While the full results of San Francisco’s Railyard Alternatives and I-280 Boulevard Feasibility Study, which is now known as Planning’s five-point Rail Alignment and Benefits (RAB) Study, are slated to be revealed this evening, here’s the executive summary:

1. As we expected and first reported last month, the recommended rail alignment to reach the Salesforce Transit Center is along Pennsylvania Avenue, with the rail lines undergrounded starting near the current 22nd Street Caltrain station and tunneled to the Transit Center, with one stop between.

2. Assuming the Pennsylvania Avenue alignment and undergrounding is followed, the Caltrain station at 4th and King, which is the current terminus in San Francisco, would be undergrounded and the railyard moved south.

3. The redevelopment of the railyard site could accommodate “between 1.05 and 2.43 million square feet of commercial [development] and up to 1.46 million square feet of residential space based on current zoning in the area.”

4. An additional rail extension/loop past San Francisco’s Transit Center could both increase capacity along the new line and enable a future rail extension to the East Bay.  And…

5. With respect to the potential of moving the north end of I-280 south: “While preliminary analysis shows removal is feasible, such dramatic change will require future analysis with Caltrans and other regional and city partners.”

While you might have seen “2026” bandied about with respect to the year the downtown extension (DTX) of Caltrain to the Transit Center could be completed, or “2027” for the full alignment, there’s a rather important footnote to that projection which some seem to have missed or ignored: “Presumes all money [for the extension was] available January 1, 2017.”

And as we outlined last month, “we’re looking at a mid-point expectation of around 9 years to complete the Downtown Rail Extension (DTX) once the preferred approach has been approved, which would suggest rail service to the Salesforce/Transbay Transit Center isn’t likely to be operational until 2030, at the earliest” (and with a couple more years to complete the full re-alignment to the south).

75 thoughts on “Results of SF’s Railyard, Tracks and I-280 Redevelopment Study”
  1. Please just build 1 or 2 tall towers rather than 5 blocky ugly mid-rise buildings that span the whole block. If it looks anything like the Berry street buildings across the street, it’ll be a huge failure.

    1. The greatest failing of the Berry Street blocks is not the height of the buildings but the total lack of any urban amenities at street level.

      1. that’s what i’m getting at. 1 or 2 sledner tall towers with open green space around it. As opposed to 3-4 ugly mid-rises that span horizontally across the whole block. Half of Berry street is loading docks and sterile apartment windows.

        1. You are most certainly not getting it. I’m talking blocks of stoops, subterranean entries, bars, restaurants, cafes. Certainly not some damn empty lawn surrounding a disassociated tower.

          1. One does not preclude the other. No empty lawns, but pocket parks, small greenways, outdoor artwork, water features and places to sit a read a book or people gaze. Also, of course, a shrub or two. The two important boundaries for a building are where it meets the ground and touches the sky. How well the ground level boundary is determines whether one has a delightful urban space – or not. The Pearl is an example of street level urban amenities done correctly.

    2. Would be great to have a Vancouver-like high rise surrounded by open green space in each parcel built.

      1. The “tower in a park” design is awful and not appropriate for San Francisco. Cities with a continuous streetwall of storefronts are much more walkable and interesting, and the design you propose is what ruins a lot of third world cities. Sure–build taller–but at least put midrise between the towers as in the new buildings going up along Folsom, not grass and shrubbery which will be allowed to die the next dry spell and will bed a prime sleeping/bathroom spot for the homeless in the meantime.

        1. Agree 100%. Put parks along the waterfront, and activate our streetscape. But mostly get rid of all the parking lots!

        2. “the design you propose is what ruins a lot of third world cities. ”

          Brazil! Sleek upscale towers with absolutely awful streetscapes. Brazil street, park and open space maintenance makes SF look first class.

    3. It’s a wall on the water and an example of failed urban planning. Too little was done to engage the creek. This is a case where substituting high rises for the blocky wall to wall buildings would have been preferable. If, and only if, much of the ground level space was left open and greened. As they do in Vancouver. The trouble is with SF planning and developers you’d get an 8 story podium building covering the whole block with a high-rise plinked on top of it. That would have been worse.

