Saturday, 7 September 2019

Should the Airport Rail line use existing tracks?

There's been some speculation in the media recently about whether the Andrews government might not build a tunnel between Sunshine and the city, as part of the Airport Rail line they're currently planning. There's been a lot of ink spilled on this topic already, and my PTUA colleague Daniel has done a good job responding to this, so I won't retread the same ground - but I did have a few brief points I wanted to emphasise.

Changi Airport Station in Singapore

Travel times and stopping patterns

The rumours that suggest the Airport line might use "existing tracks" aren't very specific about which tracks they are - the Metro suburban lines, or the Regional Rail Link lines. Not so long ago it was envisioned that Airport Rail would use the suburban lines and plug into the Melbourne Metro 1 tunnel - but I don't think that would be the preferred option today. Pretty much everyone agrees that the Airport line needs to be very quick, which means at least a limited express service - stopping at every small station along the Sunbury line would slow the train down way too much, as well as potentially causing issues with overcrowding from regular non-airport commuters.

VLocity trains are timetabled to travel from Southern Cross to Sunshine in 12 minutes, including one stop at Footscray station. This takes into account the very slow speeds between Southern Cross and South Kensington, creeping over the North Melbourne flyover; and the fact that even the nice new straight bits further out are limited to 130km/h. Even with these issues, it's only 12 minutes. With some work at the Southern Cross end, this could easily be reduced to 10 minutes.

The Sunshine alignment to the airport (via Big Build)

By track distance, Sunshine is just under halfway to the Airport; brand-new dedicated infrastructure on this leg should have far fewer issues than the mess around Southern Cross, so the train should easily complete the second leg in 10 minutes, even if it stopped somewhere like Keilor East along the way. (I'm assuming the Airport line would have electrified trains similar to the Metro X'Trapolis, with fast acceleration and a top speed of 130km/h)

So looking at it purely from a speed perspective, we don't need to worry about expensive tunnels, or avoiding Footscray to shave a minute off our travel time - a total travel time of 20 minutes is eminently achievable along this alignment, even with a few stops.


This is where things get trickier, and it's where the case for building this tunnel becomes clearer; it's more about adding capacity than increasing speed.

Because of the issues with speed and stopping patterns, we can effectively rule out putting the Airport trains on the Sunbury line, so if it's to use existing tracks, that means Regional Rail Link. The problem is that there isn't really any room on the RRL tracks, particularly in the peaks. But can we make room?

RRL currently carries three lines - Bendigo, Ballarat and Geelong. Two of these lines are effectively pulling double-duty, serving both large regional centres and bits of suburban Melbourne - this above all else is the source of the strain on the RRL tracks.

V/Line services serving Melbourne's western suburbs

As I've said many times before, we urgently need to electrify the lines from Sunshine to Melton and Wyndham Vale, and add an extra track pair for the Ballarat and Geelong trains to run express. The new track pair would get plugged into the RRL corridor, while the converted existing track pair would get plugged into the Sunbury line.

This would free up a lot of space on the RRL corridor; freed from the burden of running a suburban service, you could run a lot fewer diesel trains along here and it'd still be plenty enough to cover Ballarat and Geelong.

So what could you do with all the space you've just freed up? You could put the Airport trains on it.

Free lunch

Now - just because it would be theoretically possible to do Airport Rail using existing tracks, doesn't mean it's free.

To make this happen, they would probably need to:
- completely revamp the Melbourne Yard, between North Melbourne and Southern Cross Stations
- electrify the RRL tracks (since the Airport train will need to be electrified, for the ~6km of tunnel between the Ring Road and the airport)
- extend the MM1 tunnel's High Capacity Signalling from West Footscray (where it's currently planned to end) to Sunshine, to cope with Sunbury + Melton + Wyndham Vale trains
- do all the aforementioned electrification and quadding west of Sunshine

V/Line and Metro trains at Footscray Station

I suspect the above shopping list would be cheaper than digging a massive tunnel under Footscray, and would have a lot of non-airport benefits as well (everything listed, except possibly the RRL electrification, needs to happen anyway). But clearly it wouldn't be free.

If they tried to shoehorn Airport Rail onto the existing tracks without these measures, it wouldn't really work. So I think it's very important to be clear when we have this debate - there's taking the more pragmatic and economical option, and then there's just half-arsing it. Money needs to be spent and work needs to be done, whichever way they go.

So should the government just forget the tunnel and use the existing tracks?

Honestly? No, I don't really think so.

For one thing, we will need more capacity between Sunshine and the CBD in the very near future. The growth in both Melbourne's western suburbs, and the big regional cities in the state's west, is huge and will continue. If they did use the existing tracks, they might defer the need to build the tunnel - but they'd still need to build it (or something very like it) within the next 15-20 years - and it'd take a good 5ish years to build. It buys us time, but not much.

For another, this is kind of a rare moment in politics. After decades of inaction on Airport Rail, and presented with a federal government that seems pretty reluctant to spend on infrastructure in Victoria - especially public transport infrastructure - we find ourselves in the rare situation where the feds are willing to put serious money towards the project. The planets have aligned, and if there were ever a time when we could build a whopping great tunnel and future-proof ourselves for the western growth that's coming, it's now.

The state and federal governments are finally both behind Airport Rail (cropped from @tminear)

There is a growing sense that the state government has bitten off more infrastructure projects than it can chew - both the financial and the practical resources (workers and raw materials) are getting a bit strained. So there's a worry that they might start trying to de-scope or delay some of the projects they've pledged.

If push comes to shove, and the choice is between delaying Airport Rail with the tunnel for several years, or building it ASAP using existing tracks, then yes - use the existing tracks and get it done now. It's already been delayed for decades, we can't let it be delayed further.

But if the government is looking to consolidate their construction agenda, the multi-billion dollar megaroads that will only make congestion worse should be first on the chopping block - not vital public transport projects like this.


  1. Great post. As you say, a tunnel is about future proofing.

  2. I like the option of a new tunnel going from Sunshine to Southern Cross via the Brooklyn freight line alignment then via Fisherman's Bend along the Melbourne Metro 2 (MM2) route. This would open up Fisherman's Bend to high-density development, and without it many of the planned 80,000 jobs and 80,000 residents won't happen there.

    - State Government is unlikely to fund MM2 without some Federal assistance, because it has already made commitments to the Suburban Rail Loop
    - Fisherman's Bend will be like a second CBD, attracting sufficient patronage (perhaps from new hotels) to justify a stop on the way to the airport
    - Provides a way to access jobs in Fisherman's Bend from the west via Sunshine
    - Provides a connection between airport services and the Werribee line at Fishermans Bend

  3. Is it possible to use that freight line? That could be a decent option.

    1. It probably wouldn't be practical to run Airport trains and freight trains on the same tracks, just because of the huge difference in speeds (and the high frequency of the Airport trains). I suppose theoretically you could repurpose the existing freight line for Airport trains, and then build a new freight line to replace it, but you'd probably end up with broadly similar problems