Follow us on Twitter

Keeping it lean and mean: getting the maximum ‘bang for buck’ from the CBD tunnel (Part 1)

As a long time advocate of public transport in Auckland I feel a certain sort of glee at the huge groundswell of support for a CBD rail tunnel in Auckland. With the likes of our new mayor Len Brown pushing hard it seems we will soon have unstoppable political momentum for the project. After a hundred years or so of waiting it seems Auckland will kick start the 21st century with a city rail tunnel.

The benefits of this project are many: new underground stations within walking distance of anywhere in the CBD, a faster route for the western line, improved interchange opportunities and of course relief of the capacity constraints at the current Britomart terminus of the rail system. Another oft-overlooked benefit is one of ‘prestige building’, a process that has already started with Britomart. Once we have new shiny new trains zipping through flash new underground stations at speeds of over 100km/h, the old perception that public transport is for ‘losers, cripples and dole bludgers’ will be left behind faster than a driver sitting in gridlock on the motorway.

Taken together these all mean much faster door to door trips, far better service frequencies and greatly improved reliability across the network, which all leads to massively improved desirability. With a CBD tunnel the rail system won’t merely offer peak hour travel times that are competitive with the car, it will blow them out of the water. In short, patronage on the rail system will skyrocket once the CBD tunnel is built.

This however leads us to one very important question: Will it be enough?

This question has been an ongoing debate on the CBT forums for years now. Will extending a two track CBD tunnel from Britomart to the Western Line provide enough capacity to meet the growth it is bound to cause? Should be we aiming to spend more and build it bigger to start with?

On one hand we have the ‘do it right the first time’ camp. These guys point to past planning failures like the four lane harbour bridge, and suggest that we should double or triple the budget for the city tunnel to ensure it has at least four tracks and decades of room for adding new lines and growing patronage. On the other hand we have a ‘step by step’ camp, who see the CBD tunnel as relatively affordable and realistic option that can be added to in the future to meet growth, perhaps with new links and additional tunnels.

Personally I lean towards the second camp. I certainly would love to see a four track mega tunnel under Auckland central but to be realistic we are lucky to see even the proposed two track one funded and built. The idea of doubling the budget to three or four billion is simply ludicrous given the political climate, plus there isn’t any need to build everything the city will ever need in the first instance. In fact to spend a lot of money to build infrastructure that isn’t actually going to be used for ten or twenty years is akin to setting money on fire. Ten years of compounded interest payments on a billion dollar loan for an empty pair of tracks is a disastrous waste of public funds.

Next train departs in two minutes

However, the point about needing to meet the growth in patronage still stands, so the answer needs to be somewhere in between. To be realistic we just can’t afford anything more than a two track tunnel, so we need to maximise the benefits of those two tracks to the very last drop. I’ve previously posted a concept of having the new tunnel bypass Britomart rather than join to it, and extend all the way to Quay Park junction. This way the entire capacity of the new tunnel would be in addition to the existing capacity of Britomart, and rather than mixing all trains in Britomart we get to keep a terminal station separate to the new metro style one.

I still think this would give a very good outcome for a relatively small extension to the project, but at the end of the day it would still require several hundred million dollars more funding. In the current climate this option seems pretty unrealistic. So with this in mind I have gone back to the drawing board to see what could be done very cheaply to maximise the capacity of the proposed CBD tunnel once Britomart is one of its stations. Based on the idea that we will only get a two track city tunnel, here are a few ideas that might make the tunnel as lean and mean as it can be:

1) Maximise frequencies of trains through the CBD tunnel

Designing the tunnel for very fast headways is a great way to maximise the capacity of the system without the need to build extra tracks or lines at great expense.

