Long-term Transit Plans: Asking the Real Questions

For several years I worked on a Strategic Public Transport Network Plan for Australia’s national capital, Canberra, so I’m happy to report that the plan has now been released for public comment.
The concise Executive Summary pulls together a number of key ideas about long-term transit planning that I’ve found useful in many cities, so even if you don’t know or care about Canberra you might find it interesting.

Canberra 2031

The most important single idea in the plan (as in much of my long-term planning work) is the Frequent Network, which consists of services that will run every 15 minutes or better all day, every day of the week.  This is the level of service that can motivate people to choose a transit-dependent lifestyle, because it assures you of the ability to get around without building your life around schedules. The proposed Frequent Network includes Rapid service (red lines in this image, stopping at “stations” every 1 km or so, averaging 40 km/hour) and Local service (orange, stopping every 200-400m, averaging 20 km/hour).   (As always, click image to enlarge.)

The plan is intentionally silent on questions of bus vs. rail.  The red lines on this map could eventually be light rail, and the orange lines could even be trams (streetcars).  The crucial point is that long-term planning must proceed without waiting for those decisions.  The point of identifying the Frequent Network over 20 years in advance is so that land use planning and development can start to respond to it. Ultimately, the adopted version of this map should be the authority on future service not just for the transit agency, but also for the land use agency, social service agencies, and ultimately for anyone who makes a decision about where to locate anything.  The long-term message is simple:  If you want really good transit service, locate on the Frequent Network.  If good transit isn’t important to you, locate somewhere else.
In the public consultation period now beginning, we’re asking the public three really hard but really important questions:
  • Where should the Frequent Network be?  We’ve drafted a network based on current and planned patterns of density, and we’re asking if the proposed network is in the right places.  But this question comes with a warning: the Frequent Network a necessary conditions for increased urban density.  If you live in a low-density suburb and you really don’t want more density there, you probably DON’T want the Frequent Network!
  • How much should be spent on service to low-density suburbs where ridership will always be low?  Such service meets a different set of goals based on the severity of needs met, rather than the number of people who need it.  These “social inclusion goals,” such as basic mobility for senior, disabled, and low-income persons who live in these low-density areas, are prominent in the community but expensive to serve given the low ridership those services achieve.  (A professional paper of mine on this crucial topic is here.)  So we need some direction on how much should be spent on those services, because that spending meets social goals but not environmental ones.  (You could also ask this question the opposite way:  “What percentage of our services should be devoted totally to the goal of high ridership?”)
  • How much should service grow overall?  The draft plan increases service by about 50% per citizen over 20 years.  We’re asking whether that sounds like too much or not enough, because ultimately, the subsidy for that growth will need to come from the government, and thus from the public.

What I’m trying to do in constructing these questions is separate technical thinking from value judgments. The three questions above have no technical answer because they are questions about what the community wants and values. As a planner my role is to ask those questions but not express my own values on them, and I recommend that the local transit staff take the same view. However, the questions are technically framed; I’ve constructed them so that when the community answers them through its elected officials, that answer can actually be implemented.

This is an important difference from many of the “visioning” processes that are often used to elicit a community’s desires and hopes. Such processes often seek consensus by steering away from the hardest choices that need to be made, while my goal is to steer right into them.
One of the main reasons that skilled and talented people leave the transit industry is the impossibility of meeting conflicting political demands.   For example, I’ve actually seen cases where elected leaders told a transit agency to cut service on high-ridership route A in order to add new service to low-ridership rural area B, and then complained to staff when that change caused total ridership to go down. Obviously, if you move a service from a high-density area to a rural area, you should expect exactly that result. These hard-but-important questions are designed to elicit direction about the real choices that transit requires us to make, so that talented staff feel supported and encouraged as they follow that direction in building and operating the transit system.
(Frequent Network map by Transit Graphics.)

15 Responses to Long-term Transit Plans: Asking the Real Questions

  1. EngineerScotty July 27, 2009 at 3:44 pm #

    You state that “frequent service” is a necessary condition for increased urban density–and then imply the contrapositive of that position, that increased urban density (now or in the future) is a necessary condition for frequent service. Given that the two go hand-in-hand, that certainly makes sense. How tied in is the transit authority with officials responsible for zoning, or other urban planning functions which may be used in Australia?
    (And as an aside–how do the various parts of Canberra compare to various communities in terms of things like lot density, home ownership/rental, automobile use, and highway development?)

