A farewell to farms

The state government announced yesterday that it is going to fast track the release of massive amounts of greenfields land along Melbourne’s growth corridors, in response to new information which suggests that the city population will reach 6.2 million in just 12 years.

Whilst there’s no doubt that the supply of land needs to be increased, the outskirts of the city isn’t the place to do it. It is terribly inefficient on account of the large distances from major trip generators, as well as the cost of new infrastructure which needs to be provided. Furthermore, very few people in the inner and middle suburbs or middle class newcomers are going to be prepared to live there. This still leaves significant price pressure on inner and middle ring dwellings.

Melbourne 2030 dead

This was the final nail in the coffin of Melbourne 2030, but it does represent an ongoing trend. When we look at the data on pages 53 and 54 of the excellent 2008 Transport Demand Information Atlas for Victoria, it’s obvious that Melbourne 2030 hasn’t made a real dent in outer suburban growth. In fact, it has increased slightly.

The proportion of the population living in growth corridors increased two percentage points between 2001 and 2006. These figures suggest that Melbourne 2030 was never seriously implemented.

A solution?

While the CBD and inner suburbs are getting quite a bit of apartment and infill development, It’s simply not enough to stem big population increases in the growth corridors. Given the extent to which the population is increasing (around 1500 per week) – something big needs to be done. I’d suggest large scale brownfields development in the inner west is a big part of the solution.

This is already happening in Southbank and the Docklands, but it’s time to move further west to Fishermans Bend. Put in proper high density development and run a four track cut and cover rail line down the middle, between Southern Cross and Newport. Two tracks would be for a metro style line that would provide Werribee services a quicker route to the city and the other two would cater for freight and Geelong line trains. Here’s a very rough idea of the sort of development I’m looking at – it’s big!


The whole project would be difficult to cost because any potential contribution from developers would be hard to guess at this stage. Add to that the problems with decking over the Westgate freeway and building a rail tunnel in the Yarra silt and costs could get quite high. That’s not to say they would be higher than the long term costs of providing infrastructure on greenfields sites, and we should emerge with a more amenable result. That’s not a problem though – a project like this won’t be happening any time soon!


16 Responses

  1. How many people do you think this development would cater for?

  2. Even if the high rise vs medium rise argument was settled in favour of the latter, there are still many, many sites in existing urban area yet to be done. I pass several every day:
    -Caulfield Racecourse car park

    -Docklands around the stadium and rail area

    -Docklands peninsula, right up to the end.

    -Yarra’s Edge and down Fisherman’s bend

    -the pathetic situation in Tooronga, including around the Monash Freeway

    -around Brusnwick, Fitzroy, Clifton Hill there are still far too many very low value commercial uses going on, old warehouses, junk yards and stuff from what I can see.

    -the old Freemasons and Mercy and looks look Peter Mac are going. Eye and ear might be being rebuilt but could easily be built inside one of the others eg St Vs or Royal Melbourne and would free up the Eye and Ear site

    -down Macaulay Rd is supposed to be light commercial/industrial but plenty of low value uses are visible.

  3. Re high density – a lot of Australians will say “Australians don’t want high density” but then will say “Australians don’t want to pay tax” – which is why government’s cant afford the transport system that pays for low density living.

    Packing’em in means that the govt gets a better average (negative) return from transport because same infrastructure, more riders. 30000 people or more per hour, each way, is easy with double track electrified rail signalled for 2 minute intervals.

    Even that would mean, though, that Docklands having 100000 people, stacked 50 high, all professionals working in the CBD, would still take 2-3 hours to clear. Heaven help Dandenong and Pakenham, if they get their million people, legacy trains, legacy trackage, legacy operating culture!

  4. Thanks Matt and Riccardo.

    Matt, I’m not sure exactly how many you could fit. But if Docklands can take 25 000 people – as VicUrban claims – then that should give you some idea. This sort of development would be at least 5 times the size and higher density.

    Riccardo, I fully agree that there’s plenty of low hanging fruit in terms of inner city land that could be used first. Well designed infill replacing junk buildings won’t interfere with heritage and would increase amenity. That’s on top of the compelling density/efficiency arguments.

  5. Here’s an idea – I’ll work it up on my blog. But some numbers for you to get started if you want to analyse.

    Suppose you wanted to accommodate half of Brumby’s 1 million the good old fashioned (European) way, to turn many of the interurban hamlets into true villages by simply sharing small portions of Brumby’s mill among them.

