Monday (in Australia) I shared a new frequent network diagram for Sydney, done not by the government but by a community transport organisation. In Adelaide, meanwhile, the governmetn released a new frequent network diagram in July. Here it is. (Original PDF here.)
Locals are discussing it here.
Adelaide is a very centralised city, but still, it's extraordinary to notice that you literally can't go anywhere at high frequency without going through the CBD. Only the stations of the O-Bahn, where routes converge from several directions to flow into the O-Bahn busway, is there any opportunity to make a frequent connection without going downtown.
Adelaide readers and citizens should think about the question: Do we really want it to be impossible to get around spontaneously — i.e. without much waiting — anywhere other than to and from the CBD? Since former Portland transit general manager Fred Hansen is now in Adelaide, I hope he is pitching the virtues of grid networks — which Portland has had since 1982. A full grid is probably not appropriate for Adelaide's geography and resources, but radial systems with grid elements — which I've been designing for years — could open up vast new all-day travel markets.
There are similar maps for Brisbane such as here ‘Transit Maps’ http://transitmapsetc.blogspot.com.au/ and proposals such as here ‘Do it yourself Brisbane Bus Network Review’ http://railbotforum.org/mbs/index.php?topic=8959.0
Brisbane has the same problem. Everything is radial and it is hard to imagine cross town services as there are limited bridge crossings in the suburban areas.
Here’s Fred’s report, which includes a lot on transport.
His key project appears to be a CBD tram loop.
In my view this is geometrically inferior to a mix of through-routed CBD buses (with priority), 15 min train frequencies and about five or six good east-west cross-town routes serving the big uni campuses and shopping centres.
Adelaide being long and narrow the need for orbitals is less than Melbourne or Perth, but one is probably desirable (something already exists but at low frequencies).
Here’s a map of the entire network:
(Not sure if there’s a version which isn’t horribly pixellated – they should just publish a PDF like other agencies do. But you can still basically understand things.)
It is not that the grid/circumferential lines are infrequent, but that they don’t really exist. Instead there are a bunch of very convoluted routes each of which travels circumferentially for a short period. Google Transit routes me through downtown when I try to make circumferential trips. But even if a circumferential route existed, it would be so confusing to find that few people would know how to take it.
@Eric the Adelaide orbital I was thinking of is Route 300.
It’s daytime only with a 30 min weekday/60 min weekend frequency. It does Arndale – Glenelg – Marion – Belair Rd – North East Rd – Arndale.
100/101 is another cross-suburban across the west.
Jarrett, It’s also extraordinary to notice that Adelaide’s journey to work mode share by transit (without a frequent grid) is higher than Portland’s… It’d be great if we could start seeing some more solid evidence of the virtues of grid networks in various contexts. “Vast new travel markets”…are you sure?
EN57, that’s pretty simple to explain. Portland has freeways from all four directions providing easy access directly to downtown and to each other. Adelaide… does not. The closest thing is the South Eastern Freeway, which still requires a good 3 miles of driving on a regular 4-lane street to get to the CBD, and it serves more sparsely populated suburbs in the hills. There’s also a 15-mile segment from Gawler to somewhat closer to the city, a short bypass by the port, and half-a-freeway to the south (it only goes in one direction!). None of these other freeways come anywhere close to the CBD, and none of them connect to each other. At least the buses, trains, and tram all go directly to the CBD and connect to each other there.
Jarrett- my own local bus network suffers from the same problem. If you need to go anywhere other than the CBD on a semi-frequent basis, the system does not work for you.
Given that not all cities have the destinations to justify multiple transit hubs, etc., in your network design experience, is there a percentage of resources that cities that want to improve non-core frequency should dedicate to making that style of service happen, even if productivity suffers a little?
I’m starting to think transit planners should initially just cut the CBD out when designing a network. If you look at transit demand by neighborhood or census tract or whatever, the CBD will always dwarf all other areas, sending the message that a radial pattern is all that matters. What if we ignored the CBD and designed the grid network first, then add in extra CBD service as needed? It’s kind of like the concept in Jarrett’s book that we should start out thinking about all-day headways, then add in peak service as needed, whereas most agencies focus on the peaks first and then fill in the off-peak as an afterthought.