      1. Totally disagree. The creek is engaged in a very enjoyable way. A street wall is not an example of failed urban planning. It is the way cities have been built for 5,000 years until modernist loons like Le Corbusier came along and thought they could do it “better”.

  2. So they (even) can’t finance the terrestrial portions of the extension and now they’re proposing an (even more incredibly expen$ive) sub-aqueous loop.

    Burnham said to dream big, so at least he left us with that legacy (whether/not lunacy = big I’ll leave for another day. And yes, I realize this has been shown before, but the sheer insanity of it bears repeating)

    1. It was originally supposed to connect to the embarcadero freeway but in typical SF anti freeway fashion, plans were abandoned.

      For 20 years there was a exit at around 3rd & King with a freeway stub just hanging in the air, left from when the plans were to go all the way to the bridge. Then it was rebuilt in the mid-90’s with the ramp touching down @ King we have today.

      1. The Embarcadero Freeway was a crime against humanity and the city is better off with it gone.

      2. If all the freeways proposed decades ago had been built this would have been the result. Keep in mind thousands of people’s homes and apartment building would have had to be leveled to do it. Had they built more double decked freeways like the Embarcadero and Central Frwys they would all likely have been damaged in the 89 quake and torn down.

      1. I’d rather have 280 connect to 80 at 5th street – lots of room for it to connect there; it could even be tunneled from Potrero Hill and then surface @ 5th street. the 101/80 connecting curve is a horrible bottleneck.

        1. Agreed. Its current dis- connectedness creates lots of air pollution from stop and go traffic and basically 3-4 blocks if unlivable surface street. They should just connect the 2.

  3. Time to give it up, folks, and call it a day. With a soft target of 2030 for something that should have been operational yesterday this is all talk that we’ve heard before. Also, there’s the major clean up of the rail yard itself. Several square blocks of potentially contaminated soil. All that needs to be given the green light before any tunnel is built, much less huge developments on top.

  4. Hopefully they’ll phase this and get the southern part of this (to the new 4th St station) done in the next 5 years. The logjam around 16th & 7th is getting worse all the time.

    1. The plans for tunneling the tracks are, in fact, expected to be phased. But with the completion of the downtown extension to the Transit Center (DTX) expected to precede the completion of the southern section under Pennsylvania Avenue by at least a year.

    2. The 7th street/16th and up to Mariposa exit from 280 are a big deal, and getting so much more dangerous. 280 is routinely backed up a half a mile or more on the deck. I don’t see how it’s going to get better any time soon.

  5. Is there any major city in the world with a bizarre “train station” like we’re going to have?

  6. The Pennsylvania street option is probably the right one – but it is going to be interesting to build it anywhere near on time. That whole area at Pennsylvania and 22nd slid back in 1967 when they built the freeway. Lots of homes had to be abandoned – that part of Potrero Hill is known to be unstable. Plus you had the whole Western Pacific rail tunnel collapse a little further north under Potrero back in 1960. Seems like they would have to dig under part of that too.

    I’m sure technology has advanced – and it can be done- but I bet there will be a lot of nervous homeowners on the surface wanting the City to pay to fix all their drywall cracks and leaky pipes slowing the digging down.

  7. That’s probably the best alignment we can get. The DTX tunnel was to contain 3 tracks back when tunnel begun along 7th street. With the new design, will they continue the 3-track version? It’ll mean tighter turns and slightly slower speeds, but ability for two trains to arrive in parallel will save lots of time to passengers.

    1. The recommended approach assumes “3 tracks in the DTX tunnel and predominantly 2 tracks south of the 4th/Townsend underground station with crossovers as needed to allow for flexibility in operations.”

    1. For the purposes of Planning, and “to provide decision-makers with an order of magnitude analysis,” the preliminary estimate for the Pennsylvania Avenue alignment and extension to the Transit Center was set at $6.0 billion.

    2. BART is a bloated pig of a system

      We need convention rail access to SF. Any other BART tube must have conventional rail access too

      1. a geary BART lane makes tons more sense than another tube. with a geary BART, we couldd add 10s of thousands of units to the west side, and people could commute from within SF.