    A lot of these towns (Tallarook, Clarkefield etc) have the classic rural dying disease. Corner pubs and banks left empty. Blocks marked for subdivision never used. Old town plan gazettals never taken up.

    Now I’m assuming for the larger cities below that classic brownfield/infill techniques are used, but for the small villages and hamlets, the following formula:

    2000 people = 4 per block (mum dad and 2 kids). Block = 600m2 (fairly standard aussie block). 600m2x500 blocks = a circle of land with a radius of 300 metres (sqrt(300000m2/pi))

    In other words, within walking distance of each of these minor rail stops, you can accommodate 500 blocks or 200 people. Spread this out to maybe 500m radius to cover rail corridor, roads, services, small commercial area. This is entirely low rise and does not account for existing buildings.

    The same formula works for 5000 people, 1250 blocks, 450m radius blob – expand to maybe 800 metres to allow for roads, commercial district etc. Much in walking distance, maybe cover rest with small bus.

    Suggest expand Vline service sensibly, reinstate Healesville ($20m), Nyora ($50m,Leo line too stuffed beyond) and Whittlesea ($20m on top what the Mernda project which should go ahead anyway). Fix Stony Point and rebuild some missing stations.

    Normal Vline expansion calculated on one breadwinner commuting into town, and half of these by rail. For 30000 extra people added to Seymour corridor, this would be extra 7500 commuters and 3750 by rail. This would be achieved by running 6 car velos every 20 minutes during peak hour. Assume 600 per velo, 6 velos in 2 hours, 3600 people. Discounts for early and late travellers.

    The name of the line, and the proposed numbers are shown next. Sums to 500000 people.

    Geelong 100000
    Little River 2000
    Ballarat 100000
    Rockbank 1000
    Melton 10000
    Bacchus Marsh 5000
    Ballan 2000
    Gordon 1000
    Bungaree 1000
    Bendigo 100000
    Diggers Rest 5000
    Sunbury 10000
    Clarkefield 2000
    Riddles Ck 2000
    Gisborne 5000
    Macedon 2000
    Woodend 5000
    Kyneton 10000
    Malmsbury 2000
    Taradale 2000
    Castlemaine/Chew 5000
    Harcourt 2000
    Seymour 10000
    Donnybrook 2000
    Beveridge 2000
    Wallan 5000
    Wandong/H 2000
    Kilmore East 5000
    Broadford 2000
    Tallarook 2000
    Traralgon 10000
    Nar Nar Goon 2000
    Tynong 2000
    Garfield 2000
    Bunyip 2000
    Longwarry 2000
    Drouin 5000
    Warragul 10000
    Trafalgar 5000
    Yarragon 2000
    Moe 5000
    Morwell 10000
    Clyde 2000
    Tooradin 5000
    Kooweerup 2000
    Langlang 2000
    Nyora 2000
    Stony Point
    Baxter 5000
    Somerville 5000
    Hastings 5000
    Bittern 2000
    Crib Point 1000
    Healesville 5000
    Coldstream 2000
    Yarra Glen 2000
    Whittlesea 5000

  6. Should read: In other words, within walking distance of each of these minor rail stops, you can accommodate 500 blocks or TWO THOUSAND people. Spread this out to maybe 500m radius to cover rail corridor, roads, services, small commercial area. This is entirely low rise and does not account for existing buildings.

  7. BTW I haven’t discussed the merits of Brumby’s idea, or whether power, water etc sufficient, only how they might be accommodated in towns and by rail. Sorry for all the posts

  8. Sorry one last post. You’ll notice for some of these stations the station site is at some distance from the town centre. This is ideal – you can build your 2000 people blob around the station unmolested.

    I’m thinking of Gisborne, Tooradin, Rockbank and others. Net loss of farmland low, much of it degraded.

  9. generic 2000 people town

  10. Some people might not want to live in a hamlet on account of the reduced access to ‘city’ services (as provided by regional hubs like Ballarat) such as large hospitals and the like.

    However, I am in favour of some decentralization of population – USA has many large cities, we seem determined to have maybe 5 or 6 – doesn’t really make sense, but there really doesn’t seem to be any overarching policy on this; not at State Parliament nor Federal level.

  11. Dave, you’ll see my idea has beefing up of Ballarat, Bendigo and Geelong – I don’t know if their water supplies can take it, but getting people into regional economies is, as you said, the USA way.

    Geelong, in particular, has plenty of dead patches. In terms of weatlh creation, I know a lot of people are paying rent and are on welfare, but helping the existing people of Geelong to grow their property values by injecting a little demand, will do wonders for regional prospertiy and will open up the things you mentioned people like about big cities eg hospitals, schools and so on. This of course needs to be supported by sensible govt policies.

