Bay Area Metro Rail

New BART Station Brings Infill Thinking to the Bay Area

» A new stop at West Dublin/Pleasanton could attract new riders and transit-oriented development without requiring further line extensions.

With 104 miles of track and just 43 stations, the San Francisco Bay Area’s BART system may have the most widely-spaced stopping pattern of almost any rapid transit system in the world. One wonders whether those huge inter-station distances reduce ridership by making it too difficult for people to get to and from stops by foot. Washington’s Metro, which was built in essentially the same period, has almost the same track length but twice as many stations — perhaps that is one of the primary reasons that it also has nearly twice as many daily riders?

Today, BART has taken a step forward to remediate the matter, opening a new stop at West Dublin/Pleasanton in the median of I-580, near the freeway’s junction with I-680. It is the first infill station — a stop constructed along an operating rail right-of-way — for the system and fills what had been a 10-mile gap between Castro Valley and Dublin/Pleasanton stations in the far southeast section of the region. The station cost $106 million to build and is expected to attract 4,300 daily users. $20 million of the construction funds were sponsored by Jones Lang Lasalle, a developer that plans 210 housing units, office space, and a hotel within walking distance.

The station was originally planned as a part of the Dublin/Pleasanton Extension, which opened in 1997, but implementation was delayed. The project also added 1,200 parking spaces for the large car-commuting population expected to use the stop. Reverse commuters, however, may also be expected to use the stop: It is within close distance of the Stoneridge Shopping Center and the Safeway Grocery Store headquarters.

Like Washington’s New York Avenue Station, which opened in 2004 — 28 years after the rail line on which it is located was constructed — the West Dublin/Pleasanton Station represents a new way of thinking about the right way to plan transit investments. Though BART continues to focus on suburban extensions — projects to Livermore, San Jose, and Antioch are either under construction or planned — it has plenty of room for infill stations.

These have a number of significant advantages over line extensions. For one, it costs less money to build a new station along an existing corridor than to extend the same line further out. In addition, by adding service to a neighborhood that has been overlooked by initial investments, the new station can encourage new transit-oriented projects in-town instead of encouraging further suburbanization. When done right, these sort of infill projects can bring welcome improvements for neighborhoods that suffer from a dearth of walkable urban areas — and they can be very popular, as has been proven by the new construction around the BART Fruitvale Station in East Oakland.

From an investment perspective, building infill stations could be an appropriate response to limited funding for new transit capital projects, especially since it appears private developers may be interested in helping to chip in for construction costs. There are good reasons to build new transit lines in dense sections of the Bay Area, but especially in the East Bay, there are plenty of opportunities for infill stations to fill the 2 or 3-mile gaps between stations. Though these would marginally slow down services from the far suburbs, they would more than make up for that loss by greatly increasing the number of people living in already developed areas within easy walking distance of rapid transit.

It is too bad, however, that apart from the West Dublin/Pleasanton Station, BART has no infill stations planned. Nor is it alone on this matter: Cities with extensive commuter rail and subway networks in the United States, including New York, Chicago, and Philadelphia, have been more interested on extending their lines out into the suburbs than filling them in. One notable exception is Boston, where four new stations are planned to be added to the Fairmount line to add to the transit options for people living in underserved neighborhoods south of downtown.

Image above: BART’s new West Dublin/Pleasanton Station, from BART

82 replies on “New BART Station Brings Infill Thinking to the Bay Area”

Excellent post. BART’s interstation distances are indeed a wonder of the transit world. BART is basically rapid transit operating mostly in geography that’s ideal for commuter rail. Caltrain, of course, is exactly the opposite …

To be fair, BART carries people much longer average distances than DCMetro, so a count of riders (rather than passenger-km) doesn’t quite capture its value. (And yes, you can retort that high average trip lengths are actually a signal that you may be facilitating sprawl.)

As inner Bay Area populations grow denser, you’ll see more agitation for infill stations — not just at relatively blank-slate places like West Dublin Pleasanton but also at existing points that are already natural TODs. The most obvious are 30th Street in San Francisco and Albany just north of Berkeley. These would have huge local impacts, so local governments will be ambivalent, but expect them to continue being raised as options.

I honestly find it difficult to read about BART expansions nowadays. This infill (and the one they have studied at 30th/mission) is wise, but their continued extensions to the fringe arent going to help at all.

On this same topic, im tempted to disagree with this articles assertion that DCs double ridership is due to more closely spaced stations, most of which are in the suburbs. Stop spacing appears much more comparable to DC metro within SF, it just needs more lines. Note that DC has multiple lines through the core city. This is a little more difficult in the Bay Area due to SF not being in the geographic center of the area. DCs additional core city lines are not intracity subway lines as they would essentially be in SF, they are on the way to suburbs in opposite directions from the city.

In Oakland some infill stations on the existing lines would definitely be useful.

To me, this map demonstrates that DC metro’s ridership profile is much like BART’s. The way to double ridership is to build more in the core, not more stations like West Dublin/Pleasanton, as nice an infill station as it is due to the gap that existed there.

Transit is tribal in the SF Bay Area. The best we can hope for is universal ticketing and easier connections between BART, Muni Metro, Caltrain, Santa Clara County light rail and ACE commuter rail.

Thanks for the update. I wasn’t aware that Clipper had finally been approved after many years of trying. How well is it working?

Now it will take another 15-20 years to make smoother connections between them.

Fair-to-middling. There are the usual problems – re-load machines are more complicated than they need to be; some users are un-used to the sensor location on BART’s handicapped gates (on the side, about knee-high, instead of on top); SFMuni buses and LRV’s often have one or more down sensors (I suspect the drivers haven’t been trained to do a reset); and some boarders haven’t mastered the ‘touch’.

