Revisiting a Melbourne orbital rail line
On Saturday Plan Melbourne's refresh was finally unveiled to the public and contained within it is the map shown above. At first glance your eyes are drawn to the huge 'blob' of central Melbourne - issues with the resolution of the map aside, the high concentration of jobs outside of the central blob appear to correlate with other maps of job clusters right across Melbourne.
Fair warning: this is a very long article and if you can't read it all now, scroll down to the map, open it in a new window/tab and explore. The bulk of the article is about the thought processes used to conceive the accompanying map.
Infrastructure Victoria was set up by the current state government to provide independent advice to government on all things infrastructure and whilst its remit is broad, many different transport-focused projects were present in its Options Book to support the 30-year strategy. Now that the strategy has been submitted, the government is now in the process of formulating its reply via its legislatively required response and 5-year plan.
Infrastructure Victoria also has a remit for providing research, and it's already growing body of knowledge as well as its published intentions for 2017 is looking quite impressive.
While its 30-year strategy had a strong emphasis, at least in the transport realm, on how can we make our existing infrastructure sweat more through alterations or minor improvements, the big ticket items will always be there - be it in the foreground or background.
Much of the big ticket road network items in the strategy are focused on non-radial connections - case in point: North East Link (the linking of the Metropolitan Ring Road and the Eastern Freeway). The North East Link has now received support for getting the planning work underway from government (and opposition).
It wasn't present in IV's 30-year strategy, but there are a lot of advocates for a public transport equivalent to the North East Link, an orbital rail line in Melbourne. Curiously after speaking with a few transport planners and consultants over the past few years, it appears not much modeling has been done in this area. Could this change, however?
Over at Crikey, Alan Davies has given up his blog to a guest writer & transport consultant Garry Glazebrook who has written an advocacy piece for a massive $200billion 'Building Australia Fund' that would set a fire cracker off under infrastructure investment in Australia. And to follow up on some of the initiatives mentioned in Garry's original piece, a follow-up article was recently published which deals specifically with an orbital Melbourne rail line.
I found myself constantly nodding - furiously in agreement - when reading the most recent piece, particularly this bit on the what the characteristics of a Melbourne orbital rail line should be:
A proposal for a Melbourne Ring Metro, Garry Glazebrook - Crikey/The Urbanist
- It needs to operate at high frequencies throughout the day – typically headways of 3-4 minutes – since many people using it will be transferring from the radial network.
- It needs to have well-integrated interchanges with the radial train and tram networks, as well as with major bus routes.
- It needs to operate in its own right of way and be grade separated from roads and existing public transport systems so as not to cause delays to its own patrons or to the rest of the transport system.
- It needs to have reasonable capacity (e.g. 3,000 – 10,000 passengers per direction per hour in peaks) to accommodate current travel patterns and potential growth, especially as the sub-centres themselves grow in importance.
- It needs to be relatively low cost to build and operate, and to be capable of staged construction.
Garry Glazebrook only loosely referred to metropolitan locations like Frankston, Dandenong, Monash, Box Hill, La Trobe University, Epping, Broadmeadows, Melbourne Airport Sunshine, and so I submit the following map (and very long rationale brain-dump) as a possible route.
I highly recommend viewing the map in a separate window/tab. If you'd like to view the map in Google Earth, follow the previous link and click 'Download KML' from the settings menu and then import into the GE client.
What follows is a summary of the thought processes put into the map.
Firstly, a bit of background
It's the gravitational effect of the major job clusters in Moorabbin, Monash, Box Hill, Heidelberg as well as the smaller clusters in Doncaster, Burwood, Preston East and the 'fingers' that appear northwards and westwards from the larger central Melbourne 'blob' in the Plan Melbourne map - Brunswick/Coburg, Moonee Ponds and Footscray/Maribyrnong - that has primarily influenced the route as shown in the map.
