Auckland recently passed 16 million annual rail boardings, quite a momentous achievement and in some respects the culmination of 10 years of effort: the implementation of the 2006 Rail Development Plan. At the heart of this plan, and in fact many of Auckland’s transport strategies over the past decade, was the creation of a true “rapid transit network”. Improvements to the rail network and construction of the Northern Busway during this time has meant that Auckland has gone from having no rapid transit network to one that extends out in most major directions from the city. It is improvements to the rapid transit network that have driven almost half of Auckland’s public transport ridership growth over the past 10 years.

Further developing the rapid transit network still rightly sits at the heart of Auckland’s transport strategies and formed the basis for our “Congestion Free Network” concept. There’s a remarkable similarity between the two actually. Firstly, the most recent official rapid transit network plan out to 2045:

AT Rapid Transit Network 2015-2045

And the Congestion Free Network:


Since our work on the CFN (which is actually coming up on three years old), Auckland Transport has become interested in light-rail as a way of resolving city centre bus congestion and providing a high quality public transport option to the central isthmus area. However, these are not dinky trams, but rather big multi-articulated vehicles capable of each carrying up to 400-500 people and travelling up to 80 kph. These are serious people-moving machines and sound almost identical to what’s now being built in Sydney:


The use of such high-capacity and potentially high-speed light-rail vehicles that also have the capability of travelling safety at street level along some of our major arterial roads creates a really interesting question: are these “rapid transit”? In a pure sense they are not as rapid transit is meant to be, fully grade separated from other traffic while our light-rail will need to stop at traffic lights (but presumably have some pretty amazing signal priority). But the same applies for the Northern Busway currently and the proposed AMETI busway – both of which are considered rapid transit.

AMETI Buslane - Pakuranga Rd

The speed and capacity of these vehicles means that they could potentially be used on some of the other proposed rapid transit corridors without the need for such expensive tunnelling that heavy rail requires, or the terminal capacity issues that bus based systems seem to create. In other words, once you’ve got a bit of light-rail in Auckland, it becomes a no-brainer to think about where it might be used further and if it might be an appropriate solution for some “pure rapid transit” corridors. We’ve already seen AT suggest light rail might be a viable solution for access to the airport.

Other options include to the North Shore – perhaps even linked to an airport route, a long-term replacement of the AMETI busway and maybe even out to the Northwest?


Of course it’s easy to get excited and a bit mode obsessed about something like light-rail and one of the most important rules when it comes to good public transport planning is to think about the characteristics of the corridor you’re looking at before jumping to a conclusion about the mode to use. But the ability of light-rail to be both a very high capacity “true” rapid transit system but also something that can run at street level provides us with a pretty powerful tool that may be able to deliver very high quality rapid transit along a couple of key corridors that heavy rail options have struggled to stick.

North Shore

The Harbour Bridge is Auckland’s single busiest bus corridor, with more people entering the city centre from the North Shore at peak times than arrive at Britomart station by train (although this gap has probably shrunk due to recent rail ridership growth). This is shown quite dramatically in this image from an upcoming AT presentation to the Council.

Bus Passenger Volumes to City

While the Northern Busway has been hugely successful, it’s hard to see how the city centre will cope with the rising number of buses from the North Shore – even with the City Rail Link and a Dominion Road light-rail route taking buses from other areas out of the city. Significant planned growth at Dairy Flat appears likely to only add to the need to essentially ‘upgrade’ the busway at some point in the next 10-20 years.

Yet heavy rail options for doing this have always been a bit messy. How do you deal with the grades? How do you “hook into” the existing rail system at the city end? Where on the North Shore do you go – just up next to the motorway with limited land-use transformation opportunities? Do you close the busway for years while you rebuild it for heavy rail? Do you go the whole hog and do a proper underground metro? But that seems to come with a $10+ billion price tag.

If sufficient capacity can be provided, light-rail is potentially a solution to a number of these issues. The “tie in” at the city end can work well by just connecting a cross-harbour tunnel into the end of the isthmus light-rail scheme which is proposed to end at Wynyard Quarter. That creates the potential for some fantastic one seat rides from Albany through the city to the Dominion Road corridor. Light-rail could also mean some branch lines off the main SH1 are possible in a way that just wouldn’t work for heavy rail. What about branches to Takapuna, Browns Bay or up Onewa Road to Birkenhead?


