Transit’s Role in “Sprawl Repair”

Duany Plater-Zyberk, one of the leading planning firms associated with New Urbanism, is thinking about “sprawl repair,” a process by which utterly car-dependent landscapes could be transformed into something more walkable, and thus more resilient.  Galina Tachieva of DPZ has an article explaining the concept at Planetizen. 

The only valid option is to repair sprawl – to deal with it straight on, by finding ways to reuse and reorganize as much of it as possible into complete, livable, robust communities. Pragmatism calls for the repair of sprawl through redevelopment that creates viable human settlements, places that are walkable, with mixed uses and transportation options.

As examples, Tachieva suggests turning a shopping mall into a complete mixed use town center, or dropping townhouses into the spaces between McMansions.

The ideas she presents are exclusively about the development parcel, not the street.  Clearly, though, such urbanized places will have dramatically different transport needs, which will also require re-thinking the standard suburban arterial.  Since government has much more direct control over the street than over the development parcels, we might move faster on sprawl repair if we focused on the arterial first, or at least at the same time.  Last month, on a speaking tour that took me to Portland, Fresno, and Los Angeles, I spent some time thinking about this issue.

The densified sprawl that DPZ envisions will clearly be a place that needs much more effective transit.  Parking will be more scarce and density will be higher, so there will need to be more transport options.  While cycling will take up some of that need, transit is still the most versatile sustainable transport mode for longer trips, the only one capable of delivering you from one place to another as a pedestrian, unencumbered by a vehicle.  And the only place to put that transit, in a classic sprawl landscape, is on that huge, government-controlled arterial.


As I first discussed here, the ideal geography for transit is to have a series of transit destinations (dense housing, mixed use, jobs, activities, etc) located in what feels like a straight line, so that a transit line can connect all these points in a way that feels like a direct route between any two points.  The key challenge is to avoid slowing down or deviating too much to serve the intermediate points, as this makes the intermediate points feel like obstacles to anyone whose trip requires riding through them.

The classic suburban commercial arterial already does that, up to a point.  It gathers development around it without letting that development slow it down.  So we should be looking for ways to capture this virtue for transit.  A process of “sprawl repair” should start, perhaps, with a series of steps aimed at repairing the arterial, starting with the realisation that the arterial’s straightness and speed become an asset if transit can operate reliably.

Fresno is an interesting example because it’s at a stage where it could look at what’s happened to bigger sprawl-cities and choose another path.  There’s not much congestion yet, partly because the city has a truly massive road network.  For a metro area of about 1m people, Fresno has three north-south freeways, one east-west freeway, and a complete grid of very wide arterials spaced about 800m apart (the ideal spacing for transit).  There is no shortage of road space for introducing new approaches to transport, and as anyone involved in struggles over bus lanes in big cities will tell you, the time to create transit priority, bike lanes, and other reassignments of road space is before the street is badly congested.

If and when California High Speed Rail opens, Fresno is going to boom.  With the Bay Area and Los Angeles suddenly less than two hours away, Fresno will be a perfect base for any kind of business that aims for a statewide market.  It will attract good employees who value affordable housing and outdoor recreation.  (Yosemite and Kings Canyon National Parks are both within an hour’s drive.)   If HSR were there today, and I were considering setting up a consultancy aimed at a statewide market, I’d consider living there myself.

What if the city prepared for that now by planning for the inevitable future of its abundant wide arterials?  What if Fresno’s leaders went to the people and said:  “Look, HSR is coming, and our city is going to grow and change, mostly for the better.  We can resist the change or we can get out in front of it.”

Well, you wouldn’t start by proposing to build townhouses in people’s back yards.  You’d start by saying: “Look, even if we boom horizontally, tearing up orchards to build more subdivisions, we’ll still become more and more congested; just look at Sacramento or Las Vegas.  Eventually, we’d end up like Los Angeles, struggling with congestion that chokes the economy and constrains people’s lives.  And then, we’d be desperate to get people on transit, as Los Angeles is now.  But by then, we’ll have filled up our arterials with congestion, so the remaining options for building transit will be really expensive.”