I think the breakthrough in Clipper adoption came as a result of the switch-over month’s sale of Clipper cards loaded with SFMuni “A” passes (SFMuni + BART in S.F.) at the usual Fast Pass outlets. That made it very convenient for users of that pass to get a card in the new format.

If you want some gory details and how-to’s try a local transit blog – Akit’s Complaint Dept. It’s an ex officio Clipper support site.

We I look at the Muni Central Subway, my take is that Muni plans two extensions from it in the future:

1. North Beach-Cow Hollow-Marina District
2. Geary Blvd west

When viewed from that long term perspective, those extension could easily add 100,000 Muni riders daily.

You seem to have a fair amount of knowledge about Muni & BART in SF. I’d be interested in hearing your take on the Central Subway.

The Central Subway is a ruptured duck on multiple levels :
1) Not going far enough on the northern end (one more station would be an improvement – see third link below);
2) Deep tunnel option the worst choice (cost reasons);
3) Failure to implement a transit mall on Stockton St. as an interim measure; and
4) Pencil-whipping the plans into uselessness.

The SFMTA’s (parent agency for SFMuni) outreach efforts are lacking in clarity on a key point – How close is the Union Square Stn. to Fourth + Market ? Their diagram (see 1 + 2 below) is blurry and could do with some plain English that notes both vertical and horizontal distances. I get the impression that I’ll have to walk a half-block or more underground to get to the exit. Are SFMuni’s passengers moles or tunnel rats ?

Contrast this to the Civic Center Stn. (West). There’s a staircase (no escalator on this side) just to the north of the fare gates that puts you on the surface NE of Eighth + Market in front of the theater.

1] “Central Subway Overview” (
2] CS Project Alignment diagrams (
3] “New Connections” ( report, 20 June 2007)

NB – I’m a railfan who’s lived in or near S.F. for most of his life.

Re : Geary Subway
1) The were plans in the early phases of BART’s initial construction.
2) When the streetcar tracks in the Richmond District (a sector bounded by : the Presidio, the Pacific Ocean, Golden Gate Park, and Arguello) were in the process of demolition promises were made that there would be new tracks laid, possibly BART.
3) Geary Blvd. from Arguello on west into the Forties (Aves. not years) is very wide due to it once having tracks on it.
4) Even with a TBM carving a tunnel will be both challenging (a high hill at Masonic + Geary, a ridge above Van Ness Ave., and a heavily built-up downtown) and expensive.

In short, the Richmond District got shafted and the best deal for all will be BRT. This balances cost, construction time, operational flexibility, and service improvement by upgrading the current express buses.

Part of the problem for San Francisco is that the Market Street Subway (MSS) is that city’s “Great Wall”. Crossing Market St. below the surface leads to a deep bore tunnel which put the price tag on the Central Subway Project almost into orbit. Running along the north side of the MSS, an alternative for a Geary Subway, would have to deal with a lot of skyscraper foundations in return for tie-ins to the Montgomery and Embarcadero Stns.

There is a hybrid concept of LRV’s via Duboce Tunnel, Golden Gate Park, Arguello, and then west on Geary Blvd. This could require an expansion and extension of the current Duboce Tunnel out to a portal on the north side of Golden Gate Park. Unfortunately, this concept bumps up against capacity problems in the MSS.

To borrow from MLK – “I have a dream …”

You don’t want to go there – counting passengers instead of passenger miles. In nice round numbers BART has 300k and the DC Metro has 750k.

L.A. has taken so much grief for the subway construction, admittedly some of it is deserved.

On the other hand, L.A. was supposed to have a BART-like heavy rail system until reality intervened and we were stuck with downtown L.A. to Koreatown and North Hollywood.

Yet, BART, as built out as it is, manages 300K boardings across 100+ miles. L.A.’s subway, manages more than 150K, but only along 17 total miles of heavy rail.

The other problem is that BART, unlike L.A., has already had its moon-shot. (L.A.’s moon-shot is the Century City and Westwood area, which will put the subway well over BART’s 300K and with less than half of the track miles.)

BART’s already serving its busiest markets as it runs now, and all extensions at this point are diminishing returns. Ringing the bay won’t help.


Sorry to diverge from BART and SF Bay, but please indulge me a moment before returning to BART & SF Bay Area.

LA Metro may have I-405 Corridor as it Moonshot II, depending on price of gas by 2017 or the next Surface Transportation Reauthorization date by Congress. The price of gas by 2017, will drive up public demand for more Transit funding in $22B/year level requested by Obama. At that level, additional major project grants can be awarded to LA, which will be the highest merit, least transit served Metro Area in the nation. That will make it possible for LA to extend Purple Line to Santa Monica, Red Line up to Bob Hope Airport and a new line down I-405 Corridor, but not dependent on freeway median stations.

Imagine I-405 Heavy Rail or grade-seperated Light Rail starting construction in 2018 and finishing about 2025 with these major feeder stations:

Sylmar/California HSR
Van Nuys/MetroLinkAmtrak
Van Nuys/Orange Line
Sepulveda/Sherman Oaks Galleria
Getty Center (if Getty Trust pays for it)
Sepulveda/Santa Monica
Sepulveda/Culver City
Sepulveda/Jefferson/Westfield Culver City Mall
Sepulveda/Hughes Center
LAX/Century Station

That sounds to me like at least 200-250,000 patrons daily.