    In rail terms, making Ballarat, Bendigo and Geelong bigger could also provide a reverse and counterpeak traffic as people live in outlying towns and commute.

    I’d like to see all the state government departments such as Education, Health and so on, moved out to regional centres, with only the Premiers Department and the Treasury staying in Melbourne. There must be thousands of jobs in the state bureaucracies, in a sensible plan would have people transferred over time, so people who don’t want to go leave the job when they retire or quit.

  12. I’ll stop hijacking Phin’s blog and do this as a photo essay on my own. Need to get the car and camera out for a day out at some of my favourite spots.

  13. Thanks Riccardo and Dave. Riccardo, you’re not hijacking at all -and moreover I’m inclined to agree with the idea of expanding provincial cities and the smaller towns.

    As mentioned, the V/Line service would need to be further increased, and half hourly off peak frequencies shouldn’t be out of the question. Really, Geelong should be getting a train every half hour already.

    I think RFR has gone some way to improving the situation – but additional Vlocity sets are probably required and there needs to be a serious effort to segregate V/Line and metropolitan services in the suburbs (I’ll have more on that issue soon).

    We also need to make the regional towns more interesting somehow, and I’ve no idea how to do that. Mind you, they are often a lot nicer than places like Caroline Springs etc. so maybe it’s just a job/transport problem.

  14. Re 2030 being dead, the activity centre concept was poorly thought out.

    Compare with Sydney (see my blog post Bogan’s Bash their betters – in aid of good urban planning)


    Sydney has had decentralisation with teeth for a long time. Melbourne doesn’t need “Activity Centres” it needs real full blown alternative CBDs with real jobs in them, and transport networks reconfigured to suit.

    And that’s the rub. All good urban planning ideas for Melbourne, whether it be ultra-low density or ultra-high or anywhere in between, require reconfiguration of the transport network, public and private, and this is the bullet they won’t bite.

  15. People are always talking about the failure of the ‘radial network’ yet as I see it the main failure is the “Pakenham” failure -it is now too big for the technology at hand and becoming congested.

    The size issue was never a problem for the likes of Lilydale or Hurstbridge because these places were in no danger of swamping the rail service.

    Pakenham in particular poses a great challenge because it is in danger of swamping the limited rail service that can be provided with current asset base and with political constraints of ‘fairness’, union work practices and so on.

    The Sydney cure of having multiple CBDs actually reproduces the ‘radial effect’ of the original CBD and creates mini-radial networks around each of the new CBDs.

    Hence a place like Chatswood will have its own ‘network’ – a line towards Sydney CBD, the outer North Shore line, and the new line to Epping.

    Because most of these locations are on a mid point of a rail line, they already have a ‘radial rail network’ from the momemnt they are zoned as a CBD – a network of 2 lines. Those on or near junctions eg Hornsby, Liverpool, Bankstown, already have a mini-network of even more lines. This is with existing infrastructure. Buses etc support this on other routes.

    The radial effect is good because it reinforces rail’s line haul capability.

    Being decentralised gives strong counter-peak flows (ie counter to the ‘main’ peak for the Sydney CBD) so that Chatswood will have some reverse flows from stations like Artarmon. This also supports rail’s economics by putting bums on seats that weren’t there before, on trains that have spare capacity.

    Some would call it not ‘decentralisation’ but ‘recentralisation’ ie reinforcing the idea that centralisation is good for urban economics, but that any one centre eg the historic CBD, has reached the limit (diseconomies of scale).

    Hence places like Caulfield, Footscray, Sunshine or Camberwell should be naturals for it.

  16. Believe it or not, the service standard will not work for the Hurstbridge line becasue of late running on the single line ruining everything, even with out timetables but a notional “service standard”. Try using Diamond Creek or Eltham stations every day for a year and you’ll soon see what Connex does with short terminations in times of late running. The single track is murder to any concept of regular running.

    Also in my experience a lot of the poor patronage beyond Greensborough is caused by supply-induced-demand (or lack thereof). I can say from personal experience that a lot of people from Lower Plenty, Eltham, Montmorency, Kangaroo Ground, Hurstbridge, Cottles Bridge, etc just drive their cars to Greensborough, Watsonia or Ivanhoe to reap the benefit of a much higher service frequency and a very reduced risk of short terminations effectively cancelling their trains on a regular basis.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: