Guest Opinion: The Future of Los Angeles is Bus Rapid Transit

sbx
Omnitrans’ sbX in San Bernardino is the first on-street Bus Rapid Transit (BRT) system in Southern California to feature dedicated on-street bus lanes and rail-like stations. Full-feature on-street BRT represents a key opportunity for transit expansion in Los Angeles County. Photo: Omnitrans

Los Angeles is finally on its way toward realizing the dream of a regional rapid transit system. Five rail lines are simultaneously under construction, and there is renewed momentum to fund another round of transit expansion on the 2016 ballot. Move L.A. recently unveiled a Strawman Proposal for “Measure R2” to accelerate the completion of the remaining Measure R projects and offer a new vision for transit, highway, and complete streets improvements across Los Angeles County.

Move LA's Measure R2 "Strawman Proposal" features a number of possible rail expansions, but does not identify specific bus and BRT improvements.
Move LA’s Measure R2 “Strawman Proposal” features a number of possible rail expansions, but does not identify specific bus and BRT improvements. Source: Move LA

For Angelenos and transit nerds everywhere, there is a lot to get excited about. The centerpiece of Move LA’s vision is a $27 billion expansion of Los Angeles’ rail network (right, and also mapped below). Other features of note include $9 billion toward a “Grand Boulevards” program for complete streets improvements on the region’s automobile-oriented thoroughfares, and $3.6 billion toward active transportation projects. Although Move LA’s vision is just an early draft, a measure along these lines could transform the region—on par with the development of the expansive freeway network half a century ago.

Nevertheless, there’s something missing.

Move LA’s Measure R2 proposal does not effectively articulate one of the most critical ingredients to reshaping mobility in Los Angeles County: a spectrum of bus improvements, including bus rapid transit (BRT), to enhance transit service throughout the region.

Los Angeles already has many features of a great transit metropolis, but its greatest challenge is one of geometry: even after another $27 billion rail investment, only a handful of cities, neighborhoods, and corridors will have convenient rail access. For most Angelenos, including many in densely-populated, growing, or transit-dependent areas, buses will continue to serve as the only accessible mode of transit. Rather than rehashing bus vs. rail debates, Los Angeles must embrace upgrades to its bus system (the nation’s second-busiest) in tandem with rail expansion to reach a level of transit abundance that brings frequent, quality service to as many people as possible.

A spectrum of bus improvements are necessary. In many locations, bus stop upgrades to provide adequate shelters, security, and real-time arrival information may be sufficient when combined with frequent service. For other locations, BRT—dedicated lanes and more robust rail-like infrastructure—is necessary to provide quality service and room for growth. Yet, details on bus improvements in Move LA’s Measure R2 proposal are thin: the proportion of funds allocated to transit operations remains constant, and bus enhancements are mentioned only briefly under the Grand Boulevards program.

The lack of a comprehensive regional BRT vision in Move LA’s proposal is indicative of the region’s cautious approach to reallocating street space for buses and other users. While Metro has implemented two (mostly) off-street BRT lines—the Orange and Silver Lines—and an extensive Rapid network, the on-street implementation of BRT has been limited. A handful of “peak” hour bus lanes (7-9am and 4-7pm) have been implemented on Wilshire, Sunset, and Figueroa, and similar treatments have been recommended on nine additional corridors in Metro’s Countywide Bus Rapid Transit and Street Design study. However, Metro has currently no plans for more comprehensive bus improvements, such as all-day dedicated bus lanes and rail-like stations.

The city of Los Angeles is effectively leading the charge for bus improvements and more advanced BRT features as it develops concepts for a Transit-Enhanced Network, but the city lacks funds to implement these improvements without its own citywide ballot measure. The city is also is tied to a problematic on-street advertising contract which has limited its bus stop amenities.

A step-by-step approach to BRT implementation makes sense to deliver quick benefits to riders, but it risks setting the bar too low and degrading the benefits of BRT. What Metro presently brands as BRT offers only slight improvements over Rapid service: for example, bus lanes on Wilshire are only active for five hours per day and will be absent in Westwood and Beverly Hills, which opted out. Even after full implementation Metro’s countywide BRT plan, none of the designated corridors would meet the “Basic BRT” standard set by ITDP or come close to being on-par with Metro’s rail facilities.