        1. I agree with both. Geary BART does make sense if it’s coming from a second tube and manages to connect with one of the three Caltrain stations – Transbay, 4th & King or 22nd. Today, there aren’t good transit options for anyone in East Bay to reach a job between South SF –> Hillsdale. Today’s 20-30 minute connection penalty from BART to Caltrain contributes a non-trivial amount of car traffic.

          On other hand, Bay Area has failed to provide good transit for anyone along I80 corridor north of Berkley. The BART line is merely OK, but because it’s detoured via Berkeley and Oakland, it takes an extra 25 mins vs non-traffic driving. That means that until traffic jams contribute at least 25 mins of delay, driving is STILL faster.

          However, trying to string cities north of Richmond with BART is hard due to lack of RoW and greater distances resulting in greater costs. That’s where 2nd transbay tube for conventional rail going into the new Transbay Terminal could help. South of Emeryville, dip underground (bonus if you can build a station under existing 11th St station in Oakland) and then run over to Transbay and come out as South Bound Caltrain.

          1. Traffic jams contribute to 25 minute delays on a weekly if not daily basis.

  8. Maybe they should consider just going all-in and dig a transbay train tunnel – then they can not only facilitate transfer to amtrak/capital corridor, but probably have easy access to zods of train yard space in the east bay rather than having to find space further south.

    1. An additional BART tunnel would surely payoff bigger than Caltrain to Oakland, but I do know that it would be popular. I work with scores of people who commute from greater Oakland to Palo Alto.

      1. Surely not. A convention rail tunnel means different trains can access SF and it can incrementally be added to as a network

    2. I think that the Oakland route for HSR makes much more sense, but that train has left the station (sorry).

      I think we’re stuck with the SJ route.

      1. It would go through SJ either way. The Altamont option was taken off the table years ago, and rightly so.

      2. It doesn’t make more sense. Caltrain Corridor is faster, better built up and HSR only had to chip in about $700 million towards electrification.

        BART took the best corridor in east bay and the two remaining corridors are largely single track, slow than Caltrain, and see more freight traffic than Caltrain line.

        Plus, Caltrain line can dump you right in the middle of the city with excellent connections via Muni, BART and AC Transit Transbay service.

        Oakland doesn’t have a central train station like transbay, so unless you spend billions on a new one + associated track, tunnels and right of way, you’d be dumped at the outskirts in either Jack London Square or Emeryville. Neither stations connects with BART, is in Central Business District or has much room to expand. Plus, lots of freight that will get in the way at both of them.

  9. Any guesstimate on a second BART tube pricewise? One going from Oakland through Alameda and onto the Peninsula between SSF and the airport? Given the huge number of East Bay residents commuting to the MV/PA/SM area it seems that project is far more needed, region wide, than this. The Central Subway was a waste of money in terms of bang for the transportation buck and this would be too. Let’s not make that mistake again here.

    1. Guesstimate: Take the cost of the central subway – divide by the length in miles – multiply by the length of the trans-bay tube – multiply by a factor of 2.5 for being under the water – add inflation, lawsuits, cost of delays… (disclaimer – the 2.5 factor for underwater is completely random, but so are the state estimates for any public works)

      1. Building underground in a built city is not easy and who knows if building underwater would be 2.5 times as costly. There are no obstructions to maneuver around, no need to protect the stability of surrounding structures and such. Not sure, but the Bay is shallower in parts as one goes south from the BB. It may be that a BART tube would not be 2.5 times more expensive to build.

        Either way, the cost of any transportation project is incredibly high and major cost overruns are accepted as a given. That speaks to the need for the region to identify the most bang for the buck transportation projects and go with those. A second tube to south of SSF would serve a whole new set of commuters not served today. Extending CalTrains to the TTC would not do the same. Of course this is hypothetical as I doubt Caltrains makes it to the TTC or that a second tube is built.

        1. Caltrain to Transbay captures east bay commuters who work south of San Francisco, close to Caltrain but away from BART. There’s a penalty of about 20-30 mins to connect between BART and Caltrain. Sure you can bike, but probably not when it’s raining. Muni / buses are unreliable. Walking is consistent, but long. Scooters are not always available and not everyone is comfortable due to lack of protected bike lanes between stations. A one-block underground connector between Montgomery (?) BART/Muni station will save these people about 20 mins which may make it significantly more competative than driving for many.

