Airports & aviation

Sep 20, 2017

What are the key issues for Melbourne Airport rail?

The solution is seen as a single rail line, but it should be conceived as the package of infrastructure and policies that will best improve the accessibility of Melbourne Airport

Alan Davies — Editor of The Urbanist

Alan Davies

Editor of The Urbanist

Looking from domestic airport station, Brisbane Airtrain, towards airport terminals

The idea of a train from Melbourne Airport to the CBD is never out of the news for long in Victoria. Now the Committee for Melbourne, which sees a mass transit link to the airport as one of Melbourne’s nine strategic needs, is holding a forum on the topic next month.

So I thought I’d have a stab at setting down what I see as the main issues around this topic (in no order):

When will the airport link be needed? Infrastructure Victoria doesn’t expect it will be required for at least 15 years; it thinks the existing bus system can handle forecast demand with some upgrades to roads to provide priority access. It’s possible though that forecast demand might change, either to bring the horizon closer or push it further out.

Should it be a shuttle service for the exclusive use of airport travellers? A dedicated service minimises problems with baggage and requires fewer stops, but it also limits the potential benefits. The alternative is a service like Sydney’s Airport Link that also serves metropolitan travellers.

Should the airport service be integrated with the wider transit system? The Government proposes it should share tracks with Melbourne Metro; the advantage is many travellers in the south-eastern suburbs would get a one-seat journey to the airport. Some argue this would limit capacity for suburban services and propose instead a dedicated airport line terminating at Southern Cross Station.

What route will it take? This requires optimising several variables, especially travel time (important for the B in BCA), cost of construction, patronage, and impact on the efficiency of the metropolitan rail system. The cost to construct the lowest cost option using the reserved (mostly) at-grade route is likely to be at least $3 Billion. Some options involve extensive tunnelling and could cost a lot more.

Should it be routed to support intensive urban redevelopment? Building the line via the former Footscray munitions site and/or Essendon Airport could significantly increase housing supply and generate revenue provided the development potential of the increase in accessibility is fully exploited. The cost of construction would be higher and the net benefit of relocating existing uses would need to be established.

Should it be routed to support development of dormitory regional centres? This would benefit country travellers and help relieve growth pressures in Melbourne, but has wider implications. The net benefit of decentralising residential growth compared to current growth strategies would need to be assessed. Another issue is whether there’d be enough country travellers to justify the cost.

What will be the key benefits? It’s likely the lion’s share of benefits would come in the form of faster, more predictable trips, mainly for business travellers. History indicates it will have little impact on traffic congestion or on the pressure for motorway widening. Benefits from lower emissions, less pollution and fewer crashes are likely to be relatively small.

What mode share will it win? Sydney’s Airport Link claims close to 20% mode share but Brisbane’s, which is arguably a better parallel with Melbourne, has circa 10%.

Who will own and/or operate the line? There are a number of possible models with varying degrees of government control and accountability.

How will ticket prices be set? This is likely to have significant implications for the level of use. The usual practice with airport rail, even when operated by government, is to set prices to recoup capital and operating costs. The polar alternative is to set them in line with current myki prices.

What is the scope for related travel demand policies? How road access and airport parking are priced will have a big impact on patronage and the financial performance of the airport link.

What about non-CBD travellers? Most airport trips by Melburnians originate or end outside the city centre. The warrant for providing new transit options for trips from the eastern and western suburbs requires investigation. This highlights the need for a comprehensive assessment of airport access options (see Is an airport train to the CBD the whole story?).

This list isn’t exhaustive but they’re what I see as the main issues. There are other important matters that I think aren’t as central at this time, including choice of mode, location of terminals and stations, the potential of airport-related business agglomeration, the impact on SkyBus and taxi operations, and more.

The most important point in my view is to conceive “the problem” in terms of overall airport accessibility rather than only as a single mass transit link to the CBD. That implies a suite of measures across infrastructure construction, infrastructure pricing, and land use and regulatory policies.

It’s terrific the Committee for Melbourne is facilitating discussion. The Government needs to put a big and early effort into investigating, assessing and planning options for the airport; this is a great opportunity to get it right before some half-cocked politician locks the State into an ad hoc solution.