It's also influenced the phased approach: Phase 1 would link three of the four major and two of the smaller clusters as shown in the map. Phase 2 would extend phase 1 and skirt around the periphery of central Melbourne and link all the 'fingers'. Phase 3 would complete the orbital by linking into the last major cluster or taking a different path to the Bay.
All of those clusters except for the region to the east of Moorabbin, Burwood (Deakin University), Maribyrnong, Doncaster and East Preston already have access to the radial rail network and the orbital route in the map aims to exploit that with interchanges at each and every radial line in these regions as well as connect directly with current or future bus and tram routes.
Bringing new rail services to areas currently not served is an added aim and it follows that areas which are currently zoned for more concentrated development - residential, office, commercial/retail - are favoured station locations.
A job is a job and people need to get them in the most efficient way possible - and in each of the clusters there's a diverse range of them: retail, manufacturing, academic, scientific, professional services, they're all represented and would all become more accessible through overlaying the orbital line on the current public transport network.
Each of the stations on the route map have a footprint which puts their ultimate platform length at 80 metres - or just over 1/3 of the ultimate length of stations being built as part of the Metro tunnel project. This is what Garry Glazebrook alluded to in his piece: this is how civil costs are reduced.
Similarly, 'light metro' trains are generally slimmer around the waist: 2.5-2.7m wide (compared to 3m for our current train classes) which could lead to a more positive civil construction cost scenario because underground tunnels wouldn't need larger clearances for the vehicles to run through them.
This all leads to inevitable offshore comparisons and the one which has been referenced by UM users on previous article comment threads is the Copenhagen Metro. The trains are 39m long, 2.65m wide, weigh 52 tonnes and each three-car train can seat 96 passengers and have 204 passengers standing.
On 5 minute frequencies, that's moving 3,600 passengers per hour, per direction; 4 minute frequencies: 4,500 pphpd; 3 minute frequencies: 6,000 pphpd, 2 minute frequencies: 9,000 pphpd. If the train length were to double - to utilise the ultimate platform length of 80m in each of the stations in the route map, it follows that those passengers per hour, per direction numbers also double (which incidentally is somewhat close to the capacity of some of our current rail lines in peak).
This orbital line is conceived to run trains with those broad parameters.
The Three Phases
Phase 1: Westall - Heidelberg
This appears to be a logical place to start - three of the four major job clusters from the Plan Melbourne job density map lie in the line's path. It's a mixture of both elevated track in the relatively straightforward road grid through Monash and then proceeds underground the for the entire journey 'up and around' to Heidelberg.
Yes, it's a lot of tunneling (42-43km in total across phase 1 and 2) and there's potential to use road medians and elevated structures to carry tracks at a lower civil cost in some places, however I've deliberately gone the full underground route because it's the most direct and likely the best way to integrate with existing and future high-density residential and job centres. Plus if an orbital proposal were to be seriously looked at by Infrastructure Victoria (that's the aim!), they (and we, the public) would probably want to know what the high waterline cost would be.
Westall station on the Dandenong corridor would be integrated by extending the current platforms and the structure would allow people to transfer up to/down from the orbital line platforms that would be elevated on the western side of the Westall Road bridge.
Westall was my choice for beginning the line due to land adjacent to the corridor that could be used for the line's first maintenance and storage depot. The first station after terminus would be Monash-Mulgrave which is placed near the geographic heart of the entire Monash National Employment Cluster (NEC) near the intersection of Blackburn and Wellington Roads.
As well as building the station in this location, major pedestrian 'boulevards' would need to be built to link the Monash University campus, as well as along Wellington Road and Blackburn Road.
The line would transition underground near the intersection of Ferntree Gully Road and Blackburn Road. Blackburn Road's northbound lanes would be realigned after acquiring the properties located adjacent to allow for a drive structure to be built, allowing the smooth transition of the line from the elevated median to underground.
The first underground station is Pinewood - and a candidate for a TBM launch/extraction site. As this station is designed to serve the northern part of the Monash NEC, again, large-scale pedestrian improvements are present to make getting across the Monash Freeway (and Blackburn Road interchange) more efficient.
