This page has older content

Please see Related Projects on this page for current project information. We are keeping this page as a record of SFMTA outreach.

A new side-running design for the Geary Boulevard Improvement Project

Share this:
Route(s) Affected

At the start of the COVID-19 pandemic, the SFMTA paused work on the Geary Boulevard Improvement Project to pursue emergency transit lanes. The 38 Geary Temporary Emergency Transit Lanes Project installed side-running transit lanes, located in the curbside travel lane next to the parking lane. Based on the positive results from evaluation of those lanes, and the success of the Geary Rapid Project delivery on the eastern half of the Geary corridor, the SFMTA is now pursuing a revised side-running project design, pending outreach and approvals.

The Geary Boulevard Improvement Project had envisioned transit lanes in the center of the street (or “center-running”) between Arguello and 28th Avenue. Evaluation of the side-running emergency transit lanes indicates that they have improved transit performance with minimal impacts to car traffic on Geary Boulevard and few vehicles are diverting onto parallel streets. Additionally, a majority of survey respondents indicated they preferred to maintain the emergency transit lanes permanently. Because of these positive evaluation results and in response to evolving conditions, the SFMTA is now considering pursuing side-running transit lanes instead of center-running transit lanes in the Geary Boulevard Improvement Project limits. The SFMTA plans to seek input on this change as a part of outreach activities anticipated in mid/late 2021.

Why this change?

Accelerates transit and safety benefits
A side-running project can provide tangible improvements in stages, allowing more benefits to transit and safety to begin sooner. 38 Geary/38R Geary Rapid buses are already benefiting from transit lanes throughout portions of the Geary corridor in the Richmond District where emergency transit lanes and other improvements were installed. Additional transit lanes, bus stop modifications and safety improvements could be implemented over the course of a few weeks in 2022, while capital transit and safety improvements like bus bulbs and pedestrian bulbs would follow a couple years later. With a center-running project, transit lanes cannot be operational until the project completes final design, awards construction contracts, and completes roadway reconstruction, a process that can take several years.

Limits construction disruption
A side-running project avoids the need to reconstruct and relocate curbs and medians along the entire project limits and focuses construction at only spot locations where items like bus or pedestrian bulbs would be installed. One unknown at this point is whether the SF Public Utilities Commission will want to join the project to coordinate any State of Good Repair utility upgrades at the same time as transit and safety improvements, which would contribute to a longer construction duration. If so, construction would be similar to the Geary Rapid Project, which has an approximate 2.5-year construction duration including water and sewer scope, and has remained on-time and on-budget with limited disruption to the adjacent business districts affected.

Preserves the most transit operational flexibility and preserves local stops
One of the downsides of the center-running design was that it required consolidation of Rapid and local service because the median transitway would not have bus passing lanes. Rapid buses would stop more frequently than they do today, reducing travel time savings potential; and local buses would stop less frequently than they do today, which reduces options for people making short trips and is less accessible for people with mobility impairments. With a side-running alternative, both Rapid and local service can remain.

Improves cost effectiveness of transit travel time and reliability benefits
According to the analysis conducted as a part of the environmental review stage of the project, center-running and side-running alternatives performed similarly in transit travel time savings and reliability improvement. The center-running alternative is anticipated to cost at least $235 million, while a side-running alternative is expected to cost less than $50 million. As a point of comparison, the Geary Rapid Project cost is $36 million for transportation improvements.

Avoids center median tree removal
The center-running alternative would require removal of approximately 60 trees in the median of Geary Boulevard, and new trees would be replanted after construction.  Little to no tree removal is anticipated for a side-running project.


  1. How does this relate to plans under discussion for a Geary-19th Avenue subway?
    The ConnectSF Transit Investment Strategy calls for investment in a Geary-19th Avenue subway to provide fast, high-capacity service and improve regional access, but would take more than a decade to deliver at minimum. A side-running project would complement such an investment by providing transit benefits much sooner for a relatively low cost. In addition, the alignment and stations of a Geary-19th Avenue subway have not yet been determined, but are not expected to serve the entire length of the Geary corridor. Therefore, Geary bus service will continue to have importance in connecting the western Richmond District to rail.
  2. Does the side-running project decrease transit benefits?
    Not really. While a center-running median better protects transit from cars, it also requires additional stops and prevents buses from passing one another. Additionally, the design called for two locations where buses would need to transition from side running to center and back again to side; the time it takes a bus to make those transitions detracts from the overall amount of transit benefits that would be provided. Finally, the side-running project provides much of the benefits years sooner than would be possible with the center-running design.
  3. How would this affect parking?
    With both the side-running and center-running design, some parking removal would be necessary. This would include converting blocks where there is angled parking to parallel parking to provide enough room for one transit lane and two general purpose travel lanes per direction. On a typical Geary block face, converting angled parking to parallel parking would reduce parking spaces by 1-2 spaces per block face. There would be more parking removal required with a side-running design than a center-running design because a side-running design requires maintaining bus zones on the side of the street, while existing bus zones could be used for more parking with a center-running design. During outreach, the project team will survey merchants on the corridor to learn their vehicle loading and access needs and develop a curb space management plan to best accommodate their needs.
  4. What are the next steps?
    We plan to conduct outreach in mid/late 2021 to seek input on this change in direction, as well as collect feedback to inform the detailed design of the project. Subscribe for updates to stay informed.