(Visited 31 times, 1 visits today)


Leave a comment

Your email address will not be published. Required fields are marked *

14 thoughts on “What are the key issues for Melbourne Airport rail?

  1. Vikraman Selvaraja

    Nobody has yet been able to explain to me why we can’t just build a 7km spur line off Broadmeadows station elevated across the existing roads terminating next to the Park Royal at Melbourne airport. This would probably cost a little more than the Mernda extension on South Morang line (8.2km, half elevated, 582 million) and provide most of the benefits, especially if timed with Metro completion. Seems like a no brainer, but the PTV Airport options plan costs it at many multiples of that and doesn’t favour it because of congestion at Nth Melbourne, which will be irrelevant once Metro is done. No matter how hard I read that PTV plan, I still can’t get to the justification of why it is so much more expensive per km than recent suburban rail works (Mernda, Frankston line level crossing removals etc)

    1. Alan Davies

      Vikraman Selvaraja

      The 2012 study on route options for airport rail done by Parsons Brinckerhoff for PTV looked at six Craigieburn line options and shortlisted a connection south of Coolaroo for appraisal. Relative to the baseline Albion East route, the study found the capital cost would be the same but the Craigieburn line option would be slower (35 minute trip vs 30 minute), patronage would be lower (91%), it integrates with fewer stations, at the time it had seven level crossings (Albion East has none), and would take some green space in the Attwood corridor.

      1. Airport

        ^^ I find that interesting though.
        The train would only be 5 minutes slower, but as it would have many more stops, its passenger catchment would be wider.

        As another commenter said, Sydney’s train stops at intermediate stations, as does Brisbane’s.

        1. Alan Davies


          The Parsons Brinckerhoff study gave the Albion East route a higher score for connectivity/integration because it gives access to Sunshine and Footcray activity centres, CBD (via CBD Nth and CBD Sth), and out to the Dandenong activity centre. The Craigieburn route gives access to Broadmeadows activity centre, stations on the line, and CBD.

  2. Jacob HSR

    Are you serious? Either a shuttle service or a Sydney-style service?

    What about the one in HKG or JNB or KUL or BKK or DEL?

    All of those (except SYD) are special railways with special trains containing luggage racks. All of them pick up passengers between the airport and the city – not just at the airport.

  3. DB

    For a truly successful airport link, what share of ridership should employees/workers represent? If it’s a substantial share, how do the alignment, station locations, and integration with the overall pt network need to respond in order to attain that share?

  4. Airport

    I do not believe that Melbourne Airport Rail will be much faster than an improved (i.e. dedicated lanes) bus option, particularly if it travels via the Albion corridor.

    How many minutes faster will it be? (compared to improved bus, not just the bus now)
    10 minute frequency- is that all day or only during peak hour?
    Is heavy rail the “best” mode compared with others such as Light Rail or Brisbane style “metro” buses?

    Personally, I think the State Government has failed to do a basic mode choice analysis for political optics reasons. Rapid light rail down the Tullamarine freeway corridor (similar to say Perth trains or Seattle LRT) should be looked at, and indeed all other modes as well.

  5. Tony Morton

    Thanks Alan – these are all good questions. In recent months PTUA has been developing a policy response that aims to address all of these, which we’ll be sharing more widely in due course. Some immediate reactions though:

    – $3 billion is AECOM’s upper bound estimate for airport rail construction on the existing plan. Their lower bound estimate is a little over $2 billion. Furthermore, in both these estimates 40 per cent of the figure is attributed to ‘indirect costs’ such as design expenses, profit margin, and because it’s a rail project, allowances for ‘rail occupations and approvals’. But it’s also evident that AECOM has applied the same indirect cost escalator (of about 60% of direct costs) to every rail project they assessed for IV, whether it’s ‘greenfield’ or ‘brownfield’ in nature and whether or not there’s an existing planning reservation. We’re talking here about a substantially ‘greenfield’ project with an existing surface-level planning easement – if these factors count for anything in reality they suggest AECOM’s estimate is highly conservative.

    – This also of course feeds critically into the question of whether growth in travel is best accommodated by expanding road capacity or providing rail capacity. Infrastructure Victoria is relying on AECOM’s costings in assuming the former, concluding that accommodating as much travel as possible by road and deferring a rail option is both cheaper and more desirable for Melbourne. These are highly contestable claims.