Now that the line is underground - depending on ground conditions & topography - it is free to link centres in the most direct way possible, as is the theme for the rest of the journey through phase 1 and phase 2.
Mount Waverley would see an underground station and interchange built in its commercial precinct and the line would then hit the first minor job cluster at Deakin University/Burwood. The station at Deaking would utilise the sport grounds for construction (access to the grounds would be restored after construction has completed), its main entrance would face inwards to the campus, but also have a pathway under the eastbound lanes of the Burwood highway to integrate with the main Deakin University tram stop.
Heading north, Box Hill is encountered next. Here, it is assumed the current station (and bus interchange) would be upgraded and I've taken a punt and put the underground station (and platform-to-platform integration infrastructure) to the east of the current covered platforms and put forward the notion of a new direct street entry on the pedestrian mall as well. Carrington Road would also see a secondary entrance.
Manningham would finally get some form of rail service with the line heading further north to Doncaster. I've chosen the Doncaster Shoppingtown Hotel/Pub site for the station (and another possible TBM launch/extraction site). The rationale is that it's an open-air car park rather than apartment buildings or shopping centre facilities. Property acquisition would still be required.
As can be seen by clicking on the footprint on the map, it might be worth turning this station into a JV with Westfield with the main station entrance located above the platforms on the western side of Manningham Road, but linked under the road to an expanded shopping centre.
The line now turns north-westward and the first station encountered is Bulleen. The station would be on acquired property on the northern corner of the Manningham/Thompsons Road intersection - smack bang in the middle of the residential growth zone which is prevalent in the area.
Both Doncaster and Bulleen station would integrate with DART buses towards the city and deeper into the suburbs.
Further west is the phase 1 terminus at Heidelberg which would be a station built under a surface car park at the Austin hospital. Heidelberg on the Hurstbridge line would see its platform extended and an underground interchange would be built between the two sets of platforms.
The tunnels as part of phase 1 would continue to a site in Bellfield on Banksia Street. It's here I've put another potential place to launch and extract TBMs for tunneling - the site is vacant and is zoned residential growth so upon the construction phase nearing completion, there's potential to remediate the site and turn it into another station with an integrated high-density development.
Other 'potential' stations I've added to phase 1 are located in the Dandenong Road median which is primarily there for interchange purposes should any realistic proposal to run express tracks for the Dandenong corridor from Caulfield to Yarraman in the Dandenong Road corridor ever get up.
Box Hill South features a potential station located under the Kingswood College sports grounds - the theme being: these open spaces can be temporarily used for construction and upon completion and can be restored or remediated and it's a mid-point between Box Hill and Deakin. I've also put a potential station to the east of Heidelberg that might have a positive cost-benefit ratio to serve the eastern end of Heidelberg's activities centre and integrate with north-south bus routes on Lower Heidelberg Road/Rosanna Road.
Stations with a red coloured 'M' would be day 1 operations, all potential stations (coloured blue) would be built at a later stage if they were deemed worthy/required.
Phase 2: Heidelberg - Sunshine
I originally modeled the line to stay on a direct westerly route along Banksia Street, under the Darebin Creek and then under Dundas Street - rationale was keep the tunnels as short as possible. The two stations in this section - one at Preston East and Thornbury effectively serve the same catchments as the alternatives. But I put the alternatives in any way - the orange 'alternative' route is about 800m-1km of extra tunneling compared to the more direct route.
The alternative route perhaps better serves the Preston East jobs precinct better as it follows the centre line (Bell St), it's really a toss-up at Thornbury versus Bell station interchanges: Thornbury would be easier to construct (most likely less property acquisition required) and the same amount of development opportunities exist. Bell Station does lie next to parcels of land that Darebin have zoned as PDZ2 - Priority Development Zone.
