In recent years the Melbourne Metro rail tunnel has been an on and off again project, which to the public has just been a line drawn north-south over a map of the Melbourne CBD. I pieced together whatever plans I could find a few years ago, but the recent publication of the Melbourne Metro business case dated December 2011 was a big win – it features the gory technical details I have been hanging out for. So let’s take a look!
Some background
My initial blog post covers the origins of the Melbourne Metro concept – a short summary:
- 2008: ‘East West Link Needs Assessment Study‘ produced by Sir Rod Eddington features a 17 kilometre long tunnel, to be built in two stages under Swanston Street.
- September 2009: Release of design concepts for the new underground stations.
- 2011: Victorian Government goes to Infrastructure Australia for help funding a shorter 9 kilometre tunnel, to be built in single stage and terminating at South Yarra instead of Caulfield.
- 2012: Visitors on my blog speculate regarding the depth of the tunnel – will it be deep under the existing City Loop tunnels as per early plans, or cut and cover down Swanston Street?
- May 2014: Liberal Government under Premier Denis Napthine proposes an alternate Melbourne Rail Link plan.
- April 2015: Labour Government under Premier Daniel Andrews announces Melbourne Metro is again the preferred project, via a shallow tunnel under Swanston Street.
And finally, here we are in May 2015 with the Heral Sun uploading a copy of a planning document dated December 2011 – you can download the full document here.
Planning stages
Inside the business case the reasoning behind how the chosen alignment was decided upon is given, as well as the alternate routes that were rejected.
Figure 4-1 shows the rail corridor options initially considered back in 2007-2008, which resulted in a north-south tunnel being preferred.
With a route decided upon, station locations and the alignment of the tunnel had to be decided upon.
Figure 4-3 shows the station and alignment options considered.
The options were presented in a schematic form in figure 4-4.
Options for the alignment between Domain station and Caulfield also had to be considered – figure 4-5 shows these.
A result of the above work, in 2011 the decision was to build the Melbourne Metro as a single stage, with two CBD stations, a rail tunnel beneath Swanston Street, and a southern connection to the rest of the network at South Yarra station. The business case had this to say about this choice:
Preliminary assessments of Russell Street and Elizabeth Street alignments indicated that they would be $430 million and $600 million more expensive than the default, respectively.
Investigation of a western CBD alignment via William Street was pursued at the request of City of Melbourne. A William Street alignment through the CBD would also cost an additional $470 million.
A single CBD station would provide a considerably poorer outcome in terms of interchange with the existing rail network, access to employment and other activities and relief to the tram network. It would result in 15% fewer passengers using MM compared to the default arrangement of two CBD stations.
Recent investigations have confirmed Toorak Road as a feasible alternative to the default scheme which provides similar benefits earlier and [snip] and significant short-medium term cost savings of $2.5 billion compared to the full default scheme.
Another decision to be made was the vertical alignment. Back in 2008 the East-West Link Needs Assessment ‘Analysis on Rail Capacity’ document featured two deep tunnel options, both of which passed below the existing City Loop and CityLink tunnels.
The 2011 business case had this to say on the alignment options:
The initial concept that underpinned the 2010 business case adopted a vertical alignment of the tunnels in Swanston Street which passed underneath the existing Melbourne Underground Rail Loop (MURL) tunnels east of Melbourne Central. This resulted in relatively deep and expensive station boxes, with CBD North being 43m deep at track level. This option included a maximum grade of 3.5%.
Following the peer review, further detailed work was undertaken during 2011 to assess constructability and feasibility of a shallower tunnel and stations along Swanston Street and to refine the alignment and station configurations. A shallower (higher) alignment would result in a better outcome for passengers and may provide cost savings.
The constructability assessment showed that a high alignment was feasible, though the tunnels would be more expensive to build and disruption would be more significant. The Yarra River crossing would also present greater challenges and risks. Significant cost savings would be expected in construction of the stations however.
A comparison of the high and low level alignment is shown diagrammatically in Figure 4-11.