    – Alternatives to the existing plan will cost more, but could also provide higher benefit, so we’re open to considering these – but that should not unduly delay the planning process. The existing plan has already been sitting on the shelf for almost two decades while Melbourne has grown around it.

    – ‘History suggests’ there would be little benefit…in fact there’s very little relevant ‘history’ at all. Melbourne hasn’t had a significant suburban rail extension for nearly a century. The RRL comes closest, but that’s all of 2 new stations with people being accommodated on a maximum 20 minute V/Line service designed for regional travel. Though judging by patronage, I’d suggest it is nonetheless taking significant pressure off the West Gate corridor. And what do you think makes Brisbane’s airport situation a better parallel than Sydney’s? Brisbane’s airport trains only run half-hourly during the day, and the airport itself doesn’t operate around the clock.

    – Accommodating non-CBD travel will be a vital planning consideration, but keep in mind that travelling via the city is still going to be the quickest and most convenient public transport option for anyone in the broad sweep of the metro area from the Lilydale rail catchment due south; likewise, starting from south-west of Sunshine will favour a route via that interchange point. The north-west and northern suburbs and the Yarra valley are where the city-to-airport link fails to be useful, and other options such as an improved 902 Smartbus service should be developed to cater for them.

    1. Alan Davies

      Tony Morton

      Infrastructure Victoria’s Options Book (p475) puts the cost of airport heavy rail as “Direct option cost $3 billion–$5 billion”. There’s plenty of evidence that these sorts of early, preliminary cost estimates for big projects – and more so for rail than road – are almost always optimistic. I’ll stick with my “at least $3 Billion”.

      1. Tony Morton

        Yes, the $3-$5 billion is the total deiivery cost, including operating and maintenance costs over some decades. So not just the ‘cost to construct’ as you implied above. It’s also a gross figure – it doesn’t take into account revenue offsets.

        1. Alan Davies

          Alternatively, the AECOM/Pwc technical report commissioned by Infrastructure Australia estimates the capital cost range, assuming an elevated station at the airport, as $2.1 – 3.2 Billion, but notes that “significant cost risks exist for services and planning with numerous interfaces with existing infrastructure”.

          The report adds this qualifier in terms of cost risk:

          Elevated structure construction along the line will cause significant impacts of the operation of the roads and intersections. This may include road closures and diversions which will incur additional costs.

          There has not been intensive geotechnical investigation into the ground conditions for this project and this should be undertaken. Depending on the outcomes the required foundations for elevated rail structures and associated
          works will be affected and costs related to these.

          Services through the airport area have not been investigated and are likely to include communications and fuel as well as water, gas and sewerage.
          The Maribyrnong River bridge has been costed at approximately $50 million, but the interface with existing structures, environment and services has not been taken into account and any realignment through this area may
          add significant cost to the project.

          The bridge across the M80 will also present several challenges with long spans and high clearances required for the crossing as well as the proximity of high voltage power lines which may need to be lifted to allow for
          construction of the track.

          I don’t think it’s at all safe to rely on the lower bound estimate when costings are so preliminary and optimism bias is chronic. There are big gaps in technical information and considerable risk of scope creep e.g. underground station at airport.

          1. Tony Morton

            It does say all that Alan, but presumably AECOM’s (already wide) quoted range of estimates is supposed to take all that into account, otherwise it’s a pretty meaningless exercise.

            And as I said, the easement already exists. In fact much of it runs parallel to recently built roads, so a lot of the constructibility issues have already been encountered in the recent past.

            1. Alan Davies

              The risks highlighted by AECOM point to the wisdom of assuming the upper end of the range i.e. $3.2 billion. My early stage estimates for the cost of Melbourne Metro, East Coast HSR and Melbourne level crossing removals were all criticised in these pages for being excessively pessimistic, but all were shown by more detailed work (and in some cases the letting of contracts) to be significant under-estimates.

  6. Peter Hill

    Another option could be to provide a rail corridor easement that could provide for a Melbourne CBD – Melbourne Airport link in common with the much touted Melbourne-Canberra-Sydney High Speed Rail. That is, one rail line making those linkages. There could be a major infrastructure saving – but, would either transit service justify itself?

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details