Options for a north-south rail tunnel
Given the discussion about a north-south city tunnel on this blog recently, combined with an excellent post on problems with the city loop over at Riccardo’s blog, I thought it might be a good idea to have a closer look at what such a plan might involve.
What’s the plan?
A north-south rail tunnel through the city should really be expected to handle services towards the north, from Doncaster and Upfield (or the airport if I have my way), and towards the south, from St. Kilda. This isn’t a new idea, but it’s certainly a good one. The map below shows the proposed line.
I’ve included two separate route options for the northern section of the line. Option A is based on station locations suggested by Riccardo, and option B shows my thinking for station placing. Ultimately however, the plans are basically the same and would serve exactly the same purpose. The biggest difference is that option A is somewhat shorter than option B (just over 8 km to Flinders St. compared with 10.25 km), largely because a tunnel from Clifton Hill to the eastern freeway would need to be longer than one from Victoria Park. Both plans would require 6 new underground stations to get to Flinders St.
The southern section of the line basically runs under St. Kilda Rd. until Moubray St., where it heads curves westward to St Kilda. and then runs east to terminate under Balaclava station on the Sandringham line. The total length of the line would be 6.875 km and 5 stations would be constructed; at Domain Interchange, Commercial Rd., Fitzroy St., Acland St. and Balaclava station.
How much would it cost?
Using my old friend the NCCCS, I can estimate construction costs for the line. In 2001, the NCCCS costed double track underground railway at $40 million/km and underground stations at $50 million. In 2006 dollars (the most recent full year figures the RBA has on their calculator), that’s $46.07 million for the rail and $57.58 million for the stations. The cost of stations for the project would be in the order of $633.4 million. The cost of building rail would be $685.3 million for option A and $788.9 million for option B. Total costs would be in the order of $1318.7 million for option A and $1422.3 million for option B. Obviously, option B is not $100 million better than option A, so option A seems to be the better plan. I would however suggest placing Fitzroy station in Johnston St. between Brunswick and Nicholson, to maximise integration with route 96 as well as 112. Bundled with the $495.2 million Doncaster line and the $127 million airport via Upfield line, total cost of the project would be $2044.5 million.
Interestingly, the $100 million saving would more than pay for the construction of a tram line between Clifton Hill and North Melbourne, which would more than cover the catchment area of an Alexandra Pde. station. The 4 km line would only cost $43.5 million. Here’s what it would look like:
The bottom line
Basically, $2 billion buys us a Doncaster line, an airport line and an inner city metro line. This provides for additional growth, not only on the newly constructed lines, but on existing lines as well. Integration of the rail system would be considerably improved. When the state government is spending $3 billion on EastLink, and possibly even more for an inner northern freeway tunnel (we’ll see what Eddington says!), $2 billion to massively increase the usefulness and efficiency of the rail system shouldn’t seem like such a big outlay…
Filed under: trains
This is good.
How long would a train journey from Flinders St to the stop on Alexandra Pde/Johnston St take?
Most suggestions I’ve seen for a Domain tunnel have had the line join Sandringham or Dandenong at South Yarra. This is, I gather, meant to help reduce capacity constraints in the loop, and to allow through running. In (my understanding of) your proposal, only the Upfield line is taken out of the loop—there’s no change to the Caulfield group—and trains can’t continue south.
Any suggestions about what we do with the bypassed bit of the Upfield line? Just completely get rid of it? If I recall correctly, the Melbourne City Council proposal to the East-West Link inquiry would have it converted to light rail, but with route 55 going through Royal Park already, it doesn’t strike me as worthwhile.
This is certainly an ambitious proposal Phin. I suppose it would have to be completed in stages? e.g. Flinders St-Jewell, University-Doncaster (that could be 2), Flinders St-St Kilda, etc…
How do you see it tying in with the recent proposal for a Footscray-South Yarra Tunnel? Or do you not think that worthwhile? That prosposal would use the same alignment from the University to the Domain, and I really feel that a short connecting tunnel to South Yarra from there could take a lot of pressure off the Caulfield group. I suppose I could take or leave the tunnel from Footscray-University if it came to it though…
Interested to see your Clifton Hill-North Melbourne Tram line. I’ve been advocating a Johnston St line for a while now, running from the University to Kew Junction (in purely track terms, obviously operations could extend further). Perhaps a train line makes this redundant though…
The North Melbourne station link seems an obvious one, and another in the area you’re looking at could be Clifton Hill-Moonee Ponds, via Rushall Station, Holden St, Sydney Rd, Dawson/Dean Sts, ending in Puckle St Moonee Ponds. A few of those have track already, or have had in the past.