AMETI and Northwest

In both cases AT will be building brand new dedicated PT infrastructure where almost none exists today. If light rail is successful on other routes, could it be worth going straight to light rail in these two “rapid transit” situations too. Doing so could both (not necessarily together) save costs later on from having to convert busways to higher capacity modes later and help encourage more people in the area to use PT to get around.

Of course the biggest issue in all of this is the cost. To build a Light Rail network like discussed would be significant – especially getting over/under the harbour and these projects would need to be prioritised just like everything else. There may also be a situation where depending on the costs and funding, a busway now is better than a light rail line in 15-20 years time.

Interestingly since I wrote this post a few days ago, council’s Development Committee agenda has been released with an interesting presentation on light rail from AT. It includes this map showing almost exactly what I’ve suggested above.

Auckland Potential Light Rail Network

Share this


  1. I think AT are trolling you guys now with their LRT plans. “Whatever you bloggers come up with, I can do better!”

    What is weird though is how 2-3 years ago AT was super hostile to LRT and now they want to put it everywhere. That’s a bit weird.

  2. Yes, LRT does offer a sort of ‘surface Rapid Transit’ but much depends on the details. If the design becomes excessively focussed on speed then it risks passing communities by, as one way to speed a service is to reduce the stop frequency. And conversely if the design takes advantage of its great ability to arrive right at the front door of major destinations then it will likely become stuck in traffic and stops and cease to be Rapid. But this flexibility can indeed be a huge strength and in the designs we’ve so far seen for the Dom Rd/Queen St route [LRT I] it looks like they have the balance right for these dense urban environments.

    Then flipping it across to the Shore and up the Busway it clearly becomes a purely Rapid service.

    In many ways the need to deliver hybrid Transit models is because we keep dispersing the city so much. Denser inner neighbourhoods require access more than speed, while for the far flung places speed is the premium because they are just so far from important destinations.

    1. Great post.
      This needs to be restated time and time again to NZTA and the government together with the necessary reform to transport funding, that the latter really must make, and the sooner the better. As transport strategists the governments’ thinking and actions have been woeful.

    2. Patrick
      Surely intensification is a means not an end; if transport technology were of sufficient sophistication to allow transit times equivalent to close-proximity walking, wouldn’t dispersal be better?

      Imagine, if you will, stargate technology (deliberately chosen as absurd and perfect). If each house could have its own stargate/transporter, then why would proximity be of use?

      Now, let’s pull it back a little: if Auckland had a much more intense PT system – trains, buses, trams, and ferries – then dispersal wouldn’t be a problem.

      1. No. Why is dispersal better? Simply more energy and time spent connecting. Anyway the Nirvana you describe is not observable anywhere; everywhere there is lavish connectivity density increases. It is unaffordable to have abundant access over a huge area with a small population; just not geometrically possible, as more distance and more different routes would be required but would have to be funded by a smaller number of people.

        And of course intensification is a means not an end, the end is to create a highly productive, highly liveable, diverse and happy-making place for more.

  3. I think heavy rail still needs to be on the cards and considered as well.

    As there is serious consideration to spend 5 billion dollars for the road tunnel – to double the spend to create a heavy rail line under the main centres all the way up to Albany would just be ground breaking !
    If London can invest in a new Tube line Cross Rail – then some serious consideration with some actual data analysis with costs for heavy rail rail under the North Shore.
    Maybe doing it in stages like to under the centre of Takapuna – under the harbour and connecting to Aotea Station.

    I can understand there are serious cost considerations – but I do get sick of the ‘cheaper’ public transport option always being chosen – this is how we got electrification stopping at Papakura and not to Pukekohe.
    Do it once and do it right.
    This is even more important for public transport to the Airport – it really needs to be fast / separate from the road network through the whole route i.e. Heavy Rail.

    Light rail makes sense around the isthmus but not linking to major important areas like the airport or for long Rapid Transit Networks.

    1. But by doing one project ‘once and right’ they would be spending money that could be used for another project. Unless they had a really massive budget to do absolutely everything that is needed which is very unlikely.
      For example, all of those blue light rail lines on the AT map might cost less than just doing proper heavy rail to the shore.

    2. Not clear to me why rail to the Airport HAS to be heavy rail. I think that there has been a strong case put forward for light rail, and though there is also a case for heavy rail, I don’t think it’s as clear-cut as you might suggest. Cost is a significant factor – people throw around the billions of dollars worth of projects as if they were all easily fundable, but history shows that even relatively small improvements to the rail network (ie Pukekohe to Papakura electrification) are up there in the $100m plus bracket. I think we have to keep a balance between what is desirable and what is realistically achievable.