Right now the best a rail line can achieve in Auckland is about twenty trains an hour each way and at this level the CBD tunnel could support forty trains an hour in total. That’s not bad compared to the current capacity, but this tunnel is going to be the backbone of the public transport network for many decades to come so the more train capacity the better. Luckily the recent CBD tunnel report suggests that with the right signalling system the same tunnel could support thirty trains an hour each way, or one every two minutes. Basically this means a 50% increase in the capacity of the entire system from a small additional expenditure on superior signalling. Even if they install regular signals to begin with a little attention to the design of the tunnel would allow these to be retrofitted once they are needed.

2) Grade separate Quay Park junction

Flat junction (left) has inefficient conflicts between trains. Grade separation with a flyover (right) removes the conflict.

Sixty trains an hour through the tunnel would be fantastic, but this wouldn’t be much use as long as the flat junction at the start of the tunnel can only handle half that.

Right now Quay Park junction is a bit like a crossroads with traffic lights. If traffic coming down from Newmarket has the green light then traffic coming from Orakei had to sit stopped at a red light. Likewise once Orakei traffic gets the green, the trains from Newmarket have to stop and wait until the track is clear for their turn.This is actually the weakest point on the network and we will never get much more than 20 trains an hour each way with a flat junction.

The answer is to grade separate, which is the rail equivalent of turning a traffic light intersection into a motorway interchange. As Quay Park is like merge between two motorways heading the same way, this is actually pretty simple: it would only need one track elevated over two other to remove all conflicting connections. A simple viaduct could carry one track heading out of the Britomart throat tunnel, up over the Eastern Line tracks and then curve around to join the line heading to Newmarket again. This would be essential once the CBD tunnel is built, but it would also help get a couple of extra trains an hour into Britomart in the mean time. They should really look into doing this cheap viaduct now.

3) Level boarding on CBD tunnel stations

Having the platforms of the CBD tunnel stations at the same level as the train floor would be a very simple way to boost efficiency. Level boarding makes it a lot faster for people to board en masse as there are no steps to negotiate or trip over, and no unwieldy gap to leap. Furthermore it would allow people with bikes, prams or luggage, or those using wheelchairs or mobility scooters to board themselves rapidly with a minimum of fuss. They current wheelchair system where the train driver has to park the train, hop out of the cab and muck around with a folding ramp simply will not cut it in the city tunnel with super frequent trains.

From what I understand perfectly level boarding is not possible on some suburban platforms due to issues with freight train clearance, as full height platforms would foul the wagons as they passed through the station. But in the tunnel there will be nothing but shiny new electric multiple units, so there no reason why the platform edge can’t line up with the floor of the carriage to the millimetre. This is one solution that will cost nothing to implement, it just means the new station platforms need to be designed accordingly.

4) Lighting and identification

A very simple trick to boost efficiency would be to have lighting in the tunnel on the approach to each station. If this is designed such that the light appears about ten seconds before the train makes it to the station, it would provide an effective cue for people to make their way to the doors to be ready to disembark quickly and easily.

A further application would be to have these lights on only one side of the tunnel when approaching underground stations, so it signals which side of the train the platform will be on. People getting ready to exit the train will instinctively turn toward the source of light. Guessing which side the platform is on can be confusing, and in the case of people with prams, bikes or luggage, getting it wrong can end up delaying the whole train. A little bit of attention to the lighting would cost basically nothing, but add to the efficiency.

Similarly, having a colour theme for each city station will help people navigate and work out where they are. For example Britomart might be lit in cool blue, midtown a warm yellow, Newton in Green and K Rd could be the, er, red light district. Little touches like this cost basically nothing, but could improve the ease and efficiency of the system to no end.

To be continued in part 2

31 comments to Keeping it lean and mean: getting the maximum ‘bang for buck’ from the CBD tunnel (Part 1)

  • 5n0re

    “gridlock on the motorway”


  • Alex

    Would this be worth a presentation to the Council and/or AT, to get the maximum benefit from the tunnel when we do get it?

    Also how feasible would it be if the tunnel is built as planned, to add in the parallel approach under Quay street at a later date, to maximize the use of the tunnel?