From this most northerly point of the orbital, the line starts heading south-west towards its ultimate end at Sunshine. The next station site is a 'potential' one at Brunswick East. I've put the question mark on this one (it's a worthy location but perhaps not needed as part of day 1 operations in phase 2) because this area has a very high level of tram service which is both radial and orbital - this section of Moreland Road has regular services that link it with Moreland Station to the west.
At Moreland Station, I've assumed as part of the level crossing removals project, the station will be sunk below the road grade and I've fit the orbital platforms underneath the sunken Upfield line platforms. I've assumed the main entrance to the station will be to Moreland Road and because the platforms intersect (just on different levels), I've assumed the interchange facilities will be straight up and down - escalators and lifts between both sets of platforms.
You might have noticed the section between Moreland and Moonee Ponds is quite 'busy' with stations. I'm not advocating for all three to be built, however depending on if the Airshuttle Australia - Melbourne Airport monorail proposal sees the light of day and depending on the intensity of redevelopment at Moonee Valley there's a fair few strategic decisions to be made: I'm just highlighting them all on the one map.
Firstly the Brunswick West station location is somewhat natural as it integrates with the 55 tram and is surrounded by residential growth zone. I should disclose I live nearby in the area (and no I haven't put a station literally at my doorstep!).
The station located adjacent to the Tullamarine FWY is to facilitate a possible interchange with the Airshuttle Australia Melbourne Airport monorail proposal. The Herald Sun reported on this in October last year and the HUN said at the time the treasurer encouraged the consortium to develop and submit an unsolicited proposal.
I'm not aware of any updates on the Airshuttle Australia story/proposal however if it does see the light of day, this station position on the orbital line would most likely trump the Brunswick West and Moonee Valley (racecourse) stations. If Airshuttle Australia doesn't get up, rule out this station.
The Moonee Valley (racecourse) station has a dual purpose: firstly it's a prime candidate for a TBM launch/extraction site (after the racetrack has been reconfigured, naturally) and secondly, like the Phase 1 launch site at Bellfield, there may be an opportunity to turn it into a station upon completion of both phase 1 and phase 2 tunneling.
Normally I'd rule out a station that would only be used sporadically (when races or events are on at the track) however given the redevelopment site has received master planning approval from Moonee Valley City Council, paving the way for the race track to be reconfigured, a new grandstand to be built along Wilson Street and the redevelopment of the existing grandstand area, I'm more comfortable with putting this station on the backburner rather than completely ruling it out.
As can be seen by how MVCC have zoned central Moonee Ponds, it will grow eastward on both sides of Mount Alexander Road.
At Moonee Ponds station, like at Moreland, I've assumed a future level crossing removal project beyond the initial list of 50 will see a new station sunk below the road grade and the orbital station platforms would be linked underground. I've assumed the main entrance to Moonee Ponds station after a level crossing removal project will be to Puckle St/Holmes Road and not included any other station entrance directly above the orbital rail line platforms.
From here on things get quite straightforward - the line passes under the Maribyrnong Road and the first station encountered is a big one: Highpoint. I've put the footprint under the current Bunning's car park and like at Doncaster, with an underground link to a shopping centre extension. The station would have an entry above the platforms (after sequestering some car park space) and also have an under-road link to the tram stop located on the eastern side of Rosamond Road.
The beeline to the Sunshine terminus (and emerging National Employment Cluster) continues with a relatively straightforward station build at the intersection of Ashley Street and Ballarat Road, again in a car park, and I've called in Braybrook-Maidstone (it's the boundary of the two suburbs).
At Sunshine there's a few things to point out: I'm conscious of the Ballarat line upgrades and eventual Melton electrification projects which reportedly may include quadruplication of the track between Sunshine and Deer Park West. Sunshine was recently rebuilt as part of the Regional Rail Link project but I suspect there's going to be many changes in future - primarily concerned with grade separating the rail lines (i.e electrified metro lines to Sunbury and Melton with the regional lines to Bendigo and Ballarat/Geelong).