As for BART, once Downtown San Jose is reached, their best extensions are taken. The Central Subway pretty much assures that Geary Blvd and Marina District will be reached by Muni, who will bring in close to 200,000 more patrons with Central Subway plus those extensions.

I seem to remember, as a cost cutting measure(s), the platforms on the Central Subway keep getting shorter and shorter. They won’t be long enough for that many riders.

Though BART from Warm Springs to downtown San Jose won’t break any records for patronage efficiency, it is worth doing. There is pent-up demand for a Silicon Valley-East Bay rapid transit.

BART won’t extend from downtown SF out Geary Blvd due to the tribal nature of SF transit politicos, which favors Muni Metro. As long as it maintains full-grade separation from roadways past Divisidero, Geary Blvd Muni Metro will be a good transit option. By any FTA patronage measure, Geary Blvd deserves full-grade separation that far.

Extending BART to Downtown Livermore should not occur until Geary Blvd Muni Metro completes around 44-45th Street, even if it takes 20 years. May be then extending BART to Downtown Livermore will make sense.

It might help if they would go ahead with the proposed Irvington station. After all, more riders would tend to help this extension’s success.

Having lived in Fremont and knowing the nearby shops/restaurants, Irvington will eventually be a good infill BART Station. BART’s highest priority is to make it to downtown San Jose, so the funding to build a station isn’t there. Wisely, City of Fremont/BART set aside the space for the infill station and I’m sure they hope to attract TOD to help build the infill station.

Re : 30th St. (near Mission)
On the map this looks like a great place for an infill station. But it would be a perfect storm to build due to geology, tunnel configuration, and surface transit. The soils are a hodge-podge, the existing tunnels make a turn in 3-D as they turn away from Mission St. and head South to Glen Park, and the area around 30th + Mission Sts. is served by six transit lines (#14, #14L, #24, #36, #49, and J-Church).

The extensive service along Mission St. (#14, #14L, #49) reduces the need in general by providing a link from 24th St. on down to 30th St. and beyond. Similarly, for those traveling from Glen Park to 30th St. there are two options – #36 and J-Church.

The heavily residential nature of the neighborhood (commercial corridor along Mission St. with a little on San Jose Ave.) combined with a station proposal would be like throwing a lit flare into a tinderbox. (Note – the tinderbox has an accelerant present – the mess along Mission St. during BART’s construction.) So, it’s a great dream, but the practicalities are murder.

“System Maps” (SFMuni aka

P.S. I deliberately left out cars and trucks from the above. FYI – San Jose Ave. is a feeder to I-280 South and is fed by I-280 North. That’s still another vocal interest group that would be screaming over the impact of a station proposal.

i guess an ‘infill station’ is a new station being plopped into/onto an existing line?

if so, that’s all fine and good (actually, probably not — because this one way built to give people more places to park their cars — really), but this particular new station, like most of the existing stations, don’t have anything but parking around them.

and, as is obvious from the pictures, this station is a disaster – plunked down in the middle of a freeway, like many others.

regardless of where the BART stations are, almost without exception, it seems very few actually allow anyone to live in relative proximity to them. the one station that does, Fruitvale, has been a relative disaster for various reasons, which is probably why it continues to receive all sorts of awards from ‘new urbanists’ and other folks who just decided to declare victory when we finally built something that wasn’t a complete waste of taxpayer money.
so it goes…

Unfortunately BART was designed and built during the same era as the Interstate highway system, so similar design techniques were used. As you rightly pointed out, this resulted in a network that mostly focuses on parking at the expense of walkability.

However, in BART’s defense, some stations have succeeded in providing good transit within an easy walk of shops and homes. I am lucky enough to live less than two blocks from one such station: North Berkeley. Other “blessed” stations include Downtown Berkeley, Ashby, El Cerrito Plaza, 16th/Mission, 24th/Mission, and Rockridge. Interestingly, Rockridge works despite being in the middle of a freeway!

The success of Fruitvale is mixed. Redevelopment around stations like Hayward and Balboa Park are likely to be more successful. Infill stations, especially at 30th/Mission in San Francisco and Solano Avenue in Albany, would be clear winners. (They would also have small footprints because parking would not be needed.)

I think that Union City BART station deserves some mention here. I got on and off there a couple of times back in October and was really quite impressed by the residential developement I sae by it. On a visit back in 1992 I can’t honestly erecall any of it. People around there still have to cross a parking lot to get to the station but the developement there is still quite nice.

It’s always a good thing to bring stop spacing in line with what’s normal for rapid transit. That said, the number of places on BART where it’s feasible is quite limited. The difference between BART and the Washington Metro is not just stop spacing: it’s also that Metro has three lines in the city core creating a reasonable subway network there, whereas BART has a single line that misses some of the densest neighborhoods.

We can advocate for more stops in the inner East Bay, and maybe even hope for large capital funding to build infill stations underground in San Francisco, but BART still has a serious problem with not serving corridors with good density. That’s why urban-oriented extensions like Geary and a second Transbay Tube would be so useful.

Great Post, and interesting that they were able to get Jones Lang Lasalle to buy in to the new station. Just wanted to note that the CTA currently has two infill stations under construction.

The new Morgan Lake station will provide access to the West Loop via the Green and Pink Lines, and better access to downtown Skokie will be provided by the new Oakton station on the Yellow line.

Yeah, Chicago’s been fairly proactive with infill stations. In addition to the Morgan and Oakton CTA stations, Metra’s building a new station at 35th Street.

Under study are CTA stations in the South Loop (18th and/or Cermak), additional Yellow Line stations in Evanston, and a Metra station at 79th. The mostly-dead Circle Line plan also included several infill stations.

