Between school, Passover, and life, I’ve been extremely short on time and (especially) focus to write recently (though I did do a Twitter series about Amtrak reform, Storified here), but I do want to get something up.
Jake Blumgart has a post in the Philly Voice about SEPTA’s fabled Route 23, the former-trolley-now-bus that connects tony Chestnut Hill in the city’s Northwest with Center City and South Philadelphia. Route 23 has been acclaimed as the world’s longest trolley route, a claim I’m skeptical of, if only because Chicago’s Route 49 (Western Ave.) has run up to 20 miles at times. It’s also a flashpoint of the bus-trolley wars, with some romantics consistently calling for the return of the trolleys that were taken off the route in 1992 (much of the rail infrastructure remains in place). The problem is that, as Blumgart notes in his piece, “When trolleys rumbled along the 23, an end-to-end ride took more than three hours; the bus takes about half of that time.” (I believe that number is somewhat exaggerated–Blumgart got it from long-time riders of the route–but there’s little doubt that the trolleys were indeed slower than buses that can pull into and out of traffic)
Route 23 intrigues me because it combines two particular challenges of urban transportation planning: super-long mixed-traffic transit routes and diagonal streets that cut across the city grid.
Route 23 map from SEPTA’s schedule packet–so long it needed to be split into two images!
As you can see from the map, Route 23 runs largely in a two-way pairing on 11th and 12th Streets from its terminus in South Philadelphia through Center City, then switches onto the diagonal Germantown Avenue, running northwest to the tony Chestnut Hill neighborhood. The route is also very long for a local bus, almost 14 miles. As a result of the length and Philly’s notoriously narrow and congested streets, Route 23’s on-time performance in 2012 was only 64%, fairly atrocious by transit standards. As a result, SEPTA is considering splitting the historic route in two, with one branch running from Center City to Chestnut Hill and the other from Center City south. Since–as Blumgart documents–virtually no riders use the whole route, there seems to be little opposition to this change.
And that leads me to thinking about the viability of super-long local bus routes in other places. I used to commute on the Chicago Transit Authority’s #49 bus, the core service on Chicago’s uber-long (24ish miles) Western Avenue. Where Route 23 is scheduled to cover its 13.8 miles in about 1:15, for an average speed of roughly 11 mph, CTA’s 49 is even slower, scheduled to cover its 15.7 or so miles in 1:30, for an average speed just under 10.5 mph. Aside: this means that the CTA 49–which has, in the past, extended past its current terminal at Berwyn all the way to the Evanston Line, about 3 more miles–is both longer and slower than the much-maligned Route 23. Unsurprisingly, the 49 is a massively unreliable route, with bunching common. Since the cancellation of the former X49 limited a few years ago, it is the only transit option in the corridor, and one of the most popular routes in the city, carrying even more people than the 23.
The problem with the 49 is that, unlike the 23, there is no especially convenient place to split the route in the middle. I’m sure relatively few people ride the route from end to end, but there’s no point where the entire bus empties out and exits, as happens to the 23 in Center City. Additionally, I will personally testify that having more splits in the route would be a massive pain in the ass, since having to transfer from the 49A and 49B extensions to the core route in order to continue a linear journey is already a major problem. The 49’s reliability problems can’t be dealt with by cutting the route in half, so what options are left for the CTA and other operators faced with similar long-route challenges? (that’s a genuine question!)
The other thing that’s intriguing about Route 23, of course, is that its northern half runs along Germantown Avenue, a diagonal street that once connected that neighborhood to the Philly waterfront. Though Philly’s grid isn’t as regular as, say, Chicago’s, Germantown Ave. still stands out as an oddity on its generally northwest-southeast path.
Germantown Ave. highlighted, from Google Maps
Germantown Ave.’s odd alignment–and its Phillyish narrowness–makes it a challenge for fast, efficient transit–but an opportunity for other things, as James Kennedy of Transport Providence, a Philly native, pointed out:
This is a question that intrigues me, since Chicago’s notorious diagonal streets have proven to be a major challenge for traffic of all types in that city. Of course, Chicago’s diagonals are more regular, such that they often cross two other arterials in a nightmarish six-way intersection. Witness Lincoln, Damen, and Irving Park:
Or Lincoln, Ashland, and Belmont:
Among other terribleness, the new Google Maps is temperamental about embedding, so you get a JPEG for this one.
These intersections are horrible for pedestrians, create massive traffic jams, and just generally suck. And there’s not too much the city can do about them. The problems at the intersection of Damen, Fullerton, and Elston are so bad that the city is spending millions to realign the intersection, but low-value industrial land isn’t usually available to do that.