When it comes to the Macaulay-Royal Park section, I think we can all agree it’s no real loss, but I’d suggest mothballing it rather than closing entirely…you just never know when it might be useful. Some of the lines closed in this city seemed useless, but we’d probably use them more if they were around today: Kew, the Outer Circle, the Inner Circle, Mont Park…even the old Fitzroy spur could have been the start of a tunnel N-S under Fitzroy and Carlton into the city.
Ah, the possibilities in the magical land of Political Support and Unlimited Funding for Public Transport
Thanks for the replies everyone - much appreciated!
Jen, I’d hope it would take a lot less than 10 minutes to get to Johnston St From Flinders St.
Alexander and Yoghurt, you raise some interesting issues about Caulfield group capacity and what to do with the Upfield line between North Melbourne and Jewell.
On the topic of the Caulfield group lines, I tend to think that extra tunnelling at the city end is not the answer. The northern loop faces greater pressures (I’m pretty sure it’s at capacity already) and Sandringham already runs direct during the week. That leaves only Frankston and Dandenong using Caulfield loop. If we get 2 minute headways in the loop, that leaves us with Frankston and Dandenong each getting a train every 4 minutes.
To have a serious crack at fixing up the Caulfield group, we need to do 2 things:
Firstly, fix up the junction at Caulfield itself. This is a serious bottleneck that wrecks capacity on these lines. It needs to be grade separated straight away.
Secondly, start running proper 2-tier services. These lines are long enough to run the long distance trains (Frankston, Pakenham, Cranbourne) as loop limited express services. Then run shorter distance (say to Moorabbin or Cheltenham and Oakleigh or Springvale) stopping all stations direct to Flinders St. and possibly through route them with western suburbs trains. This would require infrastructure improvements - like 4 tracks on the Dandenong line beyond Caulfield, but would be better value for money than running extra tunnels beside a 4 track line.
On what to do with North Melbourne - Jewell, my original plan was to close it. However, it was suggested to me that converting it to light rail and extending it along Park St. to Rushall along the inner circle alignment could provide a better cross town link than my tram proposal. The problem is then what to do with the southern end of the line. I thought that cutting it at Royal Park station and using the 55 track to Flemington Rd. then the 57 track to the city (freeing the 57 to run direct along Flemington Rd.) might be ok. The problem is that North Melbourne station is bypassed.
It’s nothing a blank cheque wouldn’t fix though!
Good stuff Phin
Jewell to North Melbourne is a difficult section. Royal Park is basically a tourist station (look how little of it is near housing) and there is excellent access to the station and area by tram.
Flemo bridge is well used but does have 2 tram routes as well. Macaulay is poorly used. The Upfield paths could be used by other trains including Broadies.
I like the tram idea - cut across to Clifton Hill down the old Inner Circle. Have the tram start at North Melbourne, but not the rail easement but up on Abbotsford St, go under Dynon Rd and down to the Arden St level crossing, then from there. Not sure where to terminate at CH - a bit of a disaster area if you ask me public transport wise.
**********************
The simplest way of freeing Caulfield capacity is to take the Sandys off the system and run via a new line, then having the Caulfields take their space at FSS and SXS. Run the Willys through to Caulfield as they did before. The further south you take the Sandys off the existing line (eg from Prahran underground to Domain Rd, or from Windsor to ST K Junction then north) the more you get to keep for the Caulfields (eg you get the existing 6 tracks from SY).
While running one of the other lines eg Pak or Frank, via a new underground has some appeal, I would probably suggest KISS on this occasion.
As I mentioned elsewhere, the attraction of pairing Sandy with Upfield is they are both isolated lines like the Clarkson and Mandurah lines in Perth. You could therefore start modifying the lines with equipment you can’t afford to implement across the system - eg standard gauge, 9 car trains, 1.5 minute frequency signalling, or articulated trains with very high axle loads.
Re Caulfield, I have given some thought to whether you would actually fix this site, or make Caulfield and Malvern joint junction stations - using the HK model (Prince Edward, Yau Ma Tei and Mong Kok) with the famous three station junction.