It’s unfair to accuse Chicago of not pursuing an infill strategy. All of the CTA lines, save Yellow, are already built with stations at a 1/2-mile interval or less. Where infill stations make sense, they’re either being studied or under construction. Metra has a less-than-enlightened attitude about providing transit to inner-city neighborhoods, but even they’re doing the infill thing in a limited way.

This is not an infill station. At least not how most would define that term.

Because BART ran 100% over-budget on the original Dublin/Pleasanton extension project, the West Dublin station was left unbuilt until more money came along.

Moreover, the funding mechanism BART used to build the station is creative. They defined the station as a TOD, even though it is plonked in a freeway median, and the vast majority of the station cost was in building the gigantic parking garage. So in effect, TOD funds were used to cover cost overruns of a park-and-ride station that is anything but TOD.

Infill? Only from a technical point of view. This is just a way to get more parking; situated not only in the freeway median but next to a spawling car oriented shopping mall and business parks. BART’s extensions to Antioch (no more than a parking shuttle) and Warm Springs (in the middle of an industrial area)are no better. As Alon Levy say there are few infill locations due to the “passive” alignment BART initially chose. However, stations at Irvington (Warm Springs extension) and Solano Ave. (Richmond Line) have apparently not gained much traction due to their “impacts”. Apparently, the idea of a (car) parking free station is beyond comprehension.


BART appears to have decided housing adjacent to stations is a good idea and has been trying to encourage it near its stations. For example, the Pleasant Hill BART station has just replaced its surface parking lot with apartments and the West Dublin station was partly funded with revenue from the planned TOD at the site. Replacing surface parking lots with housing is also planned at Walnut Creek, Richmond, and Dublin/Pleasanton, and San Leandro. BART is also trying to convince the good folks of Pittsburg, San Leandro and Hayward to allow them to do the same thing.

Having lived in two and worked in the other one, I can vouch NY, Philadelphia, and Chicago all have stations at rather close intervals already. “Infilling” on most of the NYC lines, for instance, would mean having three stations within half a mile–slowing down trains for no gain in a city where buses are readily available.

As noted above, the “infill” BART station is just a previously-planned station that was underfunded. “Infills” will not improve overall service as much as extensions and new routes do. NYC, PHL, and CHI know that, as do the BART planners who had the foresight to know they wanted to add a spot and left enough of a gap to do so.

Actually, Chicago could benefit quite a bit from infill and knows it. The CTA currently has one of the highest miles/trip ratios of any American heavy rail system, which combined with a flat fare structure really hurts its revenue. Adding stations in core areas with a lot of growth is a much more cost-effective way to add passengers. The Green Line had a number of stations removed in the late forties and early nineties in response to shrinking populations, but now those areas have turned around and become the South and West Loops. An infill station’s currently under construction at Morgan, filling an approximately 1.5-mile gap between Clinton and Ashland. Chicago has also looked into filling the 1.5-mile gap between Ashland and California: a station at Western has been studied and deemed worthwhile, though there’s no active planning or funding, and either the CTA or the City of Chicago recently applied (and failed to get) a federal grant for an infill station at Damen. In the South Loop, there are two separate TIFs for stations at 18th and Cermak (although they only have enough money for one station—although the CTA chose Cermak, with Daley leaving that could change).

Morgan is costing $40 million, so if we assume two additional West Loop stations and one South Loop one we get $160 million for four new stations. In contrast, the Red Line extension south of 95th will also have four stations and cost over $1 billion. Although I can’t compare riders because I haven’t seen any non-back-of-the-envelope ridership estimates for the Green Line infill stations, the infill stations won’t require the CTA to buy new railcars or build a new shops-and-yards facility, nor will they increase the number of miles the cars need to travel, saving on operating and maintenance costs. Although you can obviously go too far (why the CTA is considering consolidating platforms on the Red Line north of Belmont, where stop spacing typically is a third of a mile or less), to say that infill isn’t as good an improvement as a new route or extension doesn’t really hold up once you bring cost-effectiveness into the picture, especially if you’re talking about a dense urban area.

Interesting read, have to agree with some of the comments that you are really not seeing either a true infill nor a TOD with this. However, it is good to see something kicked down the road happen, a developer seeing an opportunity, and maybe just maybe give some commuters a non-car optin even if it takes a parking garage as well as a shopping mall to make it happen.

A part that is missing is where the station is exactly located relative to its immediate area/surroundigns. In other words, I think a quarter mile can make a big difference in a lot of places. What is unique in the case, their was a second chance to rethink a station location.

In my opinion the immediate area around this new station was a relatively a dead zone. It seemed like their was much more of a built environment East of 680 when you are going actually going eastbound on BART. I’m sure the 580/680 interchange has a lot to do with that fact.

What I wonder and think would be an interesting question to at least discuss even though the station is already built. Would it have been better if the station was East of 680 instead of its current location? Yes, you might give up on the garage and developers money. But you might gain the opportunity to see a more TOD orientation.

Lumping New York into the in-fill argument that is made in this article isn’t a realistic comparison. The Subway and Commuter Rail in New York City are basically separate systems whereas BART is a hybrid that combines the two. There are actually very few areas where in-fill would be useful on the NYC Subway system and it is not being extended outward towards the suburbs at all. New lines are being constructed or added onto in the already dense coverage within NYC (the 2nd Ave line on the upper east side and the 7-line extension to the west side of midtown). Saying that NYC needs in-fill stations and is actively doing that doesn’t strengthen the arguments in this article. Presenting NYC as an example of somewhere that has pretty decent coverage and it has promoted denser development could be helpful to your arguments but the challenges of the BART system and NYC’s regarding in-fill really aren’t comparable.