A comparison of the station depths was also included in Table 4-9.
The business case went into the benefits of the shallower tunnel for passengers:
The improved access associated with the high alignment and shallower stations reduces passenger walking time to surface at CBD North Station by around 90 seconds and CBD South Station by around 60 seconds.
It also went into how a shallow crossing of the Yarra River would be built:
The Yarra crossing is proposed as an immersed tube construction due to the shallow depth of the alignment. This method requires dredging of the river bed prior to floating into the place the prefabricate tube sections. It is proposed to construct the tube sections on the south bank of the river on east of the existing boat sheds.
Stations
Once the stations locations were settled, the entrances to said stations also had to be decided upon.
Each of two CBD stations have been provided with two concourses (at the north and south ends of the station) while the other stations have been only provided with one (in the centre of the station).
Arden
From the 2011 business case:
Nine potential station entrances were examined. One station entrance was assessed as sufficient.
And a matching render from the PTV website circa-2012:
Parkville
From the 2011 business case:
Nineteen potential station entrance locations were proposed for the Parkville station.
Two main entrances on the east and west side of Royal Parade were identified as important. In addition, three special entrances were considered, notably the train-to-tram interchange in Royal Parade and direct concourse access to the Victorian Comprehensive Cancer Centre (VCCC) and the Peter Doherty Buildings.
A matching artists impression from 2012:
And an earlier design concept dated 2009, which still fits in with later plans.
CBD North
From the 2011 business case:
Eleven potential station entrance locations were examined, including interchanging with Melbourne Central Station. The preferred option for interchange is a subterranean concourse under La Trobe Street linking directly to the existing and proposed concourses at each station.
Entrances 2 (north-west corner of Swanston and La Trobe Sts) and 8 (south-west corner of Swanston and Franklin Streets) were assessed as the best primary entrances points accounting for passenger flows and known constraints. Entrance 9 (south-east corner Swanston and Franklin Streets) would be an alternative entrance to Entrance 8 if Franklin Street is closed to traffic east of Swanston Street.
An entrance at the south-east corner of La Trobe and Swanston Street (State Library) (Entrance 1) is not proposed due to heritage considerations of this site. Consideration was given to a more direct connection with the heart of RMIT University on the east side of Swanston Street (3, 4 and 5) but not pursued to avoid disruption to RMIT’s main campus.
Figure 5-8 from the same document shows the concept for CBD North, with the station located north of La Trobe Street, with the tram tracks splitting around the glass roof that covers the concourse.
That design matches this circa-2012 artists impression of the station concourse.
It also matches this cross section diagram of the station, which was found in the media pack for the April 2015 announcement of the preferred shallow tunnel route.
CBD South
From the 2011 business case:
Twelve potential station entrance locations were examined.
The option taken forward for the purpose of the 2010 business case was a ‘paid area’ subterranean concourse at the east end of Flinders Street, providing direct connection to the existing Flinders Street platforms via lifts (escalators were not considered feasible). During 2011, further options have been considered in the context of the shallower (high) tunnel alignment. A simpler connection between the subterranean CBD South concourse and the existing Flinders Street concourse is now proposed.
The following entrances are the optimal locations:
- Entrances 3 (Swanston Street opposite St Paul’s) and 4 (City Square) are the recommended primary entrances.
- Entrances 2 (Federation Square) and 6 (Flinders Lane/St Paul’s) are also deemed to be desirable to spread pedestrian demand to the east (less crowded) side of Swanston Street. Entrance 2 at Federation Square would be important in handling event patronage.
Direct access to the south-west corner of Swanston Street/Collins Street was considered to better support this interchange but is not proposed due to the impact and cost of land acquisition in that location.
Figure 5-9 from the same document shows the concept for CBD South, where a single concourse is located directly above the station platforms.
Figure 5-10 shows how the concourse at CBD South with interchange with Flinders Street Station.
