Why are we still so focussed on CBD rail?

Bike lanes - Hitler hates them!!

It’s bizarre, but much of the contemporary debate about expanding public transport in Melbourne is focussed on building more rail lines to the CBD even though the great bulk of jobs are now in the suburbs. It seems we’re locked in an old way of thinking when the world long ago moved on to doing things in a very different way.

The proposed Rowville, Doncaster and Melbourne Airport rail lines are evidence of our preoccupation with radial train lines. They were key issues during last year’s election campaign and the Government promised to study the feasibility of all three. They all connect to the CBD. No one refers to any of them as, for example, the proposed “Rowville to CBD” rail line – the debate is so CBD-centric that’s taken for granted.

Yet the city centre now has only a fraction of all jobs in the metropolitan area. The traditional CBD – that area bounded by Spring, Flinders, Spencer and La Trobe streets – has just 10% of all jobs in Melbourne, as mentioned here. Even the entire City of Melbourne municipality has only 19% of metropolitan jobs.

But it’s not just jobs (although they’re very important because that’s where rail does best). The vast bulk of travel by Melburnians for non-work purposes – which involves considerably more trips than the journey to work – is also not directed at the city centre. It’s local and over relatively short distances.

One consequence of  this ‘radial thinking’ mindset is the popular sentiment that ‘black holes’ in the network, like Doncaster, have to be ‘filled in’ with new rail lines to the CBD. This is despite the likelihood that the justification for a Doncaster-CBD line borders on the farcical. For example, Eddington estimated that only 8,500 workers living in the municipality of Manningham commute to the City of Melbourne, of whom 3,150 already take public transport (and this was before the new DART bus rapid transit system started!).

With 72% of jobs located more than 5 km from the CBD and 50% more than 13 km, it’s surprising that the focus of public transport expansion isn’t on improving services for cross-suburban travel.  Why, for example, isn’t there more emphasis on improving suburban orbital services and feeder services to major suburban destinations?

The fact is Melbourne’s radial train system was never intended to deliver workers to suburban jobs. The existing rail lines are the ‘spokes’ in a radial system that’s designed for the high peak loadings generated by the CBD. In contrast, economic activity in Melbourne’s suburbs is quite dispersed – no more than 20% of suburban jobs are in the largest 31 activity centres – so something far more flexible than commuter heavy rail is likely to be required.

Adapting our public transport system to the new reality requires a vastly different mindset from the traditional ‘radial’ view of the world. The task in the suburbs is not ‘mass transit’. The solutions that make operational and economic sense are more likely to be buses – using a combination of existing road space, dedicated lanes and some exclusive busways – than rail. There’s a possibility of light rail in corridors that build up traffic but in the short term any such comittment will probably be driven more by political pressure than by good policy.

Existing rail lines will nevertheless have an important role in the suburban public transport solution. Priority should be given to developing a ‘grid’ of services and to coordinating connections within and between modes. There’s room for a base or backbone high frequency network, but I suspect much of it will have lower frequencies and require much effort in coordination.

Ironically, most of the proposed Rowville line actually runs east-west, but it is conceived as a CBD service (and consequently would require substantial investment in amplification works on the Dandenong line). If the strategic focus were on the peculiar geography of suburban travel, it might instead be conceived as a shuttle service to the Dandenong line (meaning passengers would transfer), giving scope for other lower-cost options such as buses or light rail in a dedicated lane.

An alternative conception might also provide scope to develop the Clayton/Notting Hill area – which is already by far the largest employment concentration in the suburbs – as a major activity centre. Or it might be conceived as providing linkages from the Clayton area to other parts of the region.

I’ll look shortly at why buses seem so unpopular.

19 Comments on “Why are we still so focussed on CBD rail?”

  1. brisurban says:

    I would agree here. Melbourne has huge rail network with over 200 stations. But it is only one part of the transport system. The connections between lines need to be filled in. Every bus stop needs to be viewed as if it were an extension of the rail system. You could get some really excellent services “filling in” the gaps using buses. The SmartBuses are already a good start.

    I can understand the perspective. People think they are doing good by just building more and more infrastructure. But that is so costly at $80-150 per kilometre. How many buses can you get for 1km of rail? The problem I see is is mainly lack service, not lack of extended infrastructure. Cross-town extensions of rail can be done by bus on the street.