It has been suggested (albeit in the transit blogosphere) that LIRR, for instance, could be converted to more modern metro-style operation (like the RER or S-Bahn, or even BART/DC Metro), with infill stations within NYC. This could mean better rail coverage in areas that could use it, without the cost of building new subway lines. An overhaul of the fare structure would help as well.

Id personally love to see LIRR get turnstiles like BART and do away with ticket checking. Its too expensive otherwise (and LIRR has some serious labor cost issues). The more frequent trains and lower dwell times that would result could offset the extra time of stopping at infill stations.

Obviously this would mean changing many of their stations, as they are usually not designed for controlled entry, but at least from the stations ive seen (Babylon line along Sunrise Highway) its probably not that big of a deal.

Adirondacker, this is actually an easy question: Theres plenty of room in the current LIRR terminal for partitions and turnstiles.

As for what Alon Levy said, it is indeed a huge disappointment, and we can only hope they revive the station shell. This isnt an LIRR or MNRR extension to the west side yards (well, obviously not the former) where it might make sense to only have a stop at the “destination”. The subway should have an intermediate stop.

There are pedestrian traffic jams on the streets around Penn Station and in Penn Station now. Putting in turnstiles wouldn’t help that. Roughly 1000 people a minute at peak. For three different agencies. People also use the station to pass from the subway to the Avenue on the other side of the station. Where do you put the turnstiles?

New York if anything needs some station closures, for example 18th Street on the 1. It has one of the shortest interstations in the world (second only to Paris among major networks), just as BART has one of the longest. But yeah, the LIRR is something different, and some infill could be very useful there.

By the way, although it’s possible to put turnstiles at Penn Station, in place of back offices, there’s no point in doing so. POP with random inspections incurs lower collection costs, unless you’re Tokyo or another city with extreme crowding levels. But the ticket punching and the idea that every passenger on every train must be checked both have to go.

There used to be an informational display about the East Side Access Project in Grand Central Terminal. One point impressed me. Currently the rush hour trains coming in from Jamaica are so crowded with standees that they cannot serve the Kew Gardens, Forest Hills, and Woodside stations. But after the East Side Access line diverts many trains to Grand Central, crowding will be reduced. Then these stations will provide residents of Queens with new transit-like service. Hope that turns out to be true. And if it does, the map suggests room for one or two more infill stations in Queens.

Forest Hills and Kew Gardens only get hourly off-peak service, and at rush hour the East River Tunnels are not at capacity. If the LIRR were interested in offering urban service, it would easily find capacity for it.

That said… infill on the LIRR and Metro-North should not necessarily be formal infill as much as providing more trains stopping at the existing local urban stations. I’m not sure there’s a point in reopening Rego Park, but providing higher frequencies to Forest Hills is a no-brainer.

I still can’t believe NYC isn’t building a station at 10th Ave/41st St. on the 7 extension. Hopefully sometime they put a previously planned station on the 7 line there like they are doing on the BART line currently.

The city not only dropped the station, but also dropped the option for building a station shell that could later be expanded to a station. Once the tunnel is complete, building an infill station underground is prohibitively expensive.

It was one thing for them to drop that station but why did they drop the station shell? That sounds plain stupid to me.

The shell would still cost a couple hundred million dollars. Bloomberg doesn’t really care much about that area, because “it’s already developed,” so he’d rather drop the shell than admit there was a cost escalation and spend the extra money.

Infill stations are great when– like NY Avenue in DC– they are surrounded by potential redevelopment. That doesn’t mean that the area around the West Dublin/Pleasanton area can’t be redeveloped a la Tyson’s Corner; however, I think in BART’s case, I would have preferred to redirect the line directly into the heart of that commercial area south of the 580. That would have been more expensive, of course.,-121.91494&spn=0.023091,0.061798&t=h&z=15

There are a lot of reasons to look at the BART & WMATA comparison and take a harsh view of the Bay Area’s approach to regional rapid transit. Points focused on how each system addressed stop spacing, TOD, the application of the proper transit technology, and and serving (or not) the densest urban neighborhoods are on target. Some other points to consider:

First, given that the most dense urban neighborhoods not served by BART are in SF (Geary), and Muni is the primary transit agency for the city, you could make an argument that some of Muni’s light rail/trolley lines should be included in any transit comparison of the Bay Area and the Washington metro region (rather than strictly by agency). While we can debate about the usefulness of these lines as rapid transit, it’s still relevant considering Muni serves the densest part of the region. When DC and the MD MTA add the new streetcar and LRT (Purple) lines that may not be operated by WMATA, the comparison will become more appropriate.

Then there’s the spatial differences. Both regions have water within their boundaries, however the SF Bay is much larger in area, and separates the most urban, transit-friendly locations in the region, SF and Oakland by about 9-10 miles. Comparatively, Arlington, one of the most walkable urban areas outside Washington proper is just a few miles from the DC core. Consider a thought experiment where the Bay was much smaller in size and Oakland was only a few miles from Downtown SF (as Arlington is to DC). I would expect the attractiveness of having more developable land area within close proximity to the SF core would make Oakland a much larger, more populous more economically attractive urban area, which would have required much more extensive rapid transit service. Also, if BART and Muni were not separate entities, maybe BART would have been extended within the city along the Geary and other corridors where Muni operates today, adding new BART service within some of those dense urban areas not now served by rapid transit.