While the map shows only high frequency services, the regular services also funnel into ‘Town’ in the same way. This makes getting across the outer reaches of the city very difficult, if not impossible. I live in the North East, and work in the North West. Using public transport, the quickest way is via the city, taking in excess of 2 hours. The major missed opportunity is to effective use the rail and bus systems together – use the buses to collect passengers from the suburbs and ‘dump’ on the rail network. While there are interchanges, I don’t think they’re used in a way that promotes their, and often require a 3rd interconnecting bus service.
The answer is obviously no – we want frequent services on all routes. But until we can afford that, it makes sense to concentrate resources on where they will do the most good. That generally means the busiest routes, though in some places we have to make do with the busiest sections of concurrent routes.
The 100 (Circle Line) bus was originally introduced as a frequent route so people could avoid having to go into the CBD and out again, but it failed to attract many passengers and subsequently lost its frequent status (and eventually lost its complete loop status). Because it didn’t directly serve many major destinations, it just wasn’t that useful. Also the suburbs it served were fairly close in, so the CBD wasn’t a very big detour. The 300 is more useful and I wouldn’t be surprised if it were included in the next round of Go Zone expansion.
There are many things wrong with Adelaide’s transit system. Rail passengers are treated with extreme contempt by a government department that thinks any improvement justifies any temporary closure. All lines have recently been subjected to extended closure for track upgrading, and currently the Tonsley Line is closed in order to upgrade the Convention Centre (above Adelaide station) and we’re told Adelaide station itself (the only station in the CBD, which is another problem) will be closed in January for that reason. It’s going to be chaos!
Meanwhile the trams are overloaded and don’t run frequently enough (did you notice on the Go Zone map, they can’t even manage a 15 minute frequency throughout the daytime? And running every 15 minutes can equate to standing room only all the way from Glenelg.) The buses have also had problems accommodating peak loads, and reliability on some bus routes is very poor. And real time information is still unavailable on most routes.
BUT WE DO NOT WANT, AND WOULD NOT BENEFIT FROM, A GRID NETWORK! We already have something better. Look on the other end of the radial bus routes – very few of them end in the middle of nowhere. Most of them converge on interchanges in major commercial centres. So it is already fairly easy to get from anywhere to anywhere without detouring into the CBD, but with the added advantage of the bus route probably going direct to your destination. The only problem is the frequency, but as you yourself pointed out, cutbacks have resulted in Portland having the grid without the frequency.
Don’t get me wrong – a few more cross suburban bus routes would be useful. But Adelaide’s network already has better coverage than that of Portland. And remember, unless the services are coordinated, a sequence of two bus services each running every 15 minutes is worse to get to your destination than a single bus service running every 30.
DLB_84, what parts of the NE and NW? I’m surprised you’d have to detour into the City because there is a cross suburban route (361) linking them.
Aidan writes: “a sequence of two bus services each running every 15 minutes is worse to get to your destination than a single bus service running every 30.”
Yes, but no transit can offer a single bus service every 30 to wherever you might be going, whereas a grid can offer two services each running every 15 minutes to EVERYWHERE in the gridded area. Again it comes down to how much you want your transit system to be useful for everywhere-to-everywhere trips.
Having said that, I’m not advocating a grid for Adelaide. I’ve designed many networks and I’ve never laid a regular frequent grid over a city of Adelaide’s density; the resources never allowed it. More likely you’d start with one or two frequent crosstowns, as several readers have suggested, and then grow from there.
So it sounds like a raging agreement to me. 🙂
The failure to take full advantage of the existing rail and tram lines seems very noticeable from Mr. Stanger’s description. The Glenelg tram line already provides crosstown service, if you look at the map, but….
You miss my point. Having to change between uncoordinated bus routes means running every 15 minutes to everywhere in the gridded area doesn’t equate to providing a frequent service between any two points in the gridded area.
Once again I disagree. Having routes converge on interchanges in major commercial centres is almost as good as a grid for anywhere to anywhere trips, but with the added advantage of a much higher proportion of demand able to be accommodated with a one seat ride.
Interesting you’re saying that grids require higher density, but it raises a much bigger question: under what conditions (if any) do they become a better option than converging on interchanges in major commercial centres?
It seems to me a lot of the claimed benefits of grids are illusory – the result of a comparison with a load of routes ending in the middle of nowhere despite a much better alternative being available.
I think Adelaide needs to ignore the CBD and design a grid network before implementing any more CBD services that are needed. Adding in the peak service later on would be more beneficial, this is not a new idea- Jarrett mentioned it in his book.