  2. brisurban says:

    * correction. $80 – 100 MILLION per kilometre. If only infrastructure were that cheap!

  3. Siobhan says:

    Re Doncaster – someone was telling me the other day how, after much lobbying, a bus service was put in Manningham Road, but closed after 6 months due to an almost complete lack of patronage. I suspect it would take an awful lot to get the people in that district out of their cars. This makes the cost of heavy rail even more problematic.
    On the surface of it, it would seem more sensible to start with buses, then have a slow but persistent approach to switching to light rail/tram on well patronised routes.
    But problems with buses also need to be addressed, if they deter people from taking up public transport. We need to understand clearly why buses are less popular and work out how much we can change that.

    • brisurban says:

      I guess it might be one idea to have bus connections to the lilydale rail line, sort of like this: http://melbpt.files.wordpress.com/2008/08/doncaster-bus-map.jpg

      Light rail and buses are also an option. I think the Melbourne train network is already complicated enough without adding further arms or branches – splitting the frequency and adding conflicts. The Eastern Freeway is a nice alignment to perhaps put Perth-like high speed bus or LRT down and feed it using bus and park & ride.

      I’m surprised that there seems to be a presumption that better transport in the area must be heavy rail. Was there a modal choice analysis done for this area or is it just popular because it was mentioned in a 1960’s era transport plan?

      I’m a bit surprised that there hasn’t been that much discussion about the current tram network and making that run more like light rail (wider stop spacing, light prioritization, own lanes).

    • Michael says:

      “We need to understand clearly why buses are less popular and work out how much we can change that.”

      Good luck with that. I don’t believe the reasons are properly understood and it’s probably because the answer lies in territory economics and planning don’t venture. Maybe a sociologist might have the answer.

  4. Sam says:

    Look at who is calling for rail lines to the CBD:
    – people who work in the CBD and moved to an area not covered by rail; or
    – people who live in an area not covered by rail and who have obtained a job in the CBD.

    The first group are used to having the luxury or rail and when they move away from it campaign for extensions.

    People driving to their jobs in the suburbs aren’t calling for extra rail lines (except in limited circumstances, Rowville perhaps) because they would prefer to drive than use rail – or can’t perceive of rail as an alternative.

  5. rohan says:

    Probably this focus comes first from current CBD commuters who feel squeezed and disappointed with the level of service, so there is a need for more services, though more lines, who knows ? But dont forget your figues may be misleading – people dont use PT only for jobs – there would be a large proportion of students and people travelling to the CBD for other reasons, and plenty of people work in the inner suburbs, so take a train then a short tram – you will really have to add up all the jobs in Fitzroy, Carlton (hospitals, university)and St Kilda Rd to get some idea of exactly how many people use the radial network to get to the inner suburbs not just the CBD. In the case of Doncaster, like other cities, once a fast option is available (and yes could be a properly done DART) more people would swap from car to PT, and perhaps even more people would see a CBD job / pleasure trip as more feasible. And Rowville is as much about getting people to Monash Uni and other business along Wellington Rd as it is about getting Rowville people to the CBD, or Caulfeild or Sth Yarra.

    • Alan Davies says:

      The post is about expanding the rail network – the lamentable performance of the existing network is a different issue.

      Re jobs in Carlton, Fitzroy, etc, see the figures in the post on City of Melbourne’s share of all metro jobs (19%) and inner city’s share of all metro jobs (28% are within 5 km of CBD).

      Travellers are much, much more likely to use public transport for the journey to work than they are for non-work purposes. Eddington forecast a Doncaster line would carry only 24,500 passengers max per day by 2021 (all purposes, two way). For comparison, in 2007 the Frankston line carried 51,500.

      The Huntingdale-Monash crush is a real issue, but it’s not primarily a CBD issue. Take out the city centre bias and it opens up cheaper options like light rail or BRT along North Rd.

      • rohan says:

        Interesting that 28% of all jobs are within 5ks of CBD – interesting to know how many of all them take PT to work. Yes most PT use (rail and tram) is to/from work, but there must be a singificant % of students into CBD, and shoppers. But yes current rail system lamentable, and Doncaster and Rowville may be just as effective as light rail and indeed serve the local community as much as CBD commuters – Croydon in London is a ‘mini’ CBD now, with its own light rail network. (though a Doncaster line would be half the length of the Frankston so it actually compares ok !)

