It's all green lights for the Central Subway to move forward, and this is a good thing: one of America's densest transit corridors will get a significant investment. Yet while a worthwhile project, certain changes to the existing plan could at once improve the performance of the subway while lowering the costs (and thus feasibility) of future transit enhancements.

The Central Subway is a transit project funded by 1989's Proposition K which secured funding for transit improvements in four heavily-used corridors. Primarily an extension of the T Third Street, the Central Subway as currently planned will run on the surface from the Caltrain Depot before dipping underground to Moscone Center, Union Square, and Chinatown. The project is as expensive as it is critical and the purpose of this article is not to halt the project. Rather, the aim here is to support the project in reaching its maximum utility.

Where are the plans for the North Beach Extension?

Under the current plan, the tunnel boring machine (TBM) will finish it's work at the Chinatown station before proceeding north to the excavation site at Columbus Avenue and Union. Muni plans on eventually extending light rail surface to North Beach, but if Muni were to do so as part of the Central Subway project, it could do so at a much lower cost.

Finishing a subway tunnel and closing up the whole left by extracting the TBM is costly and disruptive; re-opening the tunnel and creating a surface portal at a later date doubles this disruption and adds to the costs greatly. Restarting the project to create a North Beach connection could prove so costly and disruptive that obtaining sufficient political support for such a project could be difficult--if not impossible. The longer San Francisco waits to build, the more expensive and less likely the project becomes.

SPUR--San Francisco Planning and Urban Research Association--has advocated for including a surface portion and North Beach connection in its policy paper on the Central Subway. SPUR argues that building such an extension would greatly leverage the high cost of the subway, getting more transit for the buck. In addition, a North Beach connection could reduce demand for Stockton Street bus service (and thus lower operating expenses) and provide a great opportunity to improve pedestrian and bicycle access in the area.

The major hurdle is Federal Transit Administration funding. Such an extension is considered a "construction variant" rather than plans for a future extension of the subway. As SPUR writes, "If the Central Subway is phase 2 of the T-third line, then the extension to North Beach should be considered phase 3." Should phases 2 and 3 happen to be built together, all the better for San Francisco's taxpayers and transit riders.

Why not opt for shallower & more accessible tunnels?

The current plans call for the tunnel to plunge underneath Powell Station's mezzanine level, and then under both the Muni and BART stations. At its lowest, the subway will reach down nearly 100 feet underground. The alternative is to build a shallow tunnel (at least until Chinatown), employing a cut-and-cover method. Under this process, the street is torn out and then redecked. The building of the tunnel takes place under the road deck. Finally, the deck is removed and a new road sits above a new subway.

For now, however, Muni has favored the deep-tunneling approach. This poses two significant problems for Muni.

First, deep tunneling is not necessarily the most cost-efficient option. Muni has not given the proposal much public attention and thus no cost-benefit analysis, but one can assume that there is more material and equipment (walls, escalators, floors, lights, etc.) needed in building a deeper tunnel and station than a shallow one. At least more study is needed.

Second, a deeper subway station is less accessible. With the Union Square/Market Street station slated to be built under Powell BART, raiders will have to ride several escalators or walk many flights of stairs to simply reach the platform. Using cut-and-cover, riders would simply go down one escalator, directly reaching the platform. Shallow stations would be at the same level as Powell's mezzanine level, allowing for simple transfers.

This second point is particularly important given the existing service the subway will support. Transit between Market Street and Chinatown is currently dense and slow. Though geographically close, the grid of one-way streets forces 60-foot buses to zig-zag their way along the route. What currently takes up to 15 minutes by bus will take less than 5 via subway. But if the current plan goes unchanged, riders will also be switching from convenience street-level service to walking down several flights of stairs to reach their station, then back up on the other end. It is not enough to replace slow bus rides with time-consuming walks to the stations’ platforms.
Muni's rational for not using cut-and-cover is that the disruption to street traffic--particularly on Market Street--would be too great. I disagree: several months of disrupted street traffic seems a better option than a lifetime of ineffective transit connections.

(Credit where credit's due: many of these points are also made in SPUR's report on this issue).

How is future Geary Street light rail being optimized?

No doubt about it, light rail is the ultimate plan for Geary Street. The current plans for bus rapid transit are being developed as "rail-ready," meaning that any plans will at least not prevent future rail service and at best enhance this eventual development. Let's ignore for the moment that Muni is making a major mistake in not building Geary light rail service in conjunction with the Central Subway (Muni is building bus rapid transit now, and then redoing Geary Street again later for light rail).

What is pertinent now is that the Central Subway's alignment does not optimize future Geary Transit. Geary's hypothetical light rail service would dip into a subway at Laguna then travel downtown. Where it joins with the rest of Muni metro system downtown, however, is far less clear.

A previous alignment for the Central Subway had the Union Square/Market Street Station situated on O'Farrell & Market Streets. Trains to Chinatown would take the right turn to the north, while Geary trains would keep straight heading west. Other plans have shown the Geary subway jointing the existing subway under Market street--very costly and disruptive proposition.

But the currently alignment of the Union Square station--aligned north-south along Stockton Street--allows for neither of these plans. To incorporate a Geary subway under the Central Subway's current plans, either a new station would be built around Union Street (the third subway station within three blocks, a very expensive and inefficient proposition) or the Geary subway would take a turn south into the planned Union Square station.

The problem with this final possibility is that all Central Subway stations are only built to handle one two-car train per direction. First, keep in mind that Geary is one of the busiest transit corridors on the west coast, carrying over 50,000 passengers a day. Secondly, Geary buses currently come every few minutes. Third, if Geary trains turn into the Central Subway, Union Square becomes the only downtown station (with many passengers transferring to Powell Station to go further downtown). How can this passenger load be added to a short-platform stations already carring riders though the core of the Central Subway?

How Muni responds to these questions will not only impact the effectiveness of the Central Subway. Ensuring that San Francisco leverages the very high cost of the Central Subway and procures every possible benefit, the project will go a long way in building the voting public's trust in funding future public works projects.

It would be wrong to not reinforce the most important action Muni can take (as said by SPUR): "The single most important way to make this project the most cost-effective expansion of the Muni Metro system it can be is to begin construction as soon as possible. "

Let's go!

Peter Lauterborn may be contacted at