The San Francisco Municipal Railway (Muni) has been serving the city for 90 years and today provides over 750,000 trips each weekday. While Muni strives to be efficient and comprehensive, there is a lot of room for improvement. Riders still experience overcrowding, erratic reliability, and long travel times, due in part to the fact that transit competes with automobile traffic for space on the streets. Problems with transit cause more people to choose to drive--a choice that increases congestion, takes up land for parking, pollutes the environment, and more importantly, directs more and more of the city's resources, land and energies to private, single occupant-automobiles at the expense of public transit.
This has spurred Muni to envision what a truly first-class transit system for San Francisco might look like. It would be one that moves riders quickly and efficiently throughout the city with a minimum of waiting. It would be a high capacity, easily accessible, rapid transit-style service. It would enable riders to transfer easily between Muni lines as well as to BART, Caltrain, ferries, and other regional transit services. Muni's aim is to make improvements across a network of corridors providing broad city-wide coverage. The vision includes 12 major corridors, with a minimum level of service, generally described as
• frequent (less than ten minutes waitduring peak hours)
• fast (from any part of the city to downtown in no more than one half-hour)
• comfortable (clean, comfortable, and well-maintained vehicles).
Muni, with input from stakeholders, developed a set of guiding principles, generated a "toolbox" of solutions, and identified the major corridors for improvement. This article applies those principles and tools to the corridors, and suggests a list of projects that could become a rapid transit network for San Francisco. These in turn will be further developed and prioritized as part of Muni's Capital Improvement Program and the Short Range Transit Plan, as well as through the San Francisco County Transportation Authority's (SFCTA) Countywide Transportation Plan.
• Integrate local and regional transit into a seamless transit network. A seamless transit network minimizes transfer wait times and coordinates scheduling with regional transit providers. As jobs become more dispersed around the region, more San Franciscans will work outside the city and require connections to regional transit such as BART and Caltrain.
• Physically separate transit service from automobile traffic on major corridors by creating exclusive rights-of-way (ROW). Traffic congestion is a major source of delay for Muni, resulting in increased travel time and decreased reliability. Transit-only diamond lanes are often blocked by automobiles, and are not effective without constant enforcement. A physically separated right-of-way is more effective at allowing transit to stay on schedule. This means street space must be dedicated to transit use, and not shared with automobiles.
• Provide high-capacity, rapid transit-style service in major corridors. There are a number of major corridors in San Francisco that have high volumes of riders and that suffer from chronic capacity and reliability problems. This justifies a greater investment in these corridors to establish high capacity rapid transit. This could be rail or rubber-tired transit in an exclusive right-of-way, surface or subway, with faster boarding and wider station spacing.
• Upgrade transit service in increments as ridership builds and as funding becomes available. Because funding is limited and proposed projects cannot all be built at once, it is important to take incremental steps so that multiple corridors can be improved simultaneously. For example, building exclusive right-of-way for Bus Rapid Transit (BRT, described below) can be done as a first step, with light rail replacing BRT as more funds become available.
This toolbox describes potential incremental steps for transit improvements. All corridors discussed here need basic improvements, such as Transit Preferential Street treatments described below, but some conditions justify a more robust, higher capacity mode. Costs are expressed as general, per-mile estimates and do not include companion projects such as new storage and maintenance facilities, new vehicles, and ongoing operating and maintenance costs.
Transit Preferential Street (TPS) Treatments
The TPS tool consists of a variety of low-to-medium cost treatments that speed transit vehicle flow. These can also be implemented individually but are more effective taken together.
• Signal Timing for Transit Vehicle Flow
• Signal Priority Systems
• Bus Bulbs (sidewalk extensions at bus stops)
• Boarding Islands
• Transit Lanes
• Exclusive Transit Right-of-Way
• Transit Stop Respacing and Relocation
Cost per mile: $200,000
Conversion to Trolley Coach/Trolley Coach Extension
Electric trolley coaches are rubber-tired coaches that are powered electrically from fixed overhead wires. Trolley coaches now generally operate in regular traffic, but can operate in an exclusive ROW with signal priority, or in a subway. Trolley coaches produce zero emissions and are particularly effective on steep grades. Currently, 34% of Muni's revenue hours are operated by trolley coach. Conversion of more lines to trolley coach operation is desirable because they are quiet, clean vehicles that enhance the quality of life in an urban setting.