“What if we learned from Los Angeles’s path?  Instead of waiting until it’s really expensive, as Los Angeles did, what if we take early, gradual, inexpensive steps to make our arterials safe and attractive for transit?  That doesn’t mean ripping up our single-family neighborhoods, but it does mean rethinking our arterials so that they’re safe and attractive places for pedestrians and provide appropriate levels of priority to transit.  We don’t need transit to be attractive to everyone, we’re not ‘forcing’ people to use it, but it could attract people who already want alternatives to driving.  Let’s face it, a lot of our citizens are struggling on low incomes and cars are expensive.  Many families would experience sudden improvements in wealth if they could get rid of one or more of their cars.”

“So we need to gradually repair our sprawl.  That doesn’t have to mean big increases in density.  We’d build some denser centers for people who want a more urban life, but we’re not going to build townhouses in your back yard — or at least not until you and your neighbors want us to.  Mostly, we just need to stitch things together so that people can walk and cycle more safely, both to complete local trips and to get to transit stops.  It means making sure that at every transit stop there’s a protected way to cross the street; you can’t use transit for a round trip unless you can use stops on both sides of the street.  It means adding pedestrian links to cul-de-sac neighborhoods, so that they are through-routes for bicycles and pedestrians while remaining cul-de-sacs for cars.  It means making sure that the design of bus stops and transit priority conveys a clear message that transit riders are valued as citizens and appreciated for the contribution they make to a sustainable and functional city.”

Such a pitch would need to be aimed at the practical concerns of life in a car-dependent city.  It will sound compromised and wimpy to a big-city transit advocate, but that’s not who it’s talking to.

As part of my presentation for Fresno Council of Governments, I picked an arterial intersection pretty much at random and quickly pointed out all the obvious, inexpensive opportunities for sprawl repair — not just repair of the land use, but of the streets.  Here’s the intersection of Kings Canyon Road and Peach Avenue.

Kc peach

There’s a lot to work with there.  There are already quite a few apartments, but they tend to open out onto arterials far from places where you can cross those arterials safely.  But that could be fixed over time.  With the existing development in mind, you’d identify permanent bus stop locations each of which must have a safe street crossing opportunity.  That could mean new pedestrian crossings, signalised as necessary, which could also be combined with other accesses.

Look, for example, at the southwest corner of the image.  South of the arterial there’s a patch of apartments with several possible points of pedestrian access to the arterial.  Across the street to the north is a WalMart behind a huge parking lot.   When WalMart redevelops, of course, you’d want the building brought out to the street with parking behind it, but for now I’m thinking shorter term.

So I’d start by observing that if I signalize the WalMart entrance at the far west edge of the image, I’ve also signalzed a pedestrian access point to the apartments on the other side of the street.  I’m about 400m west of the main arterial intersection, so if there’s a bus stop at that intersection (as there must be, because it’s a connection point) then I can put another bus stop right here.  So a signal there will have three uses: (a) controlling the car access to WalMart for safer egress, (2) providing safe pedestrian access to the apartments, and (3) providing the safe pedestrian crossing that the bus stops require.

Once you notice this, you notice that, in general, Fresno is a grid of arterials spaced 800m apart.  That means that for local service, where 400m spacing is about right, you’ll have stops at the arterial intersections and then one more stop midway in between them.  Although 400m stop spacing is ideal, we can slide this intermediate stop 100m one way or the other to find the best site for a pedestrian crossing, based on the development and access points that are already there.  We can then use this crossing as a focal point when redevelopment opportunities arise on the surrounding parcels.  Over time, we’ll build a more connected network for walking, cycling, and transit, without seriously obstructing road traffic.

So some principles of arterial “repair” would include:

  • Put transit stops at an ideal spacing, 400m for locals, and make them permanent.
  • Require a safe way to cross the arterial at every bus stop, as a necessary condition for a stop, because you can’t use a stop to make a round trip on transit unless you can cross the street there.
  • Look for a way to meet other pedestrian access needs with the same signals or crossing protections that the previous point requires.

I could also add more subjective values, such as …

  • Ensure that sidewalks along the arterial are adequate, including appropriate buffering if the arterial is fast.

… but we should pause before we go too far in this direction.  If we add too many design requirements that are derived from urbanist ideals, we can quickly lose focus on the reality of what the street is today, and the level of improvement that’s needed to achieve basic safety and functionality for alternative modes.  We also risk spending so much money in one place that we can’t scale our improvements to the vastness of the sprawl that needs repairing.  So when I look at a typical street intersection in typical sprawl fabric like Fresno’s, I see first of all a need to create a basic pedestrian+transit infrastructure that will provide a safe and functional transit option for getting around the city.