More than three quarters of Metro’s ridership is on buses, and many more people choose not to ride because the service and amenities are inferior to other alternatives: shelters, safety features, real-time arrival information, and way-finding elements are often lacking, even at the busiest stops. Frequency, speed, and reliability can be all-day issues given the ever-present threat of traffic congestion.

More robust bus improvements are necessary. These improvements not only benefit existing riders, they also makes transit a more useful mobility option for millions of people.

Vermont
This Rapid/Local stop on Vermont at Manchester in South LA serves approximately 1,300 people per weekday (more than many light rail stations) but it lacks basic facilities such as shelters and is blanketed with trash. Source: Google Streetview
xxxx
Santa Monica’s transit provider, Big Blue Bus, is in the process of upgrading its most heavily used stops to include shelters, seating, real-time arrival information, and way-finding features. Source: Big Blue Bus

What could a fully developed BRT network look like?

Metro currently operates 400 miles of Rapid service, while other local providers have their own. In total, roughly 500 miles of Rapid services are potential candidates for improvements. 500 miles of true BRT is likely to be cost prohibitive, but it may be conceivable to imagine a mix of improvements such as:

  • 100 miles of “comprehensive” BRT (and complete streets enhancements) that could qualify under the ITDP standards (around $4 billion at $40 million/mile – assuming the cost of Metro’s East San Fernando Valley BRT project)
  • 200 miles of BRT-lite “Select Bus” service featuring dedicated lanes and targeted bus stop and streetscape improvements (about $2 billion at $10 million/mile – assuming double the cost of Metro’s Wilshire BRT project)
  • 200 miles of signalization, bus stop upgrades, and minor street improvements (roughly $500 million at $2.5 million/mile).

All together, a fully-upgraded Rapid/BRT network could cost in the ballpark of $6-7 billion—almost the total budget for the Grand Boulevards program, for which may feature additional non-BRT projects as well. This order-of-magnitude estimate doesn’t factor in improvements to the local bus network or associated increases in operating cost to maintain appropriate all-day frequencies for a core transit service.

Despite the cost, a robust BRT investment appears doable within the framework of Move LA’s proposal combined with a mix of local, regional, state, and perhaps federal funds.

For abstract illustrative purposes only, consider the conceptual rail network from Move LA’s Strawman Proposal overlaid with a countywide BRT concept, which assumes upgrades to most existing Rapid lines plus the addition of new BRT services.

xxxx
Move LA’s rail vision in their Measure R2 Strawman Proposal. Map by author
xxxxx
Move LA’s rail vision overlaid with a sketch of potential BRT lines in red. Map by author

This map represents a very rough idea of what an enhanced Rapid/BRT network could look like. The key takeaway is that investing in BRT across Los Angeles County could triple the size of the region’s rapid transit dream for a relatively affordable cost.

These corridor-level investments, when combined with complete streets improvements and transit-oriented zoning reforms, provide the framework for Los Angeles to become a true transit metropolis. Moreover, by addressing the regional geometry challenge, such an investment could achieve greater regional equity: many more people could have access to quality rapid transit service with investments in both rail and buses, compared to just rail. Ideally, rail and bus services would converge to the point that riding the bus would be nearly as pleasant, dignified, and efficient as riding the train, yet buses will serve more places and more trips.

Transit is crucial to the future of Los Angeles: the region is dependent upon expanding mobility options to unlock new opportunities for growth while achieving key environmental, health, and equity goals. Rail expansion will play a key role in reshaping Los Angeles, but a comprehensive investment in transit must extend beyond rail.Los Angeles needs to embrace bus improvements and BRT as core elements of its 21st Century transit vision to foster abundant, high quality transit service for all.

Daniel Jacobson is a transportation consultant working with cities and transit agencies to improve transit, bicycle, and pedestrian mobility in California.  He is a transportation planner at URS and a car-free resident of Los Angeles. (The views expressed in this article are his own.)

