A SMART Pipe Dream

Please click here to read the previous post, which adds a bit of context and motivation for this post.

Sonoma Marin Area Rail Transit (SMART) will likely be going up for the ballot again this November. Most recently, in the November 2006 election, the Measure R quarter-cent sales tax for SMART failed — falling just short of the required 2/3 of votes in both counties, despite cleaning up with 70% of the votes in Sonoma, a county whose rapid growth calls out for a rail line to channel sustainable development patterns. The plan for SMART is to operate DMUs (possibly light DMUs, use of which would require approval from the FRA) along the old Northwestern Pacific Railroad right of way, a roughly 70-mile, generally single-tracked route with passing sidings, running from Cloverdale in northern Sonoma County to Larkspur in Marin County, serving North Bay downtowns, including Santa Rosa and San Rafael, and within shuttle access of other employment centers. The plan also includes a pathway for cyclists and pedestrians along the right of way. The southern terminus is a quarter-mile from the Larkspur ferry terminal, where riders can transfer from the train to a ferry for San Francisco via a connection that is not nearly as seamless as it should be for a regional connection point. Naysayers — headed by Mike Arnold and his gang, the Marin Citizens for “Effective Transportation” — emphasize that the route is predicted to capture just 5,300 daily riders and hence is not cost-effective as measured on a per passenger basis. (Of course, the original ridership forecasts date from when gasoline cost half what it does now. And prices were still substantially lower than they are now even when the updated forecasts were released, thus not weighing the potentially powerful persuasive effect of $7 per gallon gas.) In the interest of keeping this post to a reasonable length, we will postpone more detail about the actual SMART proposal and its opposing forces for another day, taking it here as a given that the project should be built. This post is meant to be about brainstorming and fun pipe dreams, not politics.

SMART’s supporters have worked to convince voters that even though trains would not directly serve San Francisco, it is nonetheless not a “train to nowhere.” In a very real sense, they are correct. Although there are details to work through in terms of station placement, the proposed SMART route goes exactly where it should, serving what is really the only corridor that will accommodate travel between Marin and Sonoma Counties. Most future SMART riders will not ride the train to transfer to a San Francisco-bound ferry in Larkspur. This may be in part due to the additional fare and time associated with the imperfect mode shift from train to ferry — but an important factor is that most North Bay commuters work in the North Bay, not in San Francisco. In 2000, a little over 75% of Sonoma County home-based work trips remained in the county; for Marin, a smaller percentage but still over half of home-based work trips remained inside the county. Clearly, then, San Francisco is not the predominant travel market, but it is far from inconsequential. There are still a lot of trips between Marin and adjacent counties across the water, and transit is carrying a small share of these trips. Annual average daily traffic at the Marin/San Francisco county line in the middle of the Golden Gate Bridge span was roughly 118,000 in 2007; annual average daily traffic at the Marin/Contra Costa county line, on the Richmond-San Rafael Bridge, was roughly 69,000 in 2007. But on an average weekday in 2007, Golden Gate Transit carried 24,109 trips, merely 8,014 of which were peak-hour transbay trips, and Golden Gate Ferry carried 6,590 trips.

The lower predicted ridership at Larkspur highlights the fact that the ferry + SMART combination will capture few additional transbay trips to San Francisco, and it will not do a great deal to increase transit share in the Golden Gate corridor in the way that BART has for travel between San Francisco and the East Bay. That this should be so is not really that surprising. Even if the connection between SMART and the Larkspur Ferry were made to be as tight as possible, the process of unloading a train out onto a ferry is not instantaneous, and this mode shift will unavoidably lengthen travel times and increase fare cost. (The analogy is not perfect, but imagine a version of BART that terminated in Jack London Square, forcing East Bay riders to hop on a ferry to access San Francisco. Not only would ridership be lower, but such a system would create an inherently lower cap on capacity.) All in all, the current SMART proposal is a worthy project to bring rail transit back to the North Bay, and it is necessary to increase capacity in a commute corridor that promises to be only more heavily used in the future. But the rail line still remains isolated from the Bay Area’s rail network. So we present the following map — very much a pipe dream — which strives to integrate an electrified SMART into the Bay Area regional rail network, while still fulfilling MTC’s vision for building a regional rail link along two “stretch and grow” corridors in San Francisco: Folsom Street and Van Ness Avenue. This branch of the SMART extension is superimposed on the San Francisco subway dream map that was posted a few months ago. Another branch extends across the Richmond-San Rafael Bridge to join the Capitol Corridor in the East Bay.

The map color-coding scheme:

	Capitol Corridor; Caltrain (the Caltrain ROW includes high speed rail).
	SMART
	BART
	Muni Metro Light Rail

The first map depicts a branch along the Richmond-San Rafael Bridge. SMART trains would meet BART trains at Richmond Station, but, what’s even better, would also meet the standard gauge Capitol Corridor, opening up new possibilities for enhanced regional connectivity.