      1. The proposed light rail to the airport is slower than the current drive time from the city, seems a good reason right there.
        If the Government can afford 100 million to remove a single intersection to the airport – the Pukekohe electrification is a much better choice opening up a current 30,000 people plus the planned 30,000 people to a service 20 minutes faster travel to the city. .
        NZTA has a huge new capital expenditure budget which could easily be switched from more roads to other transport solutions.

  4. A note on the pink lined bus access map. Even though the rail network is still building up after its recent revitalisation you can see its value in terms of reducing bus volumes. Clearly the rail-less Shore is the biggest single bus source but adding the four Isthmus routes together and you get comparable number from there as from the Shore, which explains AT’s LRT plans for that area.

    Rail and the Busway will continue to provide for growth until city end limits are hit. Happily for rail the CRL will eventually arrive to both unlock those limits and hugely upgrade the whole systems utility, especially so for those out west. But the pressure on the city from ever more Shore buses is going to continue.

    The need for a Rapid Transit additional harbour crossing is clearly more pressing than the proposed traffic one, as it solves problems rather than creates the huge new one of what to do with thousands and thousands of additional vehicles in the city, on local roads, and through the CMJ.

  5. Even tho NZTA / AT planners will say they are open minded about heavy rail to the shore, I’d say its very unlikely to happen. My prediction is that the next harbour crossing tunnel will probably be car only for through traffic, with the existing bridge to take a mix of light rail, bicycles, pedestrians and mostly local traffic.

    1. My view is that the currently planned traffic crossing will become to be seen for what it is; an unaffordable disaster that creates more problems than it fixes, and will eventually be substituted by the more effective and less expensive and destructive Rapid Transit only crossing. Laparoscopic surgery instead of a vast open wound hack.

      1. Yes I’m optimistic about that too. It’s easy to forget that it’s ten years before construction is planned to start on the AWHC. Ten years! By that time the CRL will have been running for a few years, we’ll have more cycle lanes, even more people using PT, car ownership will have fallen further, climate change will be even more of a problem and so on. The idea of a vehicle crossing will have far less support than it does today.

  6. The more I look at AMETI the more I think why not make it Light Rail from day one. I’d also suggest there could be value in extending it west to the Southern Line (maybe Ellerslie via the highway) and of course easy to Botany town centre, and then south as far as you like. Ultimately to the airport.

    1. Cost. If there’s the width and terminal capacity for buses, start with buses. In effect that’s what’s been done on Dom; the buses no longer fit. And both at Panmure and Manukau city there’s the terminal capacity and the routes can be wide enough for buses. Once the buses run properly have built the demand so high they can no longer handle it, the cost of building the LR route is justified.

    2. The issue with AMETI is it will be used by buses from not just the busway e.g. buses from Howick. Could be designed for both of course but that might be even more expensive

      1. Could there also be another issue that if AMETI (e.g. from Botany) delivers LRT loads of passengers to Panmure, the Eastern Line trains will be full up? They already are, sometimes.

  7. how viable would light rail, from a technical/engineering point be on the existing bridge?

    I know light rail can handle some steep grades.

    1. The NZTA are keen to push the line that LRT could go on the bridge as they want to build a road only new crossing. When I’ve asked them privately about whether this is actually feasible they sheepishly admit that it probably isn’t and no work has been done on it. Basically it’s a complete red herring to get another road only crossing built

  8. I observed light rail last time in Sydney and it was a complete disaster.

    All very well saying the trains can go at 80kmph but it’s a complete fallacy. There is no way you want street level transport going 80kmph through a city. If that speed was safe the speed limit for cars would be 80kmph. It simply wont happen. My observation was that light rail was handicapped in the same way vehicles are – by traffic obstructions. Add to that the stops and getting people on/off and light rail going through the city was travelling at 2-3 times walking pace. A white elephant if ever I saw one.

      1. I also saw good patronage.

        I suspect when compared against the only other option – walking that it’s a pretty good choice.

        Doesn’t mean it’s the best available solution.

        1. You also have the option to drive, however in any city it will be a pretty shitty choice due to fundamental geometry.

      1. Oh please – stop conflating light rail with the busway. The poster didn’t even raise the busway.

        The 80kph is a total fallacy – stations will have to be reasonably close together to viably replace existing buses. Throw in intersections and good luck getting anything close to 80kph.