Chicago DOT’s graphic explanation of the Damen-Elston-Fullerton realignment.
Perhaps the boldest initiative Chicago has ever undertaken to tame one of its diagonals was the 1978 pedestrianization of a short stretch of Lincoln Avenue (no, not all of the horrible intersections involve Lincoln, but it does have many of them) southwest of the intersection with Lawrence and Western, creating the Lincoln Square (mostly) pedestrian mall, the core of one of the city’s hottest real estate markets:
Perhaps the Lincoln Square mall has been buoyed (and to tell the truth, it hasn’t exactly been a smashing success) by being next to the busy Western Brown Line station. But it might also represent the potential of a new approach to those problematic diagonal streets. The luxury of a grid is that it often works best without diagonal streets cutting through it at angles that are either random (Germantown) or too regular (most of Chicago’s diagonals). Surely, the idea of making these streets into a transit, bike, and pedestrian mall is radical. But it may be a really good idea.
Itinerant Urbanist 
I never lived in Chestnut Hill, Mt. Airy, or Germantown, but if memory serves me correctly, much of the grid in the NW neighborhoods of Philadelphia are also diagonal-ish, which is one of the few places where the grid gets so turned askew in an otherwise orderly and Quaker-ified city. Your point about diagonal streets may not apply directly in this case, but it’s an interesting one regardless, and it definitely has applications to other Philadelphia streets (Passyunk, maybe, or the fingers of the West Philly arterials–Lancaster, Baltimore, Chester, Woodland, etc.).
I really hope somebody in Philly takes us up on our conversation point about Baltimore Ave. losing a lane of parking on one side from 40th to 43rd, as a test. This wouldn’t give the 34 trolley a ROW (yet) but it would test out the ability to have non-door-zone protected bike lanes, which in West Philly would be popular. Eventually, if the whole thing panned out, maybe neighbors could be convinced to go car-free on Baltimore, like what I suggested on Germantown Ave. That also could be done a few blocks at time, to let people get accustomed to it and to see if it pans out. It would make the trolleys a much better addition to the city.
I will freely admit to not knowing Philly well enough to judge where modifying the diagonals might work, so I’ll defer to your judgment on that :-). I’m hopeful that the significant modifications coming to the West Philly trolleys in the coming years will include at least stretches of dedicated lanes–it seems like there’s a real opportunity to leverage the new investment there.
And yeah, I think someone should absolutely nudge the city to try out some different kinds of closures and space re-allotments and see what works. Planners (and transportation planners in particular) tend to be scared of experimentation, but we shouldn’t be 😉
Wait, why are trolleys slower than buses that have to pull over at stops? That generally means that the buses then have to wait for cars to finish passing before they can get moving again, which tends to add a few seconds at every stop, which tends to add up pretty significantly over a very long route with many stops like the 23. A trolley (or bus that stops in the travel lane) blocks traffic and creates a nice bit of clear road so there’s no delay in getting moving after a stop, and potentially even some free road space to get a bit of a speed boost over the regular traffic speed.
You’d think it would work that way, but apparently Center City is congested enough with double-parking and generally bad driving that the buses’ ability to maneuver around obstacles is a major advantage. From what I’ve heard, this has been the experience of newer streetcars that operate through downtowns as well. It’s probably less of a problem for the West Philly trolleys.
Oh I don’t doubt that double parking and other obstacles are bigger problems for streetcars than buses (compounded by some new systems’ puzzling insistence on putting the streetcars in the outer lanes), but the ability of the bus to get around obstacles does have to be weighed against the risk that it’ll be forced to get out of the way of cars, and lose time getting going again once that’s happened.
It’s not just Center City, it’s the whole route. Germantown Ave has 2 parking lanes and 2 travel lanes. There are no extra left turn lanes at intersections. Imagine the trolley sitting behind cars backed up to make left turns while cars going straight are pulling to the right to get around the backup (unless cars are parked too close to the corner for even that, in which case, everybody’s slowed down there too!)
Mixed-traffic streetcars on streets with only 2 travel lanes basically stopped making sense after the era of mass auto use. People who fetishize old timey things would love to see the 23 trolley come back but no transit advocate should unless it got dedicated lanes. Maybe take the parking lane from 1 side of the street?
The precedent of Broadway in New York shows that on diagonal streets in a grid you can (sometimes, at least) massively remove car lanes without congesting the grid.