Once, long ago, when I was presenting a transit plan to an outer-suburban council, a council-member said: “So if we spend all this money on transit, is that going to make me leave my BMW at home?”

The answer, from a “sprawl repair” standpoint, should be:  “No, and it doesn’t need to.  Most people don’t have BMWs, and many of them struggle to afford a car at all.  These improvements are focused on people who are near a decision point where transit could be a rational choice for them.  That’s why our priority is basic safety, functionality, and civility.  We’ll incorporate quality design wherever we can do so affordably.  But our first focus is on improvements that will scale to the size of the problem.  And there’s a whole lot of sprawl to repair.”

29 Responses to Transit’s Role in “Sprawl Repair”

  1. John November 15, 2010 at 7:52 am #

    To me the major obstacle seems to be getting a low-density, low congestion, sprawling municipality to fund a grid system of relatively frequent transit service that is not needed now, but will be someday in order to repair the sprawl. There’s a time period of at least several years, probably a couple of decades when they will be spending a lot of money on mostly empty buses. Maybe with that kind of capacity, the low-demand routes should be free? Could that help sell the tax increase that would probably be necessary to fund it?

  2. Danny November 15, 2010 at 8:38 am #

    Maybe it is time to try the Market Urbanism solution: drop minimum parking regulations, and drop Euclidian zoning restrictions.
    With these solutions, they can be branded as deregulation, which is a very politically popular idea among those in the conservative valley right now. And they definitely can have a redevelopment effect that brings more density and more demand for transit…which means developers are more likely to be transit friendly in areas where there is access to transit.
    And if you could find a way to tax land value rather than property value, even better.

  3. Tom West November 15, 2010 at 8:39 am #

    My local area isn’t that sprawly – the housineg is generally detached, but close together (like 2-3m between houses), so there’s no way of increasing residentital density through in-fill. However, the commerical nodes are few and far between (literally), which leads to car dependence (and huge parking lots).
    So, how do we get more commerical/employment nodes? They don’t have to be huge, but they would have to replace existign housing.

  4. EngineerScotty November 15, 2010 at 8:53 am #

    One problem, in the Portland context at least–many of our most notorious “sprawlevards”, and we have quite a few–are also state highways and under the jurisdiction of the Oregon Department of Transportation, and not the city or county. And ODOT generally insists that its highways, even ones which run at surface-grade through well developed urban fabric, and which aren’t limited-access in the least, be optimized for through traffic and through traffic only. (One of my biggest concerns about Powell BRT is precisely this–Powell Boulevard is US26, and an ODOT facilitiy–something which will likely constrain the design).
    When you have a highway agency to deal with, the organizational tendency for it will be to try and REMOVE access, not add to it. Were one of the aforementioned Fresno streets maintained by CalTrans; my suspicion is that CalTrans would want to do the exact opposite of what you suggest: Eliminate left turns at any place other than the major intersections; eliminate pedestrian crossings altogether (unless grade-separated), and discourage busses from stopping at places where there isn’t already a signal.
    A “good” thing about many sprawl cities is that there are locally-maintained boulevards which can be repaired in this fashion without interference from the highway authority. Which is also, of course, the bad thing as well.

  5. Eric November 15, 2010 at 10:19 am #

    Jarrett, I second your cautionary note there on an overly prescribed path to applying “sprawl repair”. I think though getting the examples out there does require careful thought into new guidelines and standards. To get an interagency level of coordination going you have to inevitably bat around standards and best practices for single facility types.
    This March in Savannah, the ITE (in collaboration with CNU) released a recommended practices manual to look carefully at thoroughfare repair: Designing Walkable Urban Thoroughfares: A Context Sensitive Approach (it can be downloaded here:
    A step in the right direction and a lot of good thought went into it, even if some of the outcomes and ‘case studies’ seem a bit beyond reach of your typical municipality’s ability to implement them wholesale.
    They did leave a lot of language in the document that subordinated one size fits all approaches to ‘design’ in order to lead the practicioner to start first with the context and network. The street sections and graphics, however, seemed to tell another story. There’s always this interesting paradox in products that are the result of New Urbanists collaborating with engineers. You know they are going to talk a lot about “context” and yet nonetheless they end up with prescription pills.
    It was somewhat ironic that this manual was released in Savannah, that, with its meager sidewalks and simple amenities, starts with simple premises that add up to remarkable wholistic advantages. It seems DWUT has started from the opposite vantage point. The most most interesting design for me tends to begin with simple premises.