Cost per mile: $6.6 Million
Bus Rapid Transit (BRT)
BRT is rubber-tired vehicle operation configured to offer speeds and capacity similar to rail transit, with exclusive travel lanes, limited stops, and signal pre-emption. Other characteristics include the use of low-floor transit vehicles, a prepaid fare system that expedites boarding, and stations that provide shelter and passenger information. Because BRT vehicles are separated from other vehicles and stop less frequently, travel time decreases. BRT is appropriate in corridors with high ridership where there is sufficient ROW to provide dedicated lanes. BRT does not require as much capital infrastructure as surface light rail transit (LRT), and may serve as the first phase of implementing light rail transit.
Cost per mile: $28 Million
Surface Light Rail Transit (LRT)
LRT on surface streets operates most efficiently in exclusive rights-of-way, where traffic is prohibited (possibly by a physical barrier) from traveling in the same lane as the transit vehicle but is allowed to cross at intersections. LRT ideally operates with signal preempts, allowing vehicles to travel relatively unimpeded from station to station. Exclusive rights-of-way may be located along the curb or down the center of the street. The majority of Muni's LRT system runs on the surface, some of it in exclusive ROW, such as the N-Judah on the Embarcadero and parts of Judah Street.
Cost per mile: $59 Million
Subway Light Rail Transit (LRT)
In addition to running on the surface, light rail can operate in subways in congested areas. LRT operation is most efficient in an exclusive right-of-way with no conflicts with other vehicles and pedestrians, where speed is maximized and train control can be automated. This is only possible in a grade-separated right-of-way, such as a subway. Muni currently operates five light rail lines with one- or two-car trains, but capacity can be increased to some extent at marginal cost by adding more cars to each train. LRT subways may also accommodate electric trolley as well as light rail. The cost of building a subway is justified where there is a high density of population, destinations, and traffic, such as downtown.
Cost per mile: $366 Million
This document focuses on physical infrastructure investments, but there are other tools that would help reduce travel time and provide a higher level of service to riders.
• Proof of Payment
• Express, limited, or skip-stop service
• Low-floor vehicles
• Real-time passenger information systems.
The Corridors (as shown on the map above)
This is the first of the Four Corridors (identified in the SFCTA's 1995 plan) to be built. Construction began in 2001, and the first trains will be in revenue service in late 2004. The project will replace much of the current 15-Third motor coach route with surface LRT, operating in exclusive right-of-way except in the Bayview commercial core. Initially, Third Street trains will operate on the Muni Metro Extension on the Embarcadero north of the Caltrain Terminal, and into the Market Street subway. The Third Street LRT project will vastly improve the street with landscaping, lighting, and other urban design treatments.
Phase 2 (Central Subway) is the next major investment for Muni. This phase of the project is partially funded and is scheduled to open in 2009. The Central Subway will take Third Street trains from Fourth and King into a subway through SOMA, under Market Street, to Union Square and Chinatown, serving many destinations and connecting to BART. The cost of the Third Street Light Rail and Central Subway, with a projected daily ridership of over 92,000, is $1.3 billion.
The four Geary bus routes together carry 50,000 riders on an average weekday and taken together, Richmond District ridership accounts for 112,000 rides, nearly 16% of Muni's weekday ridership. Even with buses running every two minutes on Geary, in peak times, capacity is still insufficient.
Geary has priority after the Central Subway. Geary would be surface light rail line in its own ROW from the ocean to Laguna, where there is enough width on the street for an exclusive transit ROW. The LRT would then go into a subway and terminate in the Financial District or at the Transbay Terminal. This new line would also require a new LRT maintenance and storage facility or a major expansion of Metro East (a new facility to be built on part of the Third Street project).
This line would increase reliability by ensuring that auto traffic would not impede transit vehicles, particularly in the most congested downtown portion of the corridor. Capacity would increase and travel time would decrease. Perhaps most importantly, the quality of service for riders would improve.
Subway/Surface LRT from Pacific Ocean to Transbay Terminal: $1.7 Billion
An intermediate phase for Geary would be BRT in a physically separated ROW, with major TPS improvements east of Van Ness. This would include timed signals and proof-of-payment.