You would do this by putting the dives between the 2 stations. Express trains going inbound (from Dandy and Frankston) would stop at Malvern but not Caulfield, and pax would walk across the centre platform.
Express trains going outward would skip Malvern, and go to Caulfield. Pax would cross the centre platform of Caulfield, to connect with trains that were on the outer platforms at Malvern but dived under the other pair of tracks.
I’ll give this idea some more thought.
Riccardo - is there really any point to converting the suburban system to SG? Especially the isolated lines on which freight is never going to be carried. (Not that that’s likely for a section that’s in metro use anyhow). There’s no need for compatibility!
Surely it’s not that expensive to get wider bogies compared to resleepering the entire length of the lines (although you could just lay dual-gauge sleepers instead of regular during normal maintenance). I agree with the longer trains, signalling etc, why not? Certainly any new stations should be built with this in mind too.
Agree Dave, no need for SG - just an example. In fact, the more isolated the line is, the less need for conversion.
Although it might be worth checking with the manufacturers - if I had to order 200 cars from an overseas supplier (piggyback on someone elses order) and get both lines resleepered at the same time, probably just as easy to convert.
But definitely no need to do it. The 9 car idea is a no-brainer. All your new stations can be 9 car (and should be if you are expecting crowds, the 6 car length is too small) Alternatively, do 8 cars with current stock weights, so that you don’t need extra power cars (ie equivalent to 4 motor and 4 trailer cars).
For existing stations, you can easily tack on extra length in most places, and when you do a full rebuild you can do it properly.
If you are going to do a full resignal, then you would definitely do the 9 or 8 cars at the same time.
Thanks Riccardo and Dave. Riccardo, This joint junction using Malvern and Caulfield idea sounds interesting. Part of the problem is that many in the public service dismiss these sorts of ideas off hand because it’s not the way Melbourne has been doing things for the past 100 years. This sort of ‘it’s the way we’ve always done it - why change?’ attitude is arguably the main reason we still have a commuter rail system when we need a metro.
I haven’t thought much about 9 car running in Melbourne, but it would be sensible to start undertaking these sorts of projects on isolated lines. The north-south rail tunnel, Doncaster and Upfield/airport would be perfect starting points. There’s no excuse not to build the line with 90 second signal spacing either.
cheers,
Phin
Here it is the Prince Edward/Yau Ma Tei/Mong Kok trick, done at Caulfield and Malvern
http://img266.imageshack.us/img266/858/caulmalix6.jpg
Phin & Riccardo, on the subject of the public service dismissing this sort of idea, and somewhat off the subject of the north-south rail tunnel (sorry!), is there some sort of plan for doing something with these ideas? These, and your other, ideas look great, but in reality they’re depressing because while they could vastly improve the efficiency or use of public transport I fear they will never leave your blogs.
I don’t mean to criticise you — far from it! — but I think the best plan to develop is a plan to get these from here to there.
(Another plan I think could work wonders is to redraw the public transport maps. Zone 1 maps need to show both trams and trains — not just points where the systems connect — so interconnections can become more obvious, and tram maps are currently completely illegible. My recollection is they used to be much more useful when shown in full color.)
Alexander - the method in my madness is just to put the ideas out there.
Rail is not just badly governed - but also badly ‘lobbied’ or advocated for.
Look at the spats with the PTUA, Angelicorail, the Eastern Transport Coalition and so on - vs the RACV speaking with a single, clear voice.
You don’t see a rebel RACV splinter group set up in say Pakenham saying “RACV don’t understand blah blah blah” the way you see with rail advocacy.
Also too many dribbling gunzels, who don’t want rail progress because their favorite piece of rollingstock or station would get the can.
I will stand here like a lighthouse, pointing the way to the future and clear of the hazards of the present.
I might contribute to a public consultation, if there is one.
[...] by Phin under trains, trams A couple of weeks ago I had a look at the prospects for a north-south rail line in central Melbourne. There’s no doubt that such a line would have a substantial impact on [...]
[...] really like the idea of a north-south tunnel running under Swanston St. and St. Kilda Rd. Indeed, I posted on this issue, as well what to do about the trams, a while back. However, I planned on sending the [...]
[...] the Northern Central City Corridor Study figures I’ve been using for costing projects like a north-south tunnel and airport [...]
[...] up the Eddington series with a look at the prospects for a north south rail tunnel, but I’d already looked at the issue pre-Eddington and my ideas hadn’t changed. Moreover, we had some good discussion on the [...]