  2. Jarrett at HumanTransit.org July 27, 2009 at 4:04 pm #

    Canberra (technically the Australian Capital Territory) has a unique advantage in being a "city-state" that combines the functions of state and local government, and that unlike other Australian states is small enough to focus entirely on urban needs. As a result, the co-ordination with land use is much easier. While I was working for the transportation people (the Department of Territory and Municipal Services), we had active participation from the land use people (the Planning and Land Authority) and I believe they fully support our statement of the issue. They even worked with us to revise the layout of a greenfield development area to make it easier for transit to serve efficiently.  

  3. Jarrett at HumanTransit.org July 27, 2009 at 4:14 pm #

    One correction to your comment, EngineerScotty. The contrapositive of the statement “Frequent service is a necessary condition for increased urban density” (i.e. Increased Density -> Frequent Service) is “The absence of frequent service implies no increase in urban density.” (i.e. NOT Frequent Service -> NOT Increased Density). I do indeed want people to make that logical step, because those statements are logically equivalent.
    The statement “increased urban density is a necessary condition for frequent service” (i.e. Frequent Service -> Increased Density) is the converse rather than the contrapositive of the first statement, and logically it’s a separate statement. It is usually but not always true. (Sometimes the existing density is sufficient, and sometimes Frequent Service crosses an undevelopable gap.)

  4. Peter Parker July 28, 2009 at 3:19 am #

    Jarrett, while you mentioned ‘city state’, I’d be interested to hear your comments about the ‘overflow Canberra’ – ie that portion that is in NSW (eg Jerrabomberra, Queyanbeyan etc) and the potential to undermine planning ability (airport land is another example).
    Although the ACT itself seems to have done a good job of this themselves, though development at various office parks, Fyshwick, Symonston etc rather than in the town centres. You could include Mitchell in that but it’s at least between 2 town centres, so could be reasonably served by transit.
    Some of the frequent transit corridors don’t look that direct, and I notice the absence of a fast direct route from Barton to Woden, which could be useful instead of backtracking towards Civic). In the mid-1990s this was an express bus (peaks only) running every 10 min (Route 233) before the 1996 cuts. Or is this covered by the REDEX routes, with the route via Curtin considered good enough outside this time? Similar comments might also apply to New Parliament House, preferably with stops out the front.
    Overall I like the FS network, as the current network treats ‘group centres’ no better than local routes. FS for all group centres eg Kippax, Cooleman Ct, Dickson, Macquarie, Southland plus Kingston/Manuka, Fyswick is a big step forward. You don’t want routes from every centre to every other, but one to two others either side as per the map is good.
    The big question though is will it be funded?

  5. Jarrett at HumanTransit.org July 28, 2009 at 3:32 am #

    It's a 22-year timeframe, and it's roughly a 50% per-capita increase in operating cost.  We'll find out, in the consultation, if that seems reasonable as an implementation of the ACT's sustainable transport goals.

    Re Queanbeyan, cross-border issues are eternally vexing and I have no magic bullet in that regard.  
    The Woden-Barton market is definitely a gap.  Our view is that Barton demand will always be very peaked, so Peak Express service on this link is adequate.  But note the Frequent service via Hindmarsh Drive connecting most South Canberra suburbs direct to Woden.

  6. Max Headway July 28, 2009 at 10:44 am #

    “Re Queanbeyan, cross-border issues are eternally vexing and I have no magic bullet in that regard.”
    If Skåne, Sweden can co-operate with Denmark, there’s no reason why a state/territorial boundary has to be a hindrance.

  7. EngineerScotty July 28, 2009 at 1:01 pm #

    There are many instances of different political cultures being found in different political regions in close proximity. Many Portland planners love to bash Vancouver, WA, as a sprawl-infested cancer on Portland’s backside, dominated by right-wing and anti-transit politics, and would love nothing more than to toll the Columbia River bridges in order to teach the road-hogs a lesson. Likewise, many residents of The Couv are equally dismissive of the politics on the Oregon side of the Columbia River, and resent Portland intrusion into Vancouver politics, as well as the notion that Vancouver is little more than a Portland suburb. (It’s not; while many Vancouverites cross into Oregon for employment; the Washington side of the river has lots of its own industry, infrastructure, and culture).
    Obviously, both of the above caricatures are mostly inaccurate; but regional cooperation across the Columbia is more difficult. Vancouver’s current transit plan calls for BRT lines between downtown Vancouver and various surrounding communities in Washington; they expect MAX to eventually cross the river into downtown Vancouver but if you want to go further than that, you have to transfer to a C-Tran bus. The Washington side has no urban growth boundary; and housing (on larger lots) is generally cheaper. (Whether this is due to the lack of an UGB or to other factors is an optn question).