          Imagine if we could connect BART and ACE!

          1. “There’s a penalty of about 20-30 mins to connect between BART and Caltrain.”

            It’s a 10 minute walk from Powell station to Caltrain. Or you can take the 30 or 45 bus which between them run every 3 minutes. Or if [the] bus is too low grade for you, the N and T run every 8 minutes and take a bit longer. That’s before the T connects directly down 4th. If you bring a bike you exit BART at embarcadero and take a 5 minute ride down embarcadero/king. What’s your agenda here?

          2. 20 mins walk, plus you probably want to get to Caltrain 2-5 mins before departure + factor in a 5 minute buffer in case BART is late since if your station might not get serviced until 30 mins later. But yea, you get pretty far in 30 mins in a car even during 5:30pm rush hour.

      2. The bulk of the cost is stations, not tunneling. The BART transbay tube rests in a trench at the bottom of the bay. It was laid in sections, fused together and the water pumped out.

        The Bay Area will always have to deal with multiple rail systems (light, BART, commuter). The problem has always been connecting them for a relatively seamless integration. T-line and Caltrain still don’t have a direct connection at Bayshore. Millbrae BART/Caltrain still don’t have timed transfers which leaves many people waiting 15-20 minutes for a BART train. The Central Subway Union Square station requires a several block walk underground to connect with Powell St. station. The SFTC won’t have pedestrian tunnels connecting the bus station with Market St. to make transfers more pleasant. SMART trains are several years away from connecting to the Larkspur Ferry Terminal. The proposed 580 alignment of the Livermore BART extension won’t connect with ACE.

        1. I don’t understand why T never got a station at 22nd Street & 3rd which is center of Dogpatch. Instead we got two useless stations – with 23rd street stations placed on a corner of 3 parking lots – one for moving trucks, another for food trucks and third for ambulances. Both are also an extra 5 minute walk from 22nd street Caltrain station. Such a missed opportunity!

          1. The northbound platform is after 23rd and the southbound platform is after 20th, so in both directions you have a stop within a block of 22nd. Dogpatch is more than one intersection.

          2. Both add about 5 min walk to the 22nd Street Caltrain station and highest concentration of restaurants. Furthermore, 20th and 23rd station could’ve been combined into a single station that would not only be more convenient, it would save time for anyone passing through. Let’s not forget that money saved could’ve paid for the T Third turn-around loop.

        2. Don’t forget that SMART to Larkspur connection is a long 10 minute walk because station is built far away from the ferry terminal. I heard that was due to a law that triggers up-zoning of density for combined stations, but can’t find that reference.

      1. Thanks for the link. It’s not clear if the 12 billion includes going up Mission St as referenced in the piece. It’s hard to see how a tube from the East Bay to MB with a stop/and stopping there would cost that much. Beyond that the tube needs to go south of SSF. Going to MB and dumping folks off there leaves for a long walk to get to SOMA, the FiDi or Central SOMA. Plus it’s redundant in that many folks using BART then go on to jobs in SSF and further south. The more practical solution is to make landfall west of the SB BART station (or at the airport) and continue a line from that landfall to SB BART . And extend BART from Millbrae south the PA/MV. With BART to downtown SJ link coming it makes “bang for the buck” sense to complete the loop.

        There is a ton of planned office construction in SSF with Bayland’s 9 million feet being only part of it. Plus more office construction coming in the SV and a huge amount coming in SJ. Google alone will add 8 million feet near downtown SJ. A seamless system needs to be created that allows workers in the East Bay to get to the mid-Peninsula and south to the SV.

        1. I guess the 12-15 billion price range depends on the route. The central subway is costing close to $1 billion per mile. The existing tube is about as short as you can get it across the bay and it is 6 miles from station to station. The further south you move it, the wider the gap and the cost.

          The 6 billion to connect the TTC will be gone in no time and I am sure the intention is to get that paid by the state an federal government not local funds (BART).

          It looks like the under water multiplier is close to 2. And you have to wonder what the plan was when they proposed the TTC, they didn’t even determine how to get the trains in?