    • I’m no engineer, but I figure it would be pretty infeasible to do that. The design constraints are being pushed to the limit as it is to get the tunnel down the hill under a road corridor, around a ninety degree bend then under a proposed skyscraper and through the foundations of the historic post office building. Linking in a second tunnel after the fact to run wide out to Quay St might just be impossible. They would really have to design it in from the start.

  • john

    Add: carriages with three doors (six channels) a side to minimise dwell time.

    • I agree wholehartedly, but unfortunately the tender for Auckland’s EMUs has already gone out with only two doors in the specification. I guess the are looking to maximise seat numbers at the expense of standing and circulation room, which is a bit foolish in my opinion.

      • Matt L

        It depends on the outcome of the tender, perhaps some of the companies have submitted designs with 3 doors per car (unlikely but not out of the question)

  • This is great- yes, get trains that have 3 doors, you will increase capacity that way by reducing dwell times.
    Make sure that you have platforms that are long enough to handle 6 car trains.
    Get a good signalling system down there

    To get the project through, you need to get the foot in the door. Reduce the cost. Cut out the motorway tunnels, they are a financial poison pill designed to inflate the cost of the project package as a whole and increase the chance of rejection.

    Try and keep the lines apart- sectorisation, one line, one track if you can. Auckland really could carry 100 million trips on rail no problem IMHO if they continue with the upgrades!

    Keep it up!

    • Thanks Bris, do bear in mind these points are about the central city underground tunnel. It is the proposed harbour tunnel that has motorway tubes as well as rail tubes, not the CBD tunnel.

      Sectorisation would be great, but not possible in Auckland unless the built a four track tunnel or two twin track tunnels. However there will only be four primary lines in the foreseeable future, so it will be a case of ‘two lines, one track’ which isn’t so bad… and they can always build that second tunnel in the future right!

      • The motorway tunnel connects into these though?
        And what about integration with bus? I know the buses are on the surface, but would there be scope for a busway station above the rail station. Brisbane has something like this with its Roma St rail station Busway. The buses pull up and are just an escalator ride away from the main station concourse.

        • No, thankfully motorways have nothing to do with this particular tunnel, although it would have to pass under a motorway at depth on one part of the route.

          These is a street level bus terminal in and around Britomart at the moment (and perhaps that could be improved) but I think the best place for a new bus interchange is at the Aotea midtown station. This is proposed to be built underneath Albert St which happens to be one of the busiest bus corridors in the CBD. The station will be fairly deep under the road, but my thinking is if they build it by cut and cover from the surface then they could very easily build in a Brisbane style bus station on the first level underground with a pair of ramps at each end to connect to the road. So it would be something like:

          Street level: Top entrances from Albert St
          Level 1: Bus station platforms and waiting area
          Level 2: Rail station concourse and retail, side entrances from Queen/Darby St
          Level 3: Rail platforms

          Interchange is only an escalator away.

        • James B

          No the motorway tunnel is seperate connecting the Northcote Point/Onewa to somewhere in St Mary’s Bay.

    • Matt L

      There are no motorway tunnels as part of the CBD tunnel project, I think you are confusing it with the next harbour crossing.

      All platforms on the network will be able to handle 6 car trains from the end of this year I think so no issues there and the CBD tunnel study recommended 170m long platforms or enough for 7 cars which I assume is to make it easier for the drivers to position the trains therefore make stops faster.

  • Matt L

    Just a few points

    1. The most commonly suggested routing pattern on here is from West heads down the tunnel to Britomart and then out South and Manukau then up the Eastern to Britomart and through the tunnel then out to Newmarket and on to Onehunga (and eventually the Airport). The advanced signalling allows each of these two routes to have 4 minute frequencies each way which is something that would be amazing to see and do wonders for patronage.