Further detail of the interchange passageway was found in the 2012 Flinders Street Station Design Competition Design Brief document. From the side:
And from the top:
The design matches this circa-2012 artists impression of the Collins Street entry at Town Hall.
But this undated view of the Flinders Street end is a little confusing – to me it seems to be showing the deep level option.
Domain
From the 2011 business case:
Twenty-three potential station entrances were examined.
The preferred entrance locations are at 17 (corner St Kilda Road and Bowen Crescent) and 16 (corner St Kilda Road and Bromby Street). An additional direct access to the new tram interchange above the station in St Kilda Road is also proposed.
Figure 5-11 from the same document shows the intended ground level arrangement of Domain Station.
Again, a match for the circa-2012 artists impressions.
How to build it
With design decisions completed, chapter 5 of the business case then looked at the scope of preferred scheme.
Figure 5-3 provides a horizontal alignment overview of the tunnel.
Figure 5-4 shows the vertical alignment.
And Figure 5-5 shows the geological makeup along the route.
Characteristics and design parameters for the underground stations are presented in Figure 5-7.
Figure 5-12 shows the intended construction methods – mostly bored tunnels, with cut and cover work through the CBD, and immersed tubes under the Yarra River.
And Figure 5-14 shows potential site that can be redeveloped once construction in the CBD has been completed.
I’m sure glad the the full document finally saw the light – you can take a look at the full document here.
[…] December 3rd, 2012. Much has happened to Melbourne Metro in the years since this post – read about the current plans here. […]
i think tunnel should run halfway between swanston and elisabeth, should be deep and bored all the way, stations should be at a) halfway under the yarra, and b) halfway under melbourne central.
a) providing exits at south bank, exits connecting up with flinders street station, and further north.
b) exits at south end of melbourne central, melbourne central station, north of melbourne central.
as for deeper bore, who cares? nobody will object to sitting on an esculator. this cut and cover stuff is a joke, you are talking about cutting a 40 meter ditch. better still this tunnel is useless without awesome connection. spend more money now for less pain in the future.
I’m curious as to the Australian fear of deep level stations – various rail tunnel plans in Sydney have also avoided them for evacuation and travel time concerns.
For the purposes of comparison, the Eastern Bloc is full of stations 40 metres deep, or more:
http://www.eurogunzel.com/2014/01/world-deepest-metro-station-arsenalna-kiev/
Hong Kong has also opened a new station 70 metres below ground, with access via high speed elevators:
http://www.checkerboardhill.com/2015/04/lift-only-mtr-station-entrances/
Because deep bore is more expensive, requires a feat of engineering that might not exist in this country. Cut and cover is not really a tunnel though. I think technically it’s the lazimans tunnel. I believe Sydney has a lot of cut and cover if I am right? Not sure though.
In Sydney the City Circle had many cut and cover sections:
http://www.sydneyarchitecture.com/SUB/SUB-07.htm
But newer lines such as the Eastern Suburbs Railway, the Airport Link, and the Epping-Chatswood rail link were all bored.
It will be interesting to see what they come up with at the CBD south station in particular with the shallow tunnel option and the consequential open cut in Swanston Street in the vicinity of Finders Street. A significant open cut like Museum station with the City Loop project in a tighter area (okay the street will be just closed rather than diverted)
From there they then go at a shallow depth under the existing rail lines, themselves under the road, and Yarra somehow. At this point you go from reasonable mudstone to silts and gravels under the Yarra and perhaps some anchoring of this section down to bedrock.
I’m not sure ‘Dan’ has gone through this issue properly!
Engineering firms Aurecon, Mott MacDonald and Sinclair Knight Merz (SKM) appear to have been involved in planning so far, along with Grimshaw Architects.
http://www.aurecongroup.com/en/projects/transport/melbourne-metro-study.aspx
For the CBD South station I wonder if they will just dig the giant hole and close off the entire area for a year or two, or adopt ‘top down’ methods and build the roof first.
Thanks Marcus.