  6. Tobias November 15, 2010 at 11:48 am #

    Sprawl repair reminds me of the ill-fated attempts of remodelling European cities to fit car transport in the 1960s and 70s. Today we consider this policy a failure. Not that public transport is generally better than individual transport, but that any transport solution has to fit the urban environment and not the other way around. Even if that means to leave American suburbs as car-dependent as they are and offer just a minimum of PT there.
    If you want to create or re-create dense urban developments then do it in inner-cities or areas close to that. Suburbs 30 km off the metropolitan centre are more likely to be abandoned once energy and therefore car transport becomes unaffordable than turned into a PT-oriented areas.

  7. Jarrett November 15, 2010 at 1:00 pm #

    @John. For the record, Fresno’s buses are already very crowded! Productivity on most lines is over 40 boardings/hour, which is very impressive for a city that size with so little transit infrastructure.

  8. Jarrett November 15, 2010 at 1:06 pm #

    @Tobias. No government is going to pursue a policy that presumes that the landscape that a majority of people perceive as home, and in which they’ve invested their money, is going to be worthless anytime soon. My point in this post is that there has to be an approach to sprawl that works with the needs and values of the people who live there, and that doesn’t give the impression that we’re trying to make the whole country look like Berkeley. See also Urbanophile on “Starbucks Urbanism”:

  9. CroMagnon November 15, 2010 at 1:49 pm #

    I think some of the retrofit approaches meshes with my philosophy: do a few things, but do them well.
    Moreover, so much talk about “walkability” seems replete with all sorts of development plans and streetscaping. In reality, I’d say about 90% of walkability is having sufficient sidewalks and streets one can cross!

  10. hallam.jon November 15, 2010 at 5:13 pm #

    Using the figures from and some UK data for pedestrian crossing cost, an 800m stretch requiring two crossings would cost ~128,000, while creating two 4’6″ sidewalks would cost @288,000. So adding the sidewalks too pretty much trebles the cost of the project.
    PS: I’m not sure I believe that the best layout for a transit accessible city is a line. I think you can access more destinations in given time with a grid (and transfers between horizontal and vertical straight-line transit-lines), than with a single straight-line transit line. However, transit has a bigger disadvantage over point-to-point transit (e.g. walking) in a grid, because of the wait-time-cost of the transfer.

  11. JJJ November 15, 2010 at 7:05 pm #

    Interesting that you chose that intersection. Bus 28, which is the busiest bus line in Fresno passes through there. (50 passengers per hour, per bus)
    It is also the proposed street for a BRT system.
    20mb PDF:
    But the mayor said we can’t afford $8 million. And then she turned around and approved $55 million to “improve” the area where the 3 freeways meet (wider ramps). And as far as I could tell, she did not apply for TIGER grants for BRT.
    To truly repair sprawl you first need people in power who get it.
    And looking at the housing and retail projects under construction or planned, they don’t get it. They follow the guidelines from the 1984 plan.
    There is one man who does get it. There is an Indian gentleman who has been building lofts and townhouses downtown, and they’ve been very successful.
    And it’s not just locally where change is needed. The feds are funding a huge freeway extension project into prime farmland. Why did they say yes to this massive waste?

  12. JJJ November 15, 2010 at 7:07 pm #

    The satellite image is of a streetcar suburb, close to downtown. When the whole city is a suburb, what do you do?

  13. Alon Levy November 15, 2010 at 7:08 pm #

    The sidewalks are important. Equally important but underrated is making sure the signalized crossings are spaced close together, on the order of 200 meters or less, with a signal optimization that makes life easy for pedestrians and not just drivers. If the street is wide, some parts of the world, such as Israel, get tempted to time the signals with 3-4 parts per cycle, prioritizing conflict-free auto travel and green waves to the point that the pedestrian lights on the two sides of the median are not synchronized. This should not be done, and if it is, it should be replaced with the simpler two-cycle crossings common in Manhattan.