BRT from Pacific Ocean to Van Ness and TPS to Transbay Terminal: $346 Million
Van Ness is one of the Four Corridors in the adopted SFCTA plan, and Muni's vision is to have surface LRT in exclusive ROW on Van Ness. There are questions of how this line would fit into the existing route network and how it would connect with other lines and maintenance facilities. If the LRT line were extended into the Mission, the project would have to resolve right-of-way issues along Mission Street, which is not as wide as Van Ness.
Surface LRT from North Point to Daly City BART: $906 million
A rapid bus transitway is in the planning stages for Van Ness Avenue between approximately 12th Street and Lombard Street, and could serve as a BRT precursor to LRT. The transitway could occupy the center of Van Ness Avenue and would separate buses from other traffic using raised medians, landscaping, and boarding platforms. This option would require major TPS improvements along Mission Street.
BRT on Van Ness and TPS on Mission: $435 Million
As a first step, the 47-Van Ness bus should be electrified by extending the overhead wires at both ends of the route, allowing all-electric operation on Van Ness. In addition, the 14-Mission bus should be extended to serve Daly City BART.
Electrification of 47-Van Ness: $25 Million
Market Street is extremely transit-rich--Muni Metro and BART run below grade, and the F-Market streetcar line and numerous bus lines operate on the surface. Congestion on Market results in reliability issues, particularly east of 5th Street. Many pedestrians, bicyclists, delivery vehicles, taxis, and private transit vehicles also compete for space on the street. These delays reverberate throughout the system.
Reducing delays to transit caused by autos can be done by diverting auto trips to other streets, and restoring transit-based signal timing. Capacity increases can be achieved by enhancing rail service in the subway or by adjusting bus routes.
BRT from Castro to the Embarcadero: $95 Million
TPS treatments from Castro to the Embarcadero: $670,000
North Beach and Chinatown are among the city's densest neighborhoods, with an average of 90 dwelling units per acre. Muni envisions extending the Central Subway further north from the planned terminal at Stockton/Clay in Chinatown, through North Beach and Fisherman's Wharf. It could come to the surface and extend into the Marina on a surface alignment via Lombard or Chestnut, with a terminal at the Presidio. The subway could be built to accommodate trolley coaches as well as light rail.
Subway/surface LRT from Stockton/Clay to the Presidio: $554 Million
As a first step, this alignment needs the highest level of TPS treatments, including physically separated ROW for the route numbers 30 and 45 along their entire lengths, signal priority systems, and bus bulbs or boarding islands to speed boarding.
TPS treatments from Stockton/ Market to the Presidio: $113 Million
Fillmore-16th Street, currently served by the 22-Fillmore, is a major crosstown route, carrying almost 25,000 riders each weekday. Ultimately, this could be a surface light rail corridor with connections to the N, J, Market Street, and Third Street lines. There would be an issue with street width, since Fillmore has only one lane in each direction.
Surface LRT from Marina Green to Third Street: $643 Million
In the near term, a BRT-type service could be developed on 16th Street, where there is sufficient street width to accommodate an exclusive lane for buses, along with significant TPS improvements on Fillmore, including signal priority, bus bulbs, and prohibition of left turns at key intersections.
TPS on Fillmore and BRT on 16th Street, including electrification on 16th Street: $88 Million
In the long run, this corridor would be best served by surface light rail in exclusive ROW. The K-line would continue to operate on Ocean and an extension of the Third Street LRT would operate on Geneva with a terminal at Balboa Park BART or Phelan Loop.
Surface LRT from Bayshore/ Sunnydale to Balboa Park BART: $149 Million
An interim step on Geneva would be to implement significant TPS measures such as stop respacing and signalizing intersections. The interim step will also include establishing exclusive ROW for the K-line on Ocean Avenue.
TPS treatments from Junipero Serra to Bayshore/Sunnydale, including exclusive ROW for K-line: $51 Million
19th Avenue-Park Presidio
19th Avenue is the primary north-south artery in the western half of the city, and is appropriate for surface LRT. The alignment would follow the existing Muni 28 route, with a potential extension south to the airport. Exclusive rail ROW already exists between Eucalyptus and Junipero Serra.
Surface LRT from Golden Gate Bridge to Daly City BART: $433 Million
A more immediate improvement for this corridor is BRT with exclusive ROW, possibly extending to SFO. This line could be operated with suburban-style coaches.