  6. […] – Why are we still so focussed on rail to the CBD? […]

  7. jack horner says:

    For obsession with rail to the CBD, add the fantastically expensive proposals to increase inner city rail capacity in Brisbane (Cross River rail at $7 billion) and Melbourne (Metro rail tunnel at $4.5 billion).

    In each case the proposal does nothing to extend the reach of the PT network as a whole, and is not well justified in cost benefit terms (as far as you can judge from public information).

    In each case the project, considered as a capacity increase, is probably not needed within 20 years, providing you make some other less costly fixes in the interim. [note 1] Of course major projects should be planned with that sort of lead time, however the current political rhetoric is that these projects are near term priorities for construction, which they are not.

    Excessive focus on peak services to city centres is arguably detrimental to developing a more useful total public transport service, as inevitably it takes resources away from more widespread improvements. You could buy an awful lot of bus priority roadworks for $4.5 billion.

    Note 1: ‘less costly’: as in, HUGELY less costly. The Melbourne metro rail tunnel proposes 12 track km of new line, and five new stations, to create 14 new peak hour peak direction train paths. Connecting the Burnley and Northern loops (see p30 of the 2008 Dept of Infrastructure report ‘East West Rail Link Analysis on Rail Capacity’**) would need up to 2 track km of new line, and no new stations, to create 20 new peak hour peak direction paths. I guess it would cost a tenth as much as the metro rail tunnel.

    ** http://www.doi.vic.gov.au/DOI/DOIElect.nsf/alldocs/2DE07DE58D313680CA25741E0021CC53/$FILE/EWLNA-Analysis_on_Rail_Capacity.pdf

    • brisurban says:

      Speaking from Brisbane, we need something that is going to increase capacity for people coming on the Gold Coast, Cleveland and Beenleigh lines during the am peak hour on the Merivale Bridge Crossing. The only other river crossing by rail into Brisbane is on the Ipswich line which is a run-around route and has its own problems.

      The solution is another rail crossing of the Brisbane River. The Inner City Rail Capacity Study did touch the subject of costs and benefits for the Cross River Rail project, and the benefits come about three times more than the costs of construction. An article also touches on the CRR impacts: Link here http://www.couriermail.com.au/ipad/cross-river-rail-delays-to-cost-us/story-fn6ck51p-1226020722236

      Brisbane’s Cross River Rail is high cost, but it is also high benefit. It is not meant to directly extend the reach of the rail network, it will allow greater capacity on the network and relieve the bottleneck that is the Merivale railway bridge crossing. The busway is not a substitute as it also has its own capacity issues and the rail lines in question are long distance. http://www.couriermail.com.au/news/features/push-for-rail-tunnel-funding/story-fn4z2520-1225834056511

      • jack horner says:

        Apologies, I admit I’m not so familiar with the Brisbane example.

        By ‘less costly fixes’, for Brisbane I was thinking of –
        – three door a side cars giving shorter peak dwell times to push capacity to 24 per hour per track with existing signals;
        – longer trains (admittedly, with significant cost of extending platforms at Central);and/or
        – an extra track pair in the existing rail corridor from Park Rd to Bowen Hills.

        • brisurban says:

          Oh, yes, we need things to tide us over, so yes, you are certainly not off the mark there.
          The project has been delayed 2 years due to flooding costs. Extra doors, less seats and signalling upgrade to allow 24 trains per hour would be something very good on the current system.

  8. Simon says:

    Jobs follow the transport infrastructure, so building more radial train lines will create more jobs in the CBD. And this is a good thing IMO, as it focuses employment and activity, giving rise to the benefits of agglomeration, as seems more environmentally sustainable than encouraging jobs and development in low-density areas.

    Put simply – the “new reality” of suburban jobs you talk about is the result of past transport decisions. We can “adapt” to it, or we can make a series of new transport decisions that will create an even newer reality of more CBD jobs. What future do we want?

  9. Sean Deany says:

    While still in the developmental stages and likely never to be built I am working on a scenario plan for a Melbourne Metro System which does not entirely take people to the CBD. What Melbourne needs, and can be arranged on the existing network, are firstly cross City lines, not to mention more frequent services and little delays. I have nonetheless taken it all a bit further in my scenario where there would be other lines through out the suburbs taking into account peoples “out there”.