  14. EngineerScotty November 15, 2010 at 7:23 pm #

    Moreover, so much talk about “walkability” seems replete with all sorts of development plans and streetscaping. In reality, I’d say about 90% of walkability is having sufficient sidewalks and streets one can cross!
    And if nothing else, also improving pedestrian access from sidewalks along streets, across parking lots, to the front doors of buildings. Quite a few big boxes are difficult for pedestrians to access from the street–some having fending or shrubbery surrounding the premises with the only breaks in this barrier being driveways.

  15. Colin November 15, 2010 at 9:52 pm #

    Jarrett, I applaud your push on this issue. On the 80/20 pareto principle:
    – vast areas of sprawl represent the big opportunity for reform in our cities
    – the quantum of car parking in retro-fit development is under public control. Allowing bonus plot ratio commercial development floorspace on exisitng frontage parking lots would encourage developers to replace asphalt in their “front yards” with dense floor areas and jobs easily served by the frontage arterial transit system. Research has shown that a lack of parking at the workplace is correlated with high transit use.
    – the location of schools and government offices etc are under public control. Siting of new schools etc along arterials will encourage transit use and encourage the community to soften the arterial street environment with speed limits and safe crossings etc.

  16. Wad November 16, 2010 at 10:18 pm #

    @JJJ, a whole city by definition cannot be a suburb.
    The city can have a land use of predominantly single-family residences, but that doesn’t mean its a suburb. Architecture alone doesn’t make a suburb.
    Break the word “suburb” down. It’s a sub-urban area.
    The sub- prefix means subordinate or later-stage. This is what a suburb is.
    Suburbs have existed since the earliest days of human organization.
    All cities originate from a nucleus and move outward. The earliest civilizations all had suburbs, with the outward growth determined by the limitations of nature (hills or bodies of water) and transportation.
    Historically, cities were compact because of circumstance, not choice. Most of human civilization was defined by a compact city because its boundaries were limited by walking and draft animal distances.
    The next stage came during the railroad era. One general feature of pre-railroad cities was that human settlements tended to cluster around bodies of water.
    Landlocked cities were rare, and generally economically passive. Railroads enabled cities to grow without surface water for trade.
    The automobile allowed almost any surface to be developed and made even more space accessible.
    In each of these instances, though, the suburbs were lower-value land uses a nearby urban area transforms into a higher-value land use.

  17. Ncbarnard November 16, 2010 at 11:24 pm #

    @CroMagnon Fully agreed that sidewalks are a major step toward making areas walkable. I’ve gone hiking around many of these car centric retail areas, because thats what you do, hike through the bushes, the inadequate uneven grass strip on the side of the road etc.
    The other thing that I think is Jarett really keyed in upon on improving walkability at least initially is mid block lights. The problem I see as a pedestrian in a automobile oriented area at stop lights is that drivers are less likely to look out for you when making right or left turns, than they are in an urban area. They just don’t think about pedestrians.

  18. R. W. Rynerson November 17, 2010 at 9:13 am #

    Englewood, Colorado converted what was once the Denver region’s largest mall into a mixed-use development. It has enough good points that it draws a substantial amount of shopping out of adjacent Denver, which has nothing like it. Rents in the development are higher than for similar amounts of space in other older areas outside of Downtown Denver.
    It can be viewed by typing “Englewood Station, CO” on Google Earth. Note that it still has lots of surface parking, but with a street grid that will permit denser development incrementally, as the market warrants. Also note the sprawl development west of US85, which requires driving to the nearest transit stop. We offer something for everyone in Colorado!

  19. JJJ November 18, 2010 at 1:07 am #

    Wad, Fresno is indeed a giant suburb. The urban portion of the city, downtown, was mostly demolished or abandoned, leaving only the suburban area.
    Los Angeles indland empire, for example, is made up of various suburban cities. They don’t have downtowns, and they don’t consider downtown LA relevant.

  20. Alexis November 18, 2010 at 7:14 am #

    This is a fascinating post, but the question that came up for me is whether after the modifications the street still has the benefit of being fast that you mentioned. It may be narrower due to sidewalk additions, possible bike lanes, increased traffic due to density. Or would the effect not be significant?