BRT from Golden Gate Bridge to Daly City BART: $239 Million
Potrero and San Bruno form a north-south corridor between Mission and Third Street and include a number of residential neighborhoods as well as many commercial and industrial employment clusters. The 9-San Bruno Muni lines combined carry 38,000 riders per weekday. BRT is appropriate for Potrero Avenue but because San Bruno has only one lane in each direction, that part of the corridor requires significant TPS treatments.
BRT on Potrero and TPS on San Bruno: $42 Million
A first phase would be electrification of the 9-San Bruno, since almost half of the route has already been wired.
Electrification of 9-San Bruno: $56 Million
Muni's vision for the Embarcadero is to extend light rail service along the northern edge, from Fisherman's Wharf to Fort Mason and the Presidio, through the Fort Mason tunnel. This could be an extension of the North Beach-Marina light rail line or the F-Market historic streetcar.
LRT from Fisherman's Wharf to Presidio: $111 Million
A short-term project is implementation of the E-line service, which would enable continuous rail service from Caltrain to Fisherman's Wharf along the Embarcadero. In order to operate the E-line, a number of issues must be resolved, such as sufficient maintenance capacity, procurement of an adequate number of historic vehicles, and operating and capital funding. The primary capital cost is for design and construction of terminal improvements on the southern end.
E-line (terminal loop): $11 Million
Muni recently implemented major changes to improve SOMA service in this rapidly developing area. At least nine Muni routes pass through SOMA in all directions, and east-west streets are useful as Market by-pass routes. A clear trunk line for SOMA would improve service for passengers in this corridor. Folsom Street is the logical route, since it is midway between Market and King Streets, and it should allow two-way BRT operation. This would protect transit vehicles from added congestion. The Department of City Planning is also considering the possibility of converting Folsom to a two-way street. Folsom can thus be transformed into a transit and pedestrian street, rather than an auto-dominated street.
BRT from the Embarcadero to 16th Street: $71 Million
Hunters Point (Evans-Innes)
Bayview/Hunters Point is now seeing a large transit investment with the Third Street Light Rail Project. In the coming years, additional need for transit investment is anticipated due to planned development in the Hunters Point Naval Shipyard, which could generate up to 10,000 trips per day.
When the shipyard and neighboring areas are developed, this corridor would be best served by BRT in the Evans/Innes corridor, possibly connecting to Cesar Chavez and the 24th Street BART station or to Civic Center. This would allow connections to the Third Street LRT, the Potrero-San Bruno BRT line, and Mission Street.
BRT from Innes/Donohue to Cesar Chavez/Mission: $130 Million
Depending on the timing of new development, the 19-Polk could become a major trunk line with TPS treatments, rerouted to be more frequent and more direct to the Civic Center area.
TPS on 19-Polk route: $3.6 Million
The 19-Polk could also be converted to electric trolley operation in the short term.
Electrification of 19-Polk: $119 Million
Existing Rail Corridors: Judah, Taraval, Church, Oceanview and Ingleside
These routes have the highest ridership in the system and require special treatment to improve service for passengers and to take advantage of the substantial investment in rail. The majority of existing rail service is on surface streets in mixed flow. Although traffic is not as heavy as it is downtown, many of these routes have all-way stop signs that add running time, as well as turning movements and parking regulations that interfere with transit movement. Rail vehicles are particularly prone to delays due to automobile interference, since they cannot maneuver around obstacles.
All rail corridors should be protected with exclusive ROW and other TPS treatments: boarding islands at all stops, conversion of all-way stops to signalized intersections with priority for transit, and signal priority or grade crossing protection for M-line to cross 19th Avenue at Eucalyptus. Ultimately, the M-line could be grade-separated from St. Francis Circle to 19th and Junipero Serra.
TPS Treatments for:
J (Duboce Portal to 30th Street) $0.5M
K (St. Francis Circle to Green Terminal) $0.5M
L (West Portal to Wawona/46th Avenue) $1.2M
M (St. Francis Circle to Green Terminal) $1.6M
N (Duboce Portal to La Playa) $0.7M
SPUR is very pleased to present this summary of Muni's visionary transit expansion plan. The improvements made at Muni in the last two years give us confidence that Muni can implement this plan professionally and efficiently. The details of each of the corridors still need to be worked out as well as in the rest of the system between the 12 priority corridors. We call on the residents of San Francisco to work with Muni in the spirit of positive cooperation, and to begin to plan on how to pay for it. The price is high, but the cost of not making these needed capital expenditures is much higher.