Finally there’s the DC height limit. Would NOMA in DC, Arlington’s Rosslyn-Ballston corridor, or Silver Spring, Bethesda, etc, have developed in the way they have if skyscrapers were allowed in the DC core like they are in San Francisco? I would posit that a sizable percentage of the growth of those areas would not have occurred as it did, thereby reducing the transit demand (especially bidirectional) on the corridors between those areas and the DC core (major sources of Metro ridership). Conversely if there was a strict height limit imposed in SF or Oakland, I would think that a significant portion of the growth that occurred in those cities’ cores would have been pushed to other areas like Berkeley, Richmond, South San Francisco and San Mateo, thereby creating a larger demand for suburban TOD and a higher transit demand between those places and the cores of SF and Oakland.

At the end of the day, things didn’t happen that way, and so there’s still plenty to criticize about the results that the BART system has achieved.

Re : BART + SFMuni
You make a good point about adding a portion of SFMuni’s service to BART when comparing BART to other agencies. Inside San Francisco the two are partners – BART is an express service from Balboa Park (J/K/M) to points North and SFMuni’s Metro is a cost effective extension of BART’s service along Market St. on out to West Portal. One could also consider the N-Judah as a slower spur that begins local service* in the Cole / Haight / Parnassus (UCSF) sector.

*The N-Judah is a light-rail line that follows a ltd.-express-ltd.-local pattern. It leaves the Market St. Subway at Duboce + Church, runs for a few blocks along Duboce, and then bypasses a steep hill via the Duboce Tunnel (aka the Sunset Tunnel). SFMuni’s tunnels (see also Twin Peaks Tunnel) are a large part of why San Francisco continued to have streetcars.
NB – “ltd.” above is more a function of the ROW than the stop spacing.

Jonathan and Ted,

Here’s a few points to ponder when comparing SF Bay Area rail and ferry transit to its transit-rich peers in Washington DC, Boston, Chicago and Philly today and by 2030. I’ll leave Bus Rapid Transit for others to calculate, but I dod include LA rail transit hich is rapidly moving in the right direction.


SF Bay Area daily rail transit numbers include BART, Muni Metro, Streetcars & Cable Cars, Caltrain commuter rail, Santa Clara county light rail and ACE commuter rail. I couldn’t locate SF Bay ferry transit numbers, but I estimate 5K and welcome more accurate data. I have no idea how to measure Boston ferry transit. Washington DC numbers include VRE under commuter rail transit, but not Batlimore-DC MARC (which I treat as Baltimore’s commuter rail). I collected this rail and ferry transit data about 6 months ago, so the numbers could be +/- 2% today:

SF Bay Area 360K 200K 43K 5K = 603K
Washington 800K 0K 16K = 816K
Boston 480K 247K 145K = 872K
Chicago 640K 0K 330K = 970K
Philadelphia 337K 63K 130K = 530K
Los Angeles 160K 135K 40K = 335K


Muni Metro extension North Beach-Chinatown-Central Subway-Convention Ctr-Caltrain Station adds 80K; Muni Metro extension out Geary Blvd adds 90K; Muni Metro extension out North Beach-Fisherman’s Wharf-Marina District adds 30K. BART Fremont to Downtown San Jose extension adds 90K, without two possible infill stations. Caltrain electrification for 90 mph, positive train control, 110 daily trains from San Jose to SF completes in 2015 and fully grade separated by 2019, as part of the California HSR upgrade and Transbay Transit Center opening; combining all the Caltrain improvements, add 40K to match the additional capacity they create. Santa Clara County and Sonoma County light rail projects likely add 20K. ACE commuter rail and BART DMU add another 10K commuter rail by 2030.

Adding total patron additions for Muni Metro 200K, BART 90K, Caltrain 40K, Other Light Rail 20K, Other Commuter Rail 10K = 360K more SF Bay Area rail commuters. That would place the SF Bay Area near 1 million daily rail & ferry commuters without estimating California HSR boosting the rail transit & ferry numbers.

I’m hard pressed to envision Chicago, DC, Boston and Philly adding 360K more rail transit patrons, unless the price of gasoline is an inflation adjusted $10/gallon. Of course, that would boost everyone’s rail transit projections.

With all the construction and planning activity underway, I can envision transit underserved Los Angeles tripling current rail patronage from 335K to 1 million daily by 2030.

The comments on “New Bart Station” are much stronger than the post itself. Congratulations to the readers.

Infill station planning is happening at many locations on the MTA Metro-North’s New Haven Line, the busiest commuter rail line in the country: East Main Street in Stamford, Fairfield East (near Blackrock). Orange, West Haven.

BART can talk about infill when there’s a Solano Ave station – otherwise their latest exurb station is just a lot of the same BART BS.

Building the infill station went smoothly for BART, but for some reason this isn’t possible for NYC’s 7 line extension.

Part of the pressure of building 41 St and 10 Av station immediately is that after the tunnel is completed, doing an infill project for the missing station would require shutting down the entire subway line.

Why did this infill project work so well for BART but would cause so much disruption for NYC’s 7 line?

Additionally, while I cant find a reliable source for this at the moment, the wiki states that the foundations for this station were put in place when the Dublin/Pleasanton extension was built.

I don’t believe that’s the case. The track arrangement there was such that it would be easy to add a station and there was a paved spot in the middle, but I don’t believe they built a complete foundation.

Brilliant illustration of the failure of urban rail. If BART planners added 20 more infill stations, service would slow considerably and discourage ridership. Without OFFLINE stations, BART cannot scale up much beyond its current ridership.

SkyTran solves this problem by putting small stations every 1/2 mile that cost 100x less.

BART and Caltrain are modern day Titanics.