The second map depicts a possible San Francisco branch. South of San Rafael, stations could be added in Corte Madera, Mill Valley, and Sausalito, as shown above. In San Francisco, the line would align along Lombard, Van Ness, and Folsom to serve Civic Center and South of Market, with transfer opportunities to Muni Metro, a future Geary BART subway, and also to Caltrain and high-speed rail at the Transbay Transit Center.

The cost-effective but politically challenging version of the San Francisco extension would be to dedicate two lanes on Highway 101 for exclusive bus use, and then run buses in the dedicated BRT transitway that is currently in planning for Van Ness Avenue. But given the lack of a regional rail link to the North Bay and a goal of connecting more closely to the actual SMART proposal that will hopefully succeed on this November’s ballot, it is tempting to strive for more in the long-term. It is a general rule that rail lines attract more ridership than their BRT alternatives, but it is also interesting to note that even without an existing rail option, North Bay transit ridership trends for the past several years have shown a clear shift away from buses to the one available non-bus option: the ferry. Along with other transit agencies across the Bay Area, the year 2000 was a dot-com bubble peak ridership point for Golden Gate Ferry, with over 6,000 daily riders; this number declined over the next few years, but then increased about 32% by 2007, exceeding the 2000 figure. During the same period in which the ferry experienced a decline followed by a surge in ridership, Golden Gate Transit bus ridership decreased steadily by about 25%. Ridership on GGT’s transbay bus routes followed a downward trend similar to that experienced by the system as a whole. Rising gas costs have helped to reverse this trend just this year — bus ridership has increased, and ferry ridership has decreased, per the District, because of vessels in dry dock. It is likely that this shift is attributable to a greater interest in transit that would, in turn, extend to a rail line. The SMART line as proposed for the ballot, while diverting some bus riders onto the train, would also encourage people to travel more often without an automobile and to take advantage of bus service as a means of filling in gaps in the train schedule, with the ultimate goal of increasing overall transit share. Of course, because the real-life SMART proposal does not connect to San Francisco, no negative impact is expected on Golden Gate Transit’s transbay bus routes. It is almost certainly the case that the pipe dream version of SMART, with a direct rail link to San Francisco, would severely decrease ridership on the GGT transbay lines (though AC Transit is an immediate example suggesting that some transbay bus routes could be maintained). But once again, the overall goal here is not to preserve the currently existing transbay bus routes indefinitely. Rather, the goal is to increase transit share, balanced against the need to not adversely affect the reach of local bus service.

We should note that Marin was part of the original five-county BART district, before both San Mateo and Marin Counties opted out of the plan. The decision to opt out was related to bridge studies carried out at the time, in which it was determined that both the Golden Gate Bridge and the Richmond-San Rafael Bridge — to varying extents, and for a mixture of financial and engineering reasons — could not carry both automobiles and BART trains. So the above maps really are a pipe dream, in the truest sense of the word, and not least because it would require another look-see at the engineering, potential bridge modifications, narrow sections of the NWP right of way, and other issues. This pipe dream configuration would not be a starting point because of the extreme expense, and the San Francisco branch makes the most sense when bundled with one or more other rail services in the same corridor (in the above map, a Muni Metro line that we would not see for a long time). Not only that, but the the 2025 projection of SMART ridership per route mile, while higher than that of ACE, is closer to ACE levels than it is to Caltrain. So this discussion may be premature, but if we never brainstorm and think big, we’ll be boxing ourselves in before we even get started.

The configuration that is a starting point is the actual SMART proposal going on the ballot this November. ABAG forecasts predict that 130,000 new residents and about the same number of new jobs will be added in both counties by the year 2025, which will increase trips within the North Bay and between Sonoma and Marin. Where those new jobs and homes are placed matters. The presence of an active rail line in Sonoma and Marin can act as a catalyst to jumpstart dense growth near the stations, building a natural ridership of those who live and work near the line. Once that happens, a pipe dream extension like the one mapped here could provide the incentive for still more concentrated growth in downtowns along the route. It would bring the isolated, auto-dependent North Bay into the fold of the greater Bay Area, by connecting to the Capitol Corridor and to BART. It would also enhance a rail transit spine that could be complemented by a network of local streetcar routes serving San Rafael, Mill Valley, Tiburon, and other cities in Marin County. The land use point needs to be hammered home, though. We do not want to write off the North Bay as a hopeless case for smart growth, because the Bay Area really cannot afford to write off any of its cities as hopeless — least of all cities lying along a regional rail link. North Bay communities should reconsider the long-term feasibility of staunchly maintaining small-town charm at all costs, and they should seriously question how it is that such a policy has somehow masqueraded as being “pro-environment.” They will have to radically rethink their current land use policies to maximize home and job density within a half-mile to a mile of SMART stations. Given that we cannot simply continue to widen Highway 101 every few years, cities should really be doing this anyway.