With that disclosure, I've put the orbital line's Sunshine terminus station platforms on the western side of the current station on the surface. It's assumed the surface platforms would be integrated with the larger Sunshine station overhead platform access structure and that the tunnel portals would be on the long piece of land that's zoned for rail use (and currently sitting empty) between the grain silos and main station.
The western maintenance and storage depot would be built on this land as well.
The main idea behind this route for phase 2 works like this: there's some 'on-paper' proposals to bring rail to Melbourne Airport - it's part of the wider High Capacity Metro Line project (a future phase, according to the PTV Network Development Plan (2012)) and coupled with private proposals, in some shape or form, Melbourne Airport will be connected to a high capacity/high-frequency transit mode at some point, therefore I ruled out a more westerly (towards the airport) then southerly (towards Sunshine) route from Heidelberg.
Airshuttle Australia monorail line + orbital station at the Tullamarine Freeway would be ideal.
That allowed the focus to become more about removing congestion east-west through all the rat-race routes that dominate Maribyrnong, Moonee Ponds, Brunswick/Coburg. Moonee Ponds is a minor employment centre at the moment - compared to Sunshine, Maribyrnong, Heidelberg, Box Hill or Monash - but I wouldn't write it off entirely as a future jobs growth area, either. Residential development is a major focus at this point but it does have anchor tenants in the area, namely Foxtel.
I guess another point of the orbital line is to replicate - albeit on a much smaller scale - the centre of the city in the sense that if you can locate an office or other place of employment in a hub which has access from four directions (the centre of the city has access from all directions!), you still have great frequent access to the city (and bewyond) on the radial lines, but you make the non-central Melbourne location far more appealing a wider range of people.
I'm under no rose-coloured-glasses illusion that a single orbital rail line will put a stop to the concentration of jobs in Central Melbourne with the likes of Fishermans Bend, Arden, E-Gate and all other centrally located redevelopment zones getting underway, however as land prices climb as more and more of the land is developed closer to the city, these central city-periphery areas like Moonee Ponds & Preston East, if linked properly, might become more competitive for companies to pursue self-contained office/employment accommodation.
I ruled out sending the route via Essendon versus Moonee Ponds mainly because over time, Moonee Ponds will grow more - definitely in residential terms and is in a better position to see a higher degree of jobs density - despite the fact that an interchange at Essendon would fit neatly with the tram network (and its connection to Airport West).
Phase 3: Westall to the Bay (Sandringham or Mordialloc).
This is the part of Melbourne where my own knowledge / experience of / in the area is limited and that's evident in the map having two quite different alternatives.
I first modeled the route as entirely elevated southward from Westall - broadly in line with Westall Road's median - and then sending the line through the green wedge areas (elevated) and then down the Frankston FWY/Arterial road reservation around the jobs centre in Braeside. I saw this as relatively straightforward: keep it all off the ground, open up rail access to Dingley and Braeside and connect with the Frankston line at Mordialloc. Cheap and mostly direct (and therefore quick) to link with the Frankston line.
I then took another look the Plan Melbourne's job map and noted the high density of jobs to the east of Moorabbin, or at least, the map depicts a higher density than in Braeside. The alternative was then added,
The alternative route is essentially the same as the main route down Westall Road however the line bends through the various reservations and follows the Dingley Bypass to the eastern edge of the Kingston/Cochranes Road jobs precinct. It then proceeds underground through the jobs area and then via Southland to a bayside terminus adjacent to the existing Sandringham Station.
The Sandringham route links more existing rail lines & activity centres that are likely to grow over time, but with more tunneling comes more cost (compared to keeping the line on the surface/elevated). The Mordialloc route would be cheaper (12.5km elevated versus 15-16km mix of elevated and underground), would be a better route for people living south of Mordialloc on the Frankston line to use the orbital line to get to Westall/Monash but it avoids the most intense jobs precincts to the east of Moorabbin.
Congratulations, you made it to the end!