  10. Sean Deany says:

    The 42.7KM Frankston Line has 27 stations / 5 rail interchanges, from point of origin, before reaching Flinders Street (excluding the City Loop). While in my scenario for a Doncaster Rail Line – The Doncaster Metro it would have up to eleven new stations / 3 rail interchanges along its 26.25KM alignment from the proposed Parkville Stn to Ringwood Stn. I have limited the number of new stations – most of which are primarily inter-modal changes with existing or to be extended tram network and feeder buses. Also it is important when concerning inter-station distances that high constant-speed regimes can be met therefore allowing a faster rail service.

    With key transit orientated developments (TODs) at Parkville – Doncaster Hill – Ringwood – Knox City – Stud Park – Monash / Notting Hill a large area of the Eastern suburbs potentially can be linked by a new rapid rail / metro line, instead of depending entirely on buses or MIT (motorized individual transport) alone. Therefore the high monetary cost involved through building a Doncaster Rail Line and a proposed Rowville Rail Line could be justified by additionally linking these up via the existing Ringwood – Ferntree Gully rail corridor. This metro rail scheme would be a highly valuable to a large existing and certainly to increase population. Indeed it would be to an extent a non-CBD centric asset for transit orientated developments in the Eastern suburbs of Melbourne.

    My scenario option – Doncaster Metro:

    Distances are for platform mid point / tunnel face. Int – rail interchange / U. – sub-surface / S. – surface level station

    00.00KM Int. U. DM 01: Parkville Stn (platforms 03 / 04 – Royal Pde aligned)

    01.15KM U. DM 02: Carlton Stn (Lygon / Palmerston Streets)

    02.05KM U. DM 03: Fitzroy Stn (Alexandra Pde / East of Nicholson St)

    03.70KM Int. U. DM 04: Clifton Hill Stn (platforms 03 / 03 – Mayors Res)

    04.70KM DM Parkville Tunnel Portal / Merri Creek Bridge

    04.82KM DM Merri Creek Bridge / Cutting

    05.04KM DM Yarra Bend Tunnel West Portal / Yarra Bend Road

    05.45KM DM Yarra Bend Tunnel East Portal / Eastern Freeway median

    06.75KM S. DM 05: Willsmere Stn (Eastern Fwy median / East of Chandler Hwy)

    11.45KM S. DM 06: Bellevue Stn (Eastern Fwy median / West of Bulleen Rd)

    13.25KM Doncaster Hill Tunnel West Portal (Marjorie Cl, Bulleen)

    15.75KM U. DM 07: Doncaster Stn (East of Williamssons Rd / Westfield)

    17.75KM Doncaster Hill Tunnel East Portal (Doncaster Rd median)

    17.85KM S. DM 08: Jackson Court Stn (Doncaster Rd / Jackson Court)

    19.85KM S. DM 09: East Doncaster Stn (Doncaster Rd Tunstall Square)

    21.25KM Mitcham Road Tunnel West Portal

    22.05KM Mitcham Road Tunnel East Portal

    22.15KM S. DM 10: Donvale Stn (South of Eastlink / West of Park St)

    22.65KM existing Mullum Mullum Tunnel West Portal

    24.25KM existing Mullum Mullum Tunnel East Portal

    25.05KM Ringwood Tunnel Portal

    26.25KM Int. U. DM 11: Ringwood Stn (platforms 04 / 05 – below Railway Place)

    I have extrapolated station locations based on historical planning, current popular views, Melway and topographic map referencing, as well as knowledge gathered through Google Street View data for my scenario plan.

  11. PRT Fan says:

    Good point there…all these new train lines are designed to pump more people into the CBD.

    But the Government loves cars so much, it has built the orbital Ring Road and EastLink for them.

    The way to go is to look at the new Delhi Metro, built from scratch, first line opened in 2002. They are obviously taking a wholistic view and are building 2 orbital metro lines.

    Worth looking at their map and building an orbital rail line in Melbourne: http://upload.wikimedia.org/wikipedia/commons/thumb/6/68/Delhi_Metro_Phase_3_Route_Map.svg/2000px-Delhi_Metro_Phase_3_Route_Map.svg.png

    (^ planned orbital lines are in grey)

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 )

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