  8. Daniel July 28, 2009 at 8:22 pm #

    Something else to ponder about providing high-frequency services into low-density suburbs: if you don’t, how many of the people who live in those low-density suburbs will continue to drive every day through the high-density suburbs, adding to heavy traffic congestion?
    An example from Melbourne: I’d speculate that most of the people on the roads in the inner-suburban City Of Yarra probably don’t live in Yarra — where most residents have reasonable access to a frequent network of PT services (though not necessarily serving every trip).

  9. Jarrett at HumanTransit.org July 28, 2009 at 8:57 pm #

    Daniel. The answer to your hypothetical is “most of them will keep driving” because the fundamental mathematical fact about low density is that average walking distances to a transit stop are high while local traffic congestion is minimal. If you live on a quarter-acre lot in a suburb made of quarter-acre lots, driving is the logical thing to do, at least until you get out onto a major road where you encounter congestion. At that point, Park-and-Ride is your best option, assuming there’s such a facility.
    Locally circulating transit in low-density neighborhoods is always low-ridership.

  10. Jason October 25, 2009 at 4:32 am #

    I like the separation of value judgments and technical questions.
    But I wonder at how to interpret the answers. Who did you ask? Did you expressly convey that these are value judgments? and, in the case of disagreement, does the majority rule or is a compromise forged?
    And, lastly, how do you create the political culture to provide the best chance of actual implementation? In deomcracies, politicians love making 20 year plans. The risk is that Old mate Stanhope gets voted out (god forbid) and the next mob fail to implement, then have a crisis summit to ‘rewrite the failed policies of our pinko predecessors.’

  11. Jarrett at HumanTransit.org October 25, 2009 at 4:37 am #

    Ultimately, the value judgments are for the elected officials to make.  You're right about the challenge of long-range plans in short-range democracy.  Growing up in Portland, where there really is a broad central urbanist consensus, may have made me too idealistic.  But we try. 

  12. EngineerScotty October 25, 2009 at 3:39 pm #

    Even here in PDX, the consensus isn’t universal. In the city proper, transit and urbanism has broad political support, but in many of the suburbs, such things are viewed more skeptically–with the standard criticisms being offered: too downtown-centric, too left-wing, too expensive, insufficiently useful. And in many Portland burbs, especially those unincorporated parts of the Metro area, one finds acres of auto-centered design which, if/when served by transit, is served inefficiently at best. Given that Tri-Met includes these locales in its tax base and service district, it has to consider the opinions of the residents there–many who live in the burbs precisely because they DON’T like urban lifestyles.
    And I haven’t even gotten to Vancouver yet.

  13. Nathanael November 3, 2009 at 12:27 pm #

    ‘Daniel. The answer to your hypothetical is “most of them will keep driving”…..
    at least until you get out onto a major road where you encounter congestion. At that point, Park-and-Ride is your best option, assuming there’s such a facility.’
    Dial-a-ride has been surprisingly successful in Denver at bringing people to — or, importantly, *from* — train (light rail) stations, and may be another alternative.
    The master plan for “frequent corridors” would probably be assisted by marking possible ‘collector zones’ where dial-a-ride or park-and-ride should be used to funnel people from the low-density areas into the corridors. Did you do that?

  14. Nathanael November 3, 2009 at 3:20 pm #

    The Kingston hub appears to have direct service to very few destinations in this plan….
    but it also appears to be the main intercity rail station!
    Does Canberra need its intercity rail line extended to a more central station, or does it need better service to that station?

  15. Jarrett at HumanTransit.org November 3, 2009 at 7:39 pm #

    Canberra has 2-3 slow trains a day from Sydney, nothing more, so that Kingston station is not that important. There’s actually discussion of moving it further out from the city. In the long run, any high speed rail project from Sydney would not come to that station anyway, because the most logical HSR corridor approaches from a different direction.