While a BART infill station at 30th Street would certainly be well-located and would likely generate a lot of ridership, it would be mindbogglingly expensive and technically difficult to construct, due to the depth of the BART tunnel at that location an the grade that the tunnel is set at (too steep for a 10-car station).

An Albany (aka Solano Avenue) infill station, on the other hand, would be a downright bargain, considering the potential for ridership increase (both commuters as well as people visiting the popular commercial corridor there). The only issue with an Albany station is that there is nowhere to construct a parking lot, which, in my opinion, is a good thing, as most riders to/from that station would be within easy walking distance, and there’s gobs of parking at North Berkeley and El Cerrito Plaza stations already.

An interesting bit of trivia about the Pleasanton BART extension. I grew up in Pleasanton and while in high school I ran across the original plan for the BART extension (I did a report on it for one of my classes). The original plan was to have BART leave the I-580 median with a station next to Stoneridge Mall. It was then to run south along the east side of I-680 (between the freeway and the drainage canal). It was then going to turn east along Bernal Avenue with a station just past the fairgrounds. It was then supposed to follow the railroad tracks to downtown Livermore. It’s too bad this plan was never built (I suppose it was decided that it was too expensive).

a friend who was raised in Albany claims BART offered a station as part of the original design but the city fathers said no thanks. That said Solano would be highly useful and easy to build. Parking lot not needed, AC Transit bus routes stop under the tracks.

BART can and should IMHO re design the line east of Lake Merritt to create the San Antonio station and provide pocket tracks both for the occasional disabled train and short turns during times when three services are more than needed on that branch. The other improvement for that route in my view starts with expanding Coliseum Station to create higher capacity for events at the stadium complex. At a minimum a second platform and third track so that full length trains can be stored during events to accommodate the crowds leaving.

As to 30th, too bad the engineering is apparently so comple. It certainly is a Muni “feeder” route hotspot, which would relieve pressure on the Mission St surface lines as well as give better service to the crosstown routes there.

]I had always thought the SFO line should have turned off just beyond 24th running next to or under 101 through ‘hoods not well served by Muni.

Good post. After the 08 election, when Obama had his suggestion website up, I submitted something to the effect that DOT/FTA should create a program focused on infill stations rather than throwing everything at extensions and new lines. With low construction and labor costs now seems an opportune time to go back and build stations that were left on the table earlier. Does anyone have a count of how many delayed or deferred stations there are that were proposed and engineered, but left out of final construction?
When I worked at Tri-rail (S.FL) we looked into a few infill opportunities, but at the time we had limited equipment and another stop would have pushed us out beyond the time we needed to make tight turnarounds at Mangonia Park and stay on schedule. I wonder if that’s still the case now that they have additional equipment coming from ARRA?
Now I’m in Seattle area and there are some opportunities for infill stations on North Sounder like Ballard and something at the north end of downtown – Pike Place — Hello! Those would be much more useful if we had through routing and reverse direction trips.
I still think a program focused on infill stations is a good idea for FTA. At $250m/yr with 50% match it could fund 5 of these every year and be comparable to spending on new rail lines. And the operating cost to agencies for new passengers is probably very low because it’s territory you are already running. Has anyone done a comparison? Any grad students need a thesis topic……..

With this Green line extension going north maybe they could extend it south to Back Bay Station, using existing unused tunnels if at all possible.

Some might criticise extensions of Bart and similar regional rapid transit systems on the issue of srpawl but the fact of the matter is that the sprawl is already likely to be there because of cars and roads and extensions of these systems only helps them take advantage of it and hopefully make the situation better. Not extending these systems isn’t going to stop the sprawl one bit.

You make a good point about sprawl … it makes sense to attract more suburbanites to transit.

There can by a lower standard for suburban districts who have a right to some transit for their tax dollars, like eastern Alameda and Contra Costa counties. But thomase projects must still meet a cost per patron metric for farebox recovery that meets a reasonable national standard. They should also integrate well with large transit projects in each Metro Area.

Ted, I looked at the Muni Central Subway maps and agree with you. It makes no sense to build a tunnel exit out to Washington Square, but no station there.

IMO, it would transport more people and ultimately cost less, if Muni extended the subway 3 stations further to Leavenworth-Northpoint in Fisherman’s Wharf, rather than new Geary Blvd and Van Ness BRT lines.

I figure by 2016/17, more Transit funds will become available due gasoline prices raising public demand for more electric rail transit. Thats when SF Muni could next get big Federal Transit funds for the Geary Blvd subway/grade separation to Divisidero then surface on its way to 44th/45th Street terminus.

Then Muni should add LRT line for Van Ness, plus the Muni Metro surface extension from Leavenworth-Northpoint to Fillmore/North Point. Doing BRT on these lines is a fast choice, but ultimately a more expensive choice since they admit plans to convert to LRT down the road for capacity reasons.

Also the subway stations should support at least 3 connected cars= 230 feet.

The SPUR report (see below) is a good backgrounder on some of what’s wrong with the Central Subway. One also has to factor in SFMuni’s history of screwy choices :
1) Boeing Vertol cars (S.F. and Boston wound up being test beds for a marginal design);
2) Breda cars (overweight and overlong);
3) High-floor design rules (possibly due to the adapted upper BART platforms – the MSS was initially planned to be BART over BART instead of today’s SFMuni over BART); and
4) Not putting in the tail tracks at Embarcadero. That’s right – for quite a while the turning around at that terminus was via the cross-over just to the west of the platform. And the switches quickly proved to be a major failure point.

