 |
| Bus station, Marion, Ohio. Ben Shahn/FSA, 1938. |
I finally got around to reading SEPTA's recent report on redesigning Philadelphia's bus network, and I'm very glad I did. The Philadelphia Bus Network Choices Report was introduced to the world in June of this year. If you'd like to read the report, click here.
When it came out, Network Choices received insightful coverage from Jason Laughlin of the Inquirer (click here) and Jim Saksa of PlanPhilly (click here).
The report was prepared by consultants Jarrett Walker + Associates. This well-known firm also consulted on the redesign of Houston's bus network, which was rolled out in 2015.
What struck me most about the Walker report was that, at its base, it was a short course in geometry, as applied to transportation in cities.
The Geometry of Cars
Since the arrival of cars a little over a century ago (the Ford Model T was introduced in 1908), there has been a strong tendency to concentrate on cars for all of our transportation needs. This approach has had support from the car industry and its allies. Unfortunately, when it comes to cities, the push for an essentially monomodal transportation system based on the private automobile runs up against the laws of geometry. (See Network Choices, page 9.)
Cars are big. It takes a lot of space for cars to transport people. This is a problem, because the purpose of cities is to bring people together, and the effect of cars is to push them apart.
We can see this effect at work in Center City Philadelphia, where the Vine Street Expressway is essentially a moat dividing William Penn's original city from points north. Market Street and JFK Boulevard are too wide for pedestrians to be comfortable, and so the interesting stores tend to show up on the side streets and Walnut and Chestnut to the south, which are human scale.
Cars are also very convenient. They can get you anywhere there's a road, something that trains on tracks and buses on fixed routes do not do. There are valid reasons why people have loved cars since they were invented, and it is a simple fact that cars are not going away.
However, it seems fair to suggest that we are currently over-reliant on cars. Unlike some other parts of the country, Philadelphia has a functioning multimodal transportation system, with a variety of rail transit and a large bus network; many parts of the city are very walkable, and certain sections are heavily bicycled, despite the utter inadequacy of the city's network of bike lanes. Still, you only need to look at a street like Chestnut in Center City to see that there are too many cars, both moving and parked, and they are taking up too much space.
How do we cut down on the number of cars being used every day in this city? People tend to focus on commuters, many of whom either come from or are going to places that are inaccessible by anything but a car. Fortunately, there is a silver lining. Many of the car trips in Philadelphia are actually quite short - under two miles. And in many cases these trips could be done easily, and possibly more pleasantly, by walking, biking, or taking the bus.
Unfortunately, all of these modes have been neglected, and if you want them to fulfill their potential, they all need to be upgraded. When it comes to buses, this means we need to rethink the network.
The Geometry of a Bus Network
The older parts of Philadelphia were laid out on a grid, and the bus network in those areas generally follows the grid. In the newer parts of town, where street layouts often feel very suburban, the bus network is more radial.
The Walker firm really likes the grid, because the grid dovetails with the firm's basic design approach of short, straight routes, high frequency, and good connections. Free transfers are crucial to the success of this approach, which involves a lot of hopping on and off buses, but because of the high frequency it does not involve a lot of waiting time.
Let's have a look at South Philly. This is basically grid heaven, but that does not mean there are no improvement opportunities. It turns out that the east-west routes have comparatively wide spacing (they're generally about half a mile apart), while the north-south routes are spaced about a quarter mile apart (Network Choices, p. 73).
The north-south routes are too close together and basically try to steal passengers from one another. Who wins? The most frequent routes. Poor little Route 2, for instance, is stuck between the much more frequent Broad Street Line (aka the subway) to the east, and, to the west, Route 17. As the report puts it, "The pattern of boarding activity and route productivity is clearly showing that many people prefer to walk to higher frequency service." (P. 73.)
Route 47M
South Philly's overcrowding of north-south streets reaches its apogee with route 47M on 9th Street. To quote the report, "This is a highly specialized route that primarily serves the Italian Market on 9th Street. Until 1993, SEPTA operated all Route 47 buses via 9th Street, but the busy Italian Market meant that many buses were stuck in traffic during their trip through South Philadelphia. In May 1993, northbound Route 47 was shifted to 7th Street to improve speed and reliability. But the 47M was also added to maintain service on 9th Street.
"Route 47M operates northbound only and is effectively using one bus from the northbound 47 every 20 minutes to run through the Italian Market on 9th Street. This is done so that people going to the market do not have to walk two blocks from 7th Street to 9th Street.
"This kind of response to specialized demands takes frequency from the most productive north-south route to reduce walking distance for a few people." And guess what. Hardly anybody gets off the 47M at the Italian Market. "Most people riding Route 47M get off at Market Street." (P. 84.)