I wonder if the CBD Sth station will be ‘placed’ in the middle of Swanston like CBD north. Currently the CBD Nth section of Swanston St is closed for traffic, so like you say trams can curve around the station there, but (restricted) vehicle access still exists at the sth end.
I have a preference for the shallower tunnel, however, it seems a waste to demolish the super stops that have been installed over the last couple of years. How much has been spent on these stops in this part of the corridor? If they had of been erected 10 years ago I would be more forgiving, but the stops outside RMIT (CBD Nth) were only finished within the last 1-2 years??
The Toorak Rd option is the one being pursued (for cost reasons) but it seems illogical not to connect (subterranean pax tunnel) to Sth Yarra. Do you know what the difficulty/constraints to this are? Are the numbers of interchanging pax (to Sandringham line) just too small? It also ignores residents of Sth Yarra who may prefer to use MM.
Are these station names final? Whilst CBD Nth / Sth assists with geography (particularly for tourists) they seem a bit boring.
Whist on tourists, is there any stat of how many use the rail network? I would suspect very little as the majority would not travel to the suburbs and the tram network is obviously much more accessible to destinations within the inner ring.
Sorry this is a bit of a negative post. Its not intentional, just keen to see more detail!
Apparently the four platform stops on Swanston Street cost a total of $25.6 million:
http://thecityjournal.net/news/teething-troubles-on-swanston-st-upgrade/
The stop at the State Library forecourt cost $7 million, with the others being $6.1 million a piece.
https://www.melbourne.vic.gov.au/AboutMelbourne/ProjectsandInitiatives/MajorProjects/SwanstonStreet/Documents/4.1_NoImages.pdf
The cost was split between the City of Melbourne, the Department of Transport, and (presumably) Yarra Trams:
http://www.melbourne.vic.gov.au/AboutCouncil/Meetings/Lists/CouncilMeetingAgendaItems/Attachments/9495/CCL%205.6%20%28FMC2%205.9%29.pdf
Wow.
That’s much less than I would have expected given the inflated cost of every other (small and large) project atm.
Thanks.
As for South Yarra station, I’m guessing the reason it was excluded was cost – the current station is woefully inadequate for the number of people already using it, so to add even passengers to it would would require the construction of a massive underground concourse and additional station exits.
@Marcus, the need for a Sth Yarra reconstruction anyway might make the economics of integrating it into the new line worthwhile. Certainly the local population growth is significant combined with interchange opportunity.
@Beren (and Marcus) – the 60-90 extra seconds on an escalator is time that otherwise has to be made up elsewhere, either by reducing the ped-shed, increasing line speed, increasing line frequency to reduce average wait time, etc. It reduces the competitiveness of the option compared to surface options (tram, bus, taxi, car).
Near-surface provides ready access and dedicated ROW increasing the likelihood of use or a longer walk to access. I can’t see why you wouldn’t choose it if the option was available. The short-term additional disruption of 12-24 months (depending on block by block or single-stage construction, etc) compared to a likely hundred year plus life of the asset (look at London Underground, 125 years old now) is, while not trivial, manageable in context.
Okay, but the length of the escalator forms a buffer. See, there are multiple buffers during departing a vehicle. First, the one to the door of the train, then the one between the door and the escalator, then the one between the escalator and perhaps another escalator, further to that between the escalator and the gates, then in the case of Melb Central, those two escalator pairs, then wrapping around to the shopping centre.
The important part here is the bottlenecks, the shorter one of those buffers is, the more congestion you face at it, causing a bottleneck. Now, long escalators are a more efficient buffer. There is less likely to be congestion at the bottom because things have had longer to average out. Short escalators are a problem, where basically flow issues at the top and bottom cause issues on the escalator itself, passengers crushing up, not using the fast lane option.
None the less, what is observable is that Melbourne Central is possibly the worst offender, it contains the least amount of escalator distance then any of the three stations. Both Flagstaff and Parliament have much longer escalators, and let’s be honest here, they flow much better. Shorter escalators do not help Melbourne Central, the buffers are so short that people crush up, and this is not only dangerous but inefficient, to the point where the station itself needs many more escalators to deal with these flow issues. And technically this issue cannot be solved, this station during peak is bad, and not a good example, because by the time you’ve gotten through the gates, you aren’t any better off.