  21. John Bailo November 18, 2010 at 12:39 pm #

    It’s funny that you talk about making a mall “like a town center” because I was just in Southcenter mall in Tukwila, WA, having lunch in the food court and people watching and I thought — this is the Village Square! Yes, despite all the haranguing from “urbists” you couldn’t design a better loved and better used place than this mall. People congregate where their is shopping and nice food stuffs — and low costs! The mall is clean, and in an environment where it rains all the time — dry. And they walk. I have to laugh when you all talk about “walkable cities”. People do tons of walking in this mall … and many malls all over suburbia.
    The thing is…walking has its scale. I don’t think that every square foot has to be walkable — some of it can be driveable or bikeable — optimized for vehicles. And these vehicles can take us to the walkable places — be it a mall, or a country trail with parking lot at the trailhead.

  22. Nicholas Barnard November 18, 2010 at 1:42 pm #

    @Alexis the net effect would be there, however I think that’s fine. Given that all people pay for roads everyone should have fair use of it.
    (An aside, but the worst thing that happened for car centric proponents is the decoupling of use fees via gas taxes and for roads being paid for from
    general funds. An argument that bike riders should pay a tax to use the roads made more sense 30 years ago when roads were paid for much more via use taxes than it does today.)

  23. EngineerScotty November 18, 2010 at 10:01 pm #

    A common development model one finds in Hong Kong is apartment buildings essentially on top of shopping malls (on top of subway stops and the like). The residential areas are segregated from the shopping, but you literally can take the elevator down to the mall concourse.

  24. Jarrett at November 19, 2010 at 4:54 pm #

    @Alexis.  Arterial speeds are sometime reduced due to sprawl repair, but not by dramatic amounts.  For example, new signals can usually be timed to adjacent signals to maintain a reasonable flow of traffic.  On the other hand, sprawl repair does introduce visual features that signal motorists to take greater care, which will tend to reduce speeding.  Classic sprawl arterials practically encourage speeding.

  25. calwatch November 19, 2010 at 8:45 pm #

    The Inland Empire decidedly does have downtowns, or “town centers”, in each of the incorporated cities therein. Some of them, like Redlands, Riverside, or Ontario, are actually quite nice. Auto sprawl filled in between these small town downtown, but the original points were already established thanks to the railroad.

  26. Matt Korner January 16, 2011 at 4:55 pm #

    I strongly disagree with the thinking that transit shouldn’t be targeting BMW drivers.
    The best transit and T.O.D. projects appeal to people with a wide range of educations and incomes, and this diversity is especially necessary to a place like Fresno, which suffers from being a monoculture.

  27. Matt Korner January 16, 2011 at 5:15 pm #

    Yes, the Inland Empire has much of its pre-war planning still intact. JJJ must have never been to the region.
    Before the domination of the freeways and the rise of air travel, the 200-year-old San Bernardino was the historic core of the region, and the city held parity with Los Angeles and San Diego. The Pacific Electric Red Cars had two nexuses in the city centers of Los Angeles and San Bernardino, respectively. Additionally, a regional rail system, the Kite-Shaped Track of the Santa Fe Railroad (predecessor to Metrolink), positioned its center point in San Bernardino.
    The city is currently in the process of re-establishing itself as southern California’s third urban core with a slew of impressive new transit projects (B.R.T., light rail, commuter rail, aerial gondolas, etc.) that the city is building North, South, East, and West. These are being designed to converge at the optional station location for California High-Speed Rail in the city center. A huge transit village is also planned there, and the design is, in my opinion, the best in the entire C.H.S.R. system.
    More important is the way San Bernardino is incorporating 56 T.O.D.’s within its borders while linking surrounding cities featuring their own T.O.D.’s and downtowns. So, the Inland Empire will soon be a model of transit-oriented planning.

  28. Rick June 1, 2011 at 4:09 pm #

    Has there been any research done into this sort of “sprawl repair” for cities that aren’t exactly growing the way Fresno is? I’m thinking of Cleveland and Detroit — two cities that have become much less dense and less populated, and may fit the description of uncongested and sprawly, with large arterials. It seems the Rust Belt cities have lots of transit problems too (have you ever seen how much of downtown Cleveland is surface parking lots?), and fixing those may help improve a bad situation.

  29. Jarrett at June 1, 2011 at 8:52 pm #

    Rick. The tool is useful anywhere, though depopulating cities will
    need to think about what parts they most want to "save."