ten years, fine. It's not as if VTA doesn't make spectacular "investments" in machinery it has no idea how to use or where to use or have any reason at all for owning..
If VTA doesn't want to buy it and store this stuff somewhere, then find somebody else who does. Sadly there's not a great market for this stuff in Trumpland in 2025, and nobody outside North America would even consider this fifty-year-obsolete junk, but hey, [shrug emoji], whatevs. Just find some place somewhere else to shove it where it isn't gumming up Caltrain service and screwing Caltrain's budgets.
Prosecutors release photos of the secret Burlingame Caltrain station apartment that was built by a former Caltrain deputy director with public funds.
ReplyDeleteJoseph Vincent Navarro was sentenced Wednesday to 120 days in county jail and two years probation for embezzling nearly $40,000 in public funds to turn his office into an apartment, where he then lived for several years. The apartment was complete with a gym, a shower and furniture, and Navarro temporarily moved his ex-girlfriend into the residence with him while she was recovering from surgery.
Navarro was convicted of one felony count of misuse of public funds in a jury trial in April, where his defense attorney painted him as a “dedicated railroad man” who worked 80 hour weeks. The prosecution argued that Navarro’s use of the funds was an abuse of power and authority.
Now that they're built, why isn't Caltrain renting out these apartments??
DeleteAG: hah hah, Caltrain staff and consultants are very the definition of "rent seekers"!
DeleteIn the sense of "Oh oh oh no our hands are tied by the "standards" that we pulled directly out of own own assholes! [See electrification, grade crossings, platforms, etc.] So so sorry the project you wanted doubled in cost, but that's how standards work. They're the law! Now pay up."
Criminal self-dealing.
I ran May's Caltrian ridership guesstimates for 2025 vs 2024, to see where new (or recovered) riders are coming from.
ReplyDeleteNo surprise, SF is and remains the 400lb gorilla of regional transit.
California Avenue, as it should, taking some Palo Alto ridership. Ridership up 120% from 2024. Another clear reason to KILL THE BABY BULLET.
Sunnyvale doing great, up 105%.
San Antonio continuing to surpass expections, up 90%. KILL THE BABY BULLET.
South San Francisco still disappointing in light of big peak service improvement, "only" up 80%.
Menlo Park only up 49%, slipping in ridership rank behind Cal Ave. Lawrence up only 58%, slipping behind SSF. (Again, KILL THE BABY BULLET)
First percentage is station ridership change (2025 - 2024) compared to total system ridership change.
Second percentage is change in stations percentage of system-wide ridership from 2024 to 2025.
Example: SF went from 139265 to 222683 riders, ie from 22.1% to 22.8% of system-wide, which went from 630027 to 978563
San Francisco: 23.9% +0.7% [hors catégorie]
Sunnyvale: 8.0% +1.4%
Palo Alto: 7.8% -1.4%
San Jose: 6.6% +0.1%
Mountain View: 6.3% -0.2%
Redwood City: 5.1% -0.6%
California Avenue: 4.3% +0.8%
22nd Street: 4.2% +0.3%
Hillsdale: 3.9% -0.3%
San Mateo: 3.8% -0.0%
Millbrae: 3.3% -0.8%
San Antonio: 2.9% +0.4%
Santa Clara: 2.9% -0.0%
South San Francisco: 2.6% +0.3%
Menlo Park: 2.5% -0.1%
Lawrence: 2.1% +0.0%
Burlingame: 1.8% -0.1%
Belmont: 1.7% -0.1%
San Bruno: 1.7% +0.2%
San Carlos: 1.5% -0.2%
Hayward Park: 1.4% +0.2%
Bayshore: 0.8% +0.1%
Tamien: 0.3% -0.2%
Morgan Hill: 0.2% -0.1%
Broadway: 0.1% +0.0%
Gilroy: 0.1% -0.1%
Blossom Hill: 0.1% -0.0%
Capitol: 0.1% -0.0%
San Martin: -0.0% -0.0%
College Park: -0.0% -0.1%
Another ranking generated from the May 2025 Caltrain guesstimates.(nb: NORMAL Stadler KISS trains come with automatic passenger counters. And ETCS/ERTMS. And cost 40% less. Great job all around, Caltrian and LTK Engineering Services! Gold medals for everybody.)
DeletePercent of total ridership by station EXCLUDING 400lb gorilla SF terminal
San Francisco: - 222683
Palo Alto: 13.4% 101002
Mountain View: 8.7% 65715
San Jose Cahill: 8.3% 63114
Redwood City: 8.0% 60137
Sunnyvale: 7.2% 54567
Millbrae: 6.1% 45888
Hillsdale: 5.7% 43241
San Mateo: 5.0% 37505
22nd Street: 4.7% 35868
Santa Clara: 3.8% 28504
California Ave: 3.6% 27312
Menlo Park: 3.5% 26706
San Antonio: 2.9% 21634
South San Francisco: 2.7% 20417
Lawrence: 2.7% 20056
Burlingame: 2.6% 19798
San Carlos: 2.5% 19163
Belmont: 2.5% 18730
San Bruno: 1.8% 13547
Hayward Park: 1.4% 10870
Tamien: 0.8% 6136
Bayshore: 0.8% 5999
Morgan Hill: 0.4% 2744
Gilroy: 0.3% 2239
Blossom Hill: 0.2% 1578
Capitol: 0.1% 1012
Broadway: 0.1% 998
College Park: 0.1% 835
San Martin: 0.1% 565
Lima TV broadcasters & officials excitedly fawn over Caltrain’s old gallery cars & F40 diesel locomotives headed to Lima via cargo ship in the first of two shipments from the Port of Stockton yesterday.
ReplyDelete(Enable auto-translated English subtitles if you don’t understand Spanish.)
Plans are to use the donated trains to open a much-needed new east-west rail service ASAP running at up to 10-minute headways and carrying an estimated 200,000 riders per day.
An unrelated earlier <a href="https://youtu.be/xuBB_tuBKg4>video</a> shows gallery cars being moved out of SMART-owned storage in Schellville.
An unrelated earlier video shows gallery cars being moved out of SMART-owned storage in Schellville.
Delete