          1. To me digging up Mission St. or whatever to continue a tube coming onshore in the MB area is cost prohibitive and pointless. Going from Oakland to Alameda to SSF would be a longer route but offers more options at the point where the tube comes onshore to connect to regional transit.

            Between Baylands, HP/CP and several million more feet of biotech/tech space in the works for SSF you are talking 80K workers. Almost double the number the too jobs friendly Central SOMA plan would have. SF is otherwise approaching build out.The proper route for any future BART tube is to serve an area that will see the more intense job growth in coming decades. Also an area that is woefully underserved by public transportation as is the SSF biotech hub.

          2. Jack London Square to Bayshore is 10 miles by air – so now you are looking at 18-20 billion with no funding from Sacramento or DC. It is possible that you can tear some houses down and trench it across Alameda and save money on a mile of track, but you won’t be very popular. This is even less likely to happen than any of the other proposals, whether it is a good idea or not.

      2. The entire 75 mile BART system the voters voted on in the early 60’s was to cost about 3/4 of a billion dollars. By the mid 70’s when it fully opened that turned into $1.5 billion. Today that sounds like a bargain. $12-15billion today for a second tube sounds outrageous but in 50 years will sound like a bargain.

        1. People forget that much of these projects get amortized over 100 years – as is [the case for the] NYC subway.

        2. it would be nice to cancel the BART to livermore and use that money to get Geary BART. it could carry so many more people

          1. There is no “BART to Livermore” to cancel. Only about a third of the funding for an extension to Livermore has been identified. Of the identified funds, 80% is from Alameda County sale tax and City of Livermore impact fees. That money is not going to be spent in San Francisco.

          2. Correct – but if they were to extend BART to Livermore, why wouldn’t they continue the tracks to connect with ACE? And why does it cost 1.5 Billion when the tracks would be in the center divider of the freeway? Wouldn’t the existing grade separations work?

          3. Right, BART to Livermore is pointless without an ACE connection. Fortunately the freeway station option was not approved by the BART board. As for why the freeway centered option costs so much: this is because the freeway would have to be nudged outward to make room for BART. A large part of that $1.5B is simply to move and rebuild the freeway.

      1. Indeed. However, once Caltrain is electrified it can be converted to a more interurban-style system. In this case it can run along its current route and rather than end at the SFTC the station could be built under 2nd St. between Market and Howard which would allow it to connect directly to the SFTC and BART/Muni then continue its route under Post to Geary west of Van Ness all the way to the outer Richmond with a spur running south under 19th Ave to Daly City BART. If HSR ever gets built it can terminate at 4th/Townsend which would provide direct access to two Muni lines and the interurban Caltrain. This same interurban system can also run along the Dumbarton Bridge to the East Bay for further connections to BART/ACE/Amtrak.

        1. This would be fantastic, especially if we could run it through SFTC and under the bay and link it into capitol corridor. The ability to quickly get from the south bay to Sacramento would be fantastic.

          1. Very true. The way HSR is planned any SF to Sacto trip would require a trek through SJ and over to the Central Valley and a transfer up to Sacto. No direct connection to the Capitol Corridor for SF even after all is said and built. Regardless of BART or another rail system, a second bay crossing is required to successfully link up all these disparate systems.

  10. It seems they could incorporate much more land use and zoning changes to capture the value created through the surrounding redevelopment to offset the high costs of construction.

    The area could be upzoned to 400 feet making it more valuable and fees directly going to transit upgrade including the tunnel and a 2nd BART/Train tube. The second bay tube crossing could use the deep single tunnel method like they are doing with SJ BART (and Barcelona) and they could run 2 tracks of BART on one level and 2 tracks of conventional rail all in one new tube from Oakland under Alameda to Mission Bay.

    If Alameda doesn’t want to upzone around a new station and help pay, then just tunnel under without a stop going directly to 19th BART station in Oakland.

    1. While the study results aren’t slated to be presented to the general public until the 29th, the draft report, which we both summarized and directly linked above, was presented to the Citizen Working Group last night.

      1. Ah, ok. The way it was written in the synopsis, it sounded like the results were going to be released last night.

Leave a Reply

Your email address will not be published. Required fields are marked *