There are two good points to the Metro’s expansions – the J-Church extension added some needed redundancy to the network and the eastern yard along the T-Third (near Cesar Chavez / Army St.) relieved the crowding at the Balboa Park yards. The newer yard also allows for a more balanced schedule.

The thing that makes SFMuni watchers nervous is that SFMuni stays in its rut of making poor choices. What kind of car will they choose to succeed the Breda’s ? Will it fit both the platforms and the tracks (the Breda’s don’t like some of the tight curves and squeal in protest) ? Will any new command + control system be as glitchy as its predecessors ?

P.S. Here’s the track diagram for SFMuni’s portion of the MSS : “Live Snapshots of Muni Subway”

SPUR (S.F. Planning + Urban Research Assoc.) report – “New Connections” (20 June 2007)

Having rode on Muni Metro recently, I agree with you on the high floor/platform height and cabin length issues. I also read the SPUR report.

Though my preference for Geary would have been to schedule it later as a robust subway/dedicated MuniMetro project, I understand that SF Muni had to deliver a near term BRT for the most suffering commuters in SF. Plenty of citizens gave their blessing to the Geary BRT project AND it heads to downtown. So I no longer begrudge Muni on its decision to build a BRT-convertible-to-subway-LRT-later route. Van Ness BRT however, is different since it does not head downtown or deal with the crush loading on buses issue to the scale of Geary Blvd. For only a couple million bucks, Van Ness buses could receive Transit Preferential Street treatment to hold up for another 6-7 years.

Instead of spending$65 million for Van Ness BRT, MuniMetro from Central Subway could surfaced after Washington Square onto Columbus Ave, with a Washington Square Station and a surface Taylor/Columbus Station. Those two stations, only half a mile longer than where unused tracks & tunneling will end, would have pushed the FTA grade the Muni Metro Central Subway project higher due to significantly higher patronage. I venture to guess that for $85 million total, it could also have had a surface station to Leavenworth/North Point in Fisherman’s Wharf. That would have produced an even higher merit patron/cost score — to qualify for additional FTA funding.

I looked Muni’s Vision for Rapid Transit in San Francisco (circa 2004). This Central Subway/North Beach/Fisherman’s Wharf extension project was included. Did the local politics of needing to do “Something for Van Ness corridor now” cancel the extensions to Washington Square, Taylor/Columbus and possibly, Leavenworth/North Point?

By 2017, public demand for rapid transit funding from Congress will be higher. Then, San Francisco could apply for $4 billion/7 years from FTA to fund both Geary BRT conversion to subway/surface LRT and Van Ness surface LRT down to Mission. This would be a reasonable request level from San Francisco (or Los Angeles) since California is a long time tax donor state.

OK, stupid question; I always thought that some of DC’s high patronage was due to huge numbers of Federal workers downtown being encouraged/cajoled/subsidized into taking the Metro to work – is this a wrong impression?

about Van Ness, even BRT lite in the form of a striped off lane and signal preempts would be very useful. Not only Muni, bur Golden Gate buses use the street, and in rush the traffic interference makes the buses very slow.

As to Central Subway et al, at huge cost and disruption the platforms in the MSS could be lowered. I don’t care what the platform height is as long as it affords LEVEL boarding/roll on roll off for wheelchairs and everyone else. If low floor boarding can drive a design with wider bodies above the platforms that would be useful.
As to vendors for the next fleet, Breda needs to be blackballed. That said, the real issue is design, not who does the assembly. Boeing proved that by building good cars for CTA at the same time they built the Standard LRVs.

Heres the fun part: The way I see it, they wouldnt have to do the whole platform at once. They are currently using less than a quarter of the platform for actual boarding. Lowering part of the platform at a time, and building ramps, could lower cost and keep the platform in service while they do it.

It would temporarily slow down boarding in the MSS due to the stairs on the Bredas, but the payoff (proper Siemens or some other company low floor LRTs forevermore, no more ridiculous handicapped boarding platforms on the street portions, just curb extensions) would be so worth it.

Unless the Muni Metro can significantly increase speeds on Geary, it won’t be that much more effective than the 38 Geary. As an example, look at the N-Judah in the Sunset. It take 45-50 minutes to get from Judah&25th Ave to Market&Powell. The 71 bus which follows a similar path isn’t that much slower, about 5-10 minutes). Given that Sunset and Richmond are still in relatively short proximity from the downtown core, these transit times are quite unacceptable, whether by bus or by Muni and would be a waste of taxpayer’s money. The only realistic solution is to go underground and if you’re already paying for a boring machine, you might as well get BART in there so Richmond/Sunset riders have easier access to the East Bay and the Penninsula via a Caltrain transfer. A 15m commute time from the Outer Richmond and Sunset should be a goal for planners.

Not to be a stickler, but San Jose is not a suburb of San Francisco or Oakland. Extending the BART to San Jose would be a way of linking the 3 major cities of the region- not linking a city to it’s suburb.

I am eagerly awaiting the BART connection, but for now I’ll have to take the 181 or the 180 bus from downtown SJ to Freemont and then switch to the BART in order to get to Berkeley

Well, in fairness (and this shouldn’t have to be pointed out as the writer most likely knows this, and if not, shame) it makes sense for New York City and Chicago to look at extending their systems with more tracks as their networks do not have large gaps of missing stations screaming for a station. There may be a place here and there were a station can be squeezed in but that’s not the norm for Chicago and NYC. Actually some citizens of these cities may say the stations are too/close together as the stations can generally be anywhere from 3-8 blocks apart. So at least give your readers the truth and do not try to pull the wool over their faces; Chicago and New York do not have subway stations sitting 4-10 miles apart on their local networks.

Leave a Reply