    2. Grade separation would be needed to get those high frequencies through but it could be done so only one track needs to be crossed however if it was done now it might actually cause more problems than it solved. The new signalling system that has just been installed allows for bi-directional running of trains which means trains can easily travel each way on each track, I have already seen this used and two trains were travelling down the tunnel in the same direction at the same time. In addition to this they have added in a double slip where the yellow and green lines cross in your diagram, this means a train can travel in the ‘wrong’ direction up the yellow line and where it crosses, move on to the green line again I have been on a number of trains that have done this. These improvements have given a lot more flexibility and means that if we grade separated the junction now, we would force all trains to move across to the ‘right’ track in the tunnel to get to/from their platforms which isn’t as flexible.
    Another issue is we would also need to grade separate Newmarket which might not be as easy due to how tight it is for space and the grades on all sides of it.

    3. The new EMU’s will have level boarding on at least one of the 3 carriages, most likely the middle one, if we made the CBD tunnel stations level with the rest of the train then the low flow section will become hard to get into and out of. It would be better to keep it the same height as the rest of the network and in fact I believe that the plan is to raise the track height in Britomart as part of electrification as its platforms are currently about 100mm higher than the rest of the network due to the fact there will never be a freight wagon through there.

    4. Lighting and colouring is a good suggestion. A couple more suggestions: I think colours should actually apply to the lines themselves so the west/south trains might good just to call the Green Line with East/Onehunga trains called the Yellow Line. Back lighting on the walls of the station, like what we see at Britomart behind the advertising, could be used but change colour depending on which train is coming next. That way if you are standing at midtown and want to head out west you stand on the side of the platform that heads that way and the colour will indicate which train to get on, it would also be easier for the station announcers (even if its automated).

    Another suggestion would be to have 3 platforms per station, the idea being that when a train pulls up anyone wanting to get off does so on the outer platform and anyone wanting to get on does so from the centre platform. The reason for this is so that if there is a huge group of people trying to get on you don’t have to fight your way through the crowds which will speed things up for everyone.

    We also need to make sure that the stations have lots of access points off the platform as already at Britomart the escalators get queues forming when full trains arrive. Also more entry and exit points to the stations themselves will help to disperse people better.

    • Cheers Matt,
      On grade separation, bear in mind that once the CBD tunnel in built ‘up’ trains will only be able to use platform 1, and ‘down’ trains will only be able to use platform 5. So after conversion to a through station the bidirectional running into Britomart will be more or less futile. I’ve proposed a way to work around this which is outlined in part two of this post. And I do have one concept for a way to fudge grade separation at Newmarket despite the space constraints which i’ll stick in a later post too.

      Good point on the EMUs having effectively two boarding heights, I guess that means they tunnel platforms will need to be lower than the floor after all. Such a shame we are stuck with that just because of one model of freight wagon.

      One thing they could do with lighting is to colour each side of the station according to the respective line using that side of the platform, and even change the lighting colour to suit that of the next train. Say your Western line train is the ‘red’ line, so a minute before it arrives that side of the station glows red, and stays that way until the train is pulling away.

      As for the three platform idea, you read my mind… I’ve covered that in part two also, which is soon to come.

      • Matt L

        My grade separation comment was in relation to your suggestion of doing it now rather than wait till the CBD tunnel, it should be done as part of the overall project however.

        The lighting you mention is exactly was I was referring to, light up the side of the station as the same colour as the line the next train on that side is travelling on. LED lighting would make this very easy to do

        • Right I follow you, that is a good point. Having the flyover before the CBD tunnel might just interfere with the bidirectional running… although if the flyover was signalled both way also it could add flexibility. At the least it would give them an extra track to park trains on while waiting for a slot!

  • tochigi

    wow, the idea of having a modern, urban, electric, passenger rail network with platforms lower than the carriage floors is mind boggling. there is no railway here in Japan that has step boarding, even out in the wopwops. can they not build bypass tracks for freight trains? i.e. an extra track outside of the platform track only for the length of the stations.