I can hear the howls of protest already, as the keepers of South Philly's many "peculiar institutions" (such as parking in the crosswalk) rise up in agitated opposition. But if you're robbing Route 47 of significant frequency so that people don't need to walk two blocks to the Italian Market, and then virtually nobody is getting off at the Italian Market, perhaps sweet reason will win out in the end. Who knows. One insight the Walker report offers, which I happen to agree with, is that "the point of transit is to provide an alternative to driving, not an alternative to walking." (P. 93.)
Chestnut Street
In addition to looking at the big-picture geometry of routes, the Walker people also spent time going through their tool-box of hyperlocal solutions to specific problems. They clearly took a close look at Chestnut Street in Center City, a transit nightmare that I've been writing about for a while, and they've come up with a bunch of deft, often very minor, adjustments that I think could have very positive effects. (See page 46.)
Mayor Kenney has already started the ball rolling on Chestnut Street. In September he introduced a package of steps to increase enforcement of existing traffic regulations on the most heavily congested parts of Chestnut, which lie just east and west of Broad. And this program has had some effect.
However, the increased enforcement is temporary, and even now it's not solving the whole problem. Random people parking in the bus lane has declined substantially, but Fed Ex, UPS, and the post office still stop when and where they choose. And of course Uber and Lyft stop wherever their next passenger is standing.
I've suggested before that a big part of the solution lies in the third lane on Chestnut Street - the parking lane. Much of this lane is given over to to loading zones in the early morning, but at 10 a.m. those spots generally revert to being regular two-hour parking spots. Not coincidentally, the phenomenon of delivery trucks unloading in the bus lane explodes right after 10 a.m. It seems only logical to extend the time of the loading zones further into the day. I do think this simple and incremental move would help a lot.
That still leaves the problem of drivers filling the bus lane, which is on the right side of the street, while they try to turn right and go south on a numbered street. The right turn from the bus lane was never an ideal design, and I actually recall that when this first went in, the right turn from Chestnut onto 17th was prohibited. There was a big sign indicating No Right Turn. Basically, everybody ignored it, and after a while the sign quietly went away.
Drivers do queue regularly to make the turn on 15th and 17th, which I have observed personally. A lot. I'm told this also happens elsewhere. The issue is twofold. First, pedestrians walking along on the south side of Chestnut are often crossing 15th and 17th in great numbers, blocking the drivers who want to turn right. Second, cars headed south on 15th and 17th frequently back up into the intersection and even beyond, to the north. What can I say? It's a busy area.
The basic problem for a bus driver is that the bus stop is at the corner, just before the intersection. If turning cars back up at all, the bus can't get to the stop. The Walker folks suggest a number of changes that would likely make things, not perfect, but I think much better.
First, what would happen if we just moved the stop to the far side of the intersection? Then the bus driver could swing out into the left lane, drive around the queue, and stop at the stop.
Alternatively, we could simply eliminate the bus stop at intersections where cars turn right - in other words, space the bus stops every other block, placing them at intersections where there is only a left-hand turn.
But, but, you splutter. Take a deep breath. Stops in Philly are much closer together than they are in other comparable cities. The typical Philadelphia block is about 450 feet long, so stopping every other block would give you 900 feet between stops, putting Philly on a par with Boston (p. 93).
And often a bus rider would not have to walk any further than at present. Take the William Penn House on the 1900 block of Chestnut. It's located in the midblock, so a resident currently has a choice of two stops half a block from the apartment building's front entrance on Chestnut Street. Removing one of those stops would still leave you with the same half-block walk to a bus stop.
Implementation: All at Once or Bit by Bit?
So, should we redesign the network over a couple of years, and then implement the changes all at once? Or should we put the improvements in gradually, as they become available?
I'd do both. My thought on the 47M is that eliminating it is pretty much a no-brainer, but I've seen this community in action. I'd wait on the 47M and roll it up into a much bigger ball with things like Route 2, just west of Broad. This would encourage people to take a broader view.
On the other hand I'd do Chestnut Street now, especially since it's already started and may fail if we don't keep going. I'd even do it in several slices - expand the hours of the loading zones in the parking lane, and then later deal with the bus stops and the right-turning traffic.
The problems on Chestnut Street are widely recognized in the community, which doesn't mean that the changes will lack opposition. However, Chestnut Street is low-hanging fruit, and it might be possible to convince a City Council member or two that the results are almost certain to be very positive and the blowback transitory.
Also, it wouldn't hurt to get some successful pilot projects out there fairly early. They could show people that these changes actually do work, right here in Philadelphia, and possibly increase public confidence in the overall project.
 |
| Bus station, Marion, Ohio. Ben Shahn/FSA, 1938. |
See also Unblocking the Bus Lane on Chestnut, Taming Chestnut Street, Cars and Bikes - the Back Story, About that Parking Lot in South Philly.
About the photographer. Ben Shahn is best known as a painter, but he made a number of trips for the Farm Security Administration's photo unit, and to my mind produced work that is among the best in the whole collection, where he is up against some astounding talent, including Walker Evans, Dorothea Lange, Gordon Parks, and Russell Lee.