Trams in the city can be a terrible option too, they stop at every stop, takes ages for passengers to get off and on, and honestly, there are days where I can probably outpace a tram on foot. Then on some stretches the trams are bumper to bumper, clogged up, and just silly. Not putting trams down, just our city has grown past them, they haven’t kept pace with the expansion of the population, and neither has the train network.
The bus network seems to still be run like a Centrelink office. Meandering bus services, cutting down every side street in Melbourne on a frequency that could only be described as pathetic. Nope, trains are awesome, heavy rail. City loop is busting full of patronage, I think that if this metro tunnel doesn’t happen, you really need a major upgrade to Melbourne Central, in fact, why not upgrade Melbourne Central at the same time as the tunnel? Do them both at the same time, and link them together. And even move the station a bit so it’s not being stuffed up by the shopping centre.
Anyone else notice that the shops are basically hindering the flow of passengers? Who ever decided to make the east exit flow through a shopping mall possibly couldn’t imagine passengers brushing up against a clothing stall and customers trying to buy stuff. It’s terrible, give the passengers a direct route similar to the west exit.
Disagree on the buffer argument. The escalators at Parliament, for instance, are a real bottleneck and there’s plenty of crushing, congestion etc there. Not to mention if someone block a lane standing on the right, you’ve got to wait the 75 seconds for them to clear the escalator compared to 20 for a short flight.
Acknowledge the trams have issues, but as greater priority is rolled out, they’ll improve. And you can’t easily take similar measures to reduce travel time on an escalator, except perhaps manage ped flow better.
It’d also be less expensive to just add overflow stairs to shallow platforms, something not likely to be done nor used for deep platforms.
It seems a mistake to assume that 2 or 3 entrances will be sufficient for the CBD stations.
Sydney’s Town Hall station has about 10 entrances.
Even stations like Southern Cross and Flinders Street have multiple exits, scattered across different streets, and in the latter case, multiple levels by passing under neighbouring streets.
Yep the stations seem to be designed like moscow metro or more specifically the existing city loop stops, that is with one or two large scale major entrances, rather than many little ones like London, New York, Paris etc. Parliament for instance has a huge stair facing east Melbourne, whereas the bigger crowds use the little one facing Bourke St, practically the only one that’s like a Paris/New York entrance.
There should be big ones with escapes and lifts where they fit, but multiple others to spread the crowds, eg Swanston at Flinders with little stairs dotted along the block.
Fun fact – the Russian word for a metro station entrance is вестибюль – “vestibule” in English.
You are quite correct about Moscow Metro stations often having a single entrance – I visited Russia back in 2012, and it fits my observations.
For deep level stations, a single bank of escalators runs straight from platform level to ground level, where the ticket barriers are located. Even at major interchange stations, the same layout was common, except the platform for each line had a separate set of escalators.
However their shallow stations followed a slightly different design – I passed at least one example with escalators at each end of the platform, leading through the fare gates then into a network underground walkways, which then headed to the surface via stairs.
As for stations with lots of small entrances, the Hong Kong MTR is full of examples.
http://www.checkerboardhill.com/2015/04/lift-only-mtr-station-entrances/hku-station-layout/
This station above a total of five entrances, split across three different directions.
DOES ANYONE KNOW HOW COULD FED MINISTER FRYDENBERG REVOKE RESCIND OR CANCEL OR EVEN SIMPLY AMEND THE 2015 AGREEMENT DUE TO THE NEW NATIONAL HERITAGE AND HERITAGE VICTORIA LISTING OF THE BOULEVARD SHRINE TREES ETC>>>?TO DICTATE AN ALTERNATE ROUTE THAN THE ONE DOWN ST KILDA ROAD?
[…] Figure 5-10 in the 2011 Melbourne Metro business case […]