    • This is odd. Brisbane has platforms with 2 heights at many stations.
      At Roma Street, there is a standard height platform, but this gradually increases at height at one end so that it is much higher.

      Step free boarding is really important. Otherwise you will need to have 2 people on each train- one to drive it and one to get out and pull ramps etc, you effectively double your labor cost, when you could be running trains at higher frequency.

  • I wonder how much extra it would cost to ensure the platforms could fit 9 car trains. The EMUs are coming in three car long sets so being able to tie three of them together would create a train with massive capacity.

    • Brisbane is thinking about 9 car trains for the Cross River Rail Project (massive 9km tunnels under the CBD and Brisbane River) and these would be helpful for peak hour, when passengers fill up the train.

      Higher frequency is probably the better strategy, so future capacity expansion options would be reasonable to look at. For the record, our busways are light-rail convertible if the need arises, so it would be a good idea to look into possibilities for capacity expansion in the future. Famously, Sydney has stub tunnels in some of its central city stations to allow future expansion, and these were first thought about in the 1900s.

    • Matt L

      It would be ideal but I can’t see it happening unfortunately due to typical poor planning in the city. The other thing is the grades are already extremely steep so to get longer platforms the tunnel might need to be quite a bit longer and therefore more expensive.

    • 9 car trains would be good, Melbourne is planning them for its new metro tunnel too. However that basically means the flat station sections of the three CBD tunnel stops would need to be about 70m longer each, or around 200m in total, which might be very difficult given the grade constraints. An extra 200m of flat sections translates into an extra six or seven metres of height that would need to be made up somewhere.

      Of course if they were building platforms under Quay St instead of connecting to Britomart there would be plenty of room for this ;)

  • Lukei1

    So at this stage are we conceding that the tunnel will be a continuation from Britomart platforms, as opposed to a parallel structure?

    • Yeah, I’ve come to the reluctant conclusion that the tunnel will be extended from the existing platforms at Britomart… so the way forward from here is to ensure that the tunnel can operate to full capacity while also retaining some capacity to terminate regional and intercity trains at Britomart.

      • We have Roma Street where we terminate interstate trains. Trains may run through, but not stop at, Central.
        I think it might be the same down in Melbourne for the Southern Cross station as well. Sydney has extra platforms dedicated for terminating trains as well.

        • Yes, both Southern Cross and Sydney central have a set of terminating platforms on one side and a set of through platforms connecting to their respective underground city loops. This is what Britomart should aim to be too, a terminus for trains from out of Auckland as well as having platforms for frequent suburban services.
          The design of Britomart means only the two outer platforms could be feasibly be used for CBD tunnel trains, and as the CBD tunnel will only be two tracks there really isn’t and need for more than two platforms anyway. So this leaves us the central three platforms to use for the terminus, which should be sufficient for quite some time. The only thing is to ensure the tracks, signalling and operating patterns are designed to allow this to happen. Unfortunately so far it seems they are planning around Britomart being just a through station and the central three platforms being effectively abandoned.

  • Yes, hopefully there will be public consultation so they can pick up on these things.

    Just a question to the Aucklanders- why not rename Britomart to Auckland Central?
    Much more legible for tourists and people alike. Where did the name Britomart come from anyway?

    • It comes from “Point Britomart” which was a headland around that area. The future “Midtown/Aotea” station would be more central anyway.

      • And that point was the place where Auckland was founded as a city, and the point itself was named after the first colonist ship to the area. It has quite a lot of history to it. If anything “Auckland Downtown” would be more appropriate, otherwise we’d end up with the Melbourne situation of having a station called central that isn’t central!

    • ingolfson

      “Yes, hopefully there will be public consultation so they can pick up on these things.”

      I am not a resource consent planner, but I see no way in heck how this could NOT be publicly notified. NZ law pretty much requires it, with so many people affected. So you will get your chance to submit, and if your submission is written concisely, they will respond to your issue (and hopefully pick up all those minor tweaks like the type suggested here in this post).

Leave a Reply