  • Westernite

    I’m sorry, but BRT will just not work on some of the streets shown on the map. I know for sure it will not work on Western. I have lived along Western in central LA for over a decade and take both local 207 and rapid 757 down it on a regular basis. The required amount of space for bus lanes, whether temporary, like on wilshire, or permanent, does not exist on western. it is two lanes in each direction for most of its run through central LA and due to the narrow sidewalks and high density of the area it runs through, there is no space to widen the street at all. That means the best we can do is a rapid bus that runs in traffic, which we already have, and this takes nearly six minutes to go a mile during rush hour, when there is highest demand for transit (I timed one of my trips). I happen to pass over the 101 when I take the bus during rush hour, and while traffic on it is at a near-standstill, I know they move faster than 6 miles per hour. The only solution is a subway to escape the traffic.
    In my opinion, the majority of rail development should happen in areas that are lacking adequate public transit service currently, ie areas where buses move at 6 miles per hour, the central city. Some areas about to get rail service, such as the san gabriel valley, already have buses that move almost as fast as trains.

  • Dennis_Hindman

    Most of the advantages that some people perceive that light-rail or streetcars have over buses is narrowing.

    One example of this is the APTS Phileas 60 bus shown on page 24 of this pdf:

    http://www.nbrti.org/docs/pdf/2006_brt_compendium.pdf

    This 60-ft bus has a fully independent suspension that gives it a smoother ride.

    It has a passenger capacity that is 50% more than the Orange Line buses.

    It weighs 12,000 pounds less than the Orange Line buses, which enables it to carry 48 more passengers and yet still not obtain the curb weight of the Orange Line buses.

    This bus also has all-wheel steering which enables it to turn in a smaller radius than the Orange Line buses.

    It also has a electronic guidance system that utilizes magnetic sensors in the roadway. This produces smoother acceleration (no-jerk) and can get the bus close enough to the station that there is level boarding–with no need for a ramp.

    More information:

    http://www.apts-phileas.com/

    http://www.slideshare.net/amoplant/presentation-phileas

    Phileas buses cost about 2.5 times more than the Orange Line buses.

  • This is assuming that the average trip length spans the entire corridor. I wouldn’t be surprised to find that most trips, whether by bus or private car, aren’t actually that long. At that point, even if a BRT bus is traveling at just a 30 km/h average, it still offers a competitive service both over a local routes and to drivers who would otherwise be in traffic then searching for parking.

  • Actually it is. Come out to San Bernardino and try it.

  • BRT has been a massive success for the auto and road-building industries everwhere it’s been done — LA should be no different.

  • Kenny Easwaran

    Is there a reason that all of Western needs to have lanes that cars are allowed on? Can’t cars use Arlington or Normandie for long distances, and any of the dozens of smaller streets for short distances?

ALSO ON STREETSBLOG

Guest Editorial: Dreaming Big About Rail Lines, Grand Boulevards, Bus Rapid Transit and Measure R2

|
(Move LA’s mission is to build a broad constituency that will advocate for the development of a comprehensive, diverse, robust, clean and financially sound public transportation system for Los Angeles County. Denny Zane is the executive director. Gloria Ohland is the policy and communications director.) More and more people — from elected officials to bike […]
STREETSBLOG USA

Transit Vote 2016: Raleigh’s Chance to Grow Smarter

|
We continue our overview of what’s at stake in the big transit ballot initiatives this November with a look at Wake County, North Carolina. Previous installments in this series examined Indianapolis, Seattle, Detroit, and Atlanta. Ask Wake County Commissioner Sig Hutchinson how Raleigh’s transit system is currently functioning, and he doesn’t sugarcoat it. “I just really don’t think we’ve got […]

What a Vermont Avenue BRT Line Could Look Like

|
At this month’s board meeting, Metro staff reported that they are hiring consultants to shepherd two Bus Rapid Transit (BRT) projects. Today, SBLA previews one of those: Vermont Avenue BRT. For the uninitiated, what is BRT? Bus Rapid Transit is high-quality bus service running in its own dedicated right-of-way. It comes in a lot of flavors, […]
STREETSBLOG USA

Transit Vote 2016: Atlanta May Finally Expand MARTA and Beef Up Bus Service

|
We continue our overview of what’s at stake in the big transit ballot initiatives this November with a look at Atlanta. Previous installments in this series examined Indianapolis, Seattle, and Detroit. Back in the 1970s, both Atlanta and Washington, D.C., received federal grants to build rail networks. After finishing the first wave of Metro construction, D.C. continued to invest, […]