        Capacity is also another dropkick – 400-500, uhuh, almost never achievable in practice but also the vast majority of this is standing. And standing on a tram is about as fun as standing in the rain waiting for a bus.

        Throw in the need for redundancy (alternative lines when there’s even the slightest blockage) and you have a whole system precariously at risk from a minor car prang or even other trams.

        And Patrick, don’t assume because the NSW government is throwing money at something that it’s a good idea. Yes they’re extending the LRT system, but at huge cost and disruption. NSW government infrastructure spending is a basketcase and the Sydney infrastructure network is a shambles – that LRT funding is at the expense of far far more urgent projects. Not to mention the Sydney tram system requires inspectors on each tram – the cost is horrendous.

        And consider the Canberra LRT project, recently slammed by Infrastructure Australia as a total waste of time.

        1. I don’t; after all these are the muppets doing Westconnex, even stupider than our very own major idiocy: AWHC. But I have looked at the current Sydney LTR project and I rate it. In this I find I agree with that major rail Transit skeptic The Urbanist as I link to in my post on the subject above. Also 80kph is vehicle speed not average speed, obvs. Our LVRs will be full as boots on the Dom/Queen run; that is a sweet straight route between high demand places; I have no concerns about that.

          I do however agree with you about standing for much longer trips like to the Airport, and still feel our existing rail ought to be extended there. Sadly the sabotaging of the route has been more than a little effective over the last few decades by no doubt well intentioned auto-bots, to make it all but impossible. The latest being the incredibly over cooked and severing new pseudo m’way being fast tracked in Onehunga.

    1. Tim it isn’t, it’s that arrow next to the rail line. It’s not very clear, taken from a PowerPoint so missing the commentary, but those pink arrows are only bus pax numbers. No rail, ferry, car, or active volumes. Furthermore the map is a schematic not precisely following each route.

  9. Patrick’s post raises an issue that many are ignoring. Light rail has many advantages but brings a reduction in accessibility. AT is still investigating options but their intention at the moment is to have light rail stations at roughly 1 kilometre intervals versus buses which stop every few hundred metres. At present there are 28 bus stops (14 each side) down the 4.8km of Dominion Road between the SH20 interchange and View Road – with light rail this could be slashed to 8 stops. Assuming that stops are 1,000 metres apart and the average commuter lives say 200 metres up one of the side roads – the walking distance from home to bus stop nearly doubles from about 400 metres to 700 metres. So we may well need to consider supplementary local bus services, including the possibility of point-to-point on-demand services (similar to the Total Mobility scheme for the disabled but without the generous subsidy) – i.e. a shared mini-bus service that would come on request to the gate of people who live some way from the nearest light rail stop or town centre.

    1. There are already far, far too many bus stops as it is. What is needed is some good OA to determine peak number of stops to minimise door-to-door time (too few stops, walking time increases reduces bus time decreases, vice versa, the opposite occurs)

    2. 300m takes approximately three minutes to walk. People will more than make up that time with much faster and more reliable journeys once onboard.

      Of course this is an issue for people who have difficulty walking, but does the increase from 400m to 700m actually change that? Are there many people that can manage 400m that can’t manage 700m?

    3. Changing from 3 stops per km to slightly more than 1 per km is one of the best things LRT will do. We should do this regardless of whether we do LRT anyway.

    4. Remember that with buses & their stops not every stop gets used every time, whereas with LRT I presume it would. This adds to the unreliable/inconsistent times even more. In saying that I guess in peak times on busy corridors it pretty much would? Does this happen on Dominion Rd now?

  10. I’ve never understood why your CFN plan omits the crucial link between New Lynn and Onehunga. An outer loop line around the isthmus periphery would link all the isthmus entry and exit points, creating a more flexible and open network, supporting a polycentric development model. Why does Transport Blog seem to prefer a unicentric development, focussed on the CBD? Genuinely confused about this.

    1. Are you referring to the Avondale-Southdown line?
      I agree that long term it will be needed both as a passenger line but more importantly as a freight line when Auckland finally wakes up and closes the port and Northport+Tauranga become the main ports with freight moving by rail between them and Auckland.

    2. Because the CFN is only the top tier Rapid Transit Network the crosstown routes are lower demand and are met on the Frequent Network that sits just below the RTN in frequency [every 15mins] and right of way [bus lanes not busways], and is currently being rolled out. Of course should any route anywhere develop the demand or show the possibility of such demand it would be elevated up to RTN from FTN.

      So it isn’t missing it’s just on the next tier down.

      In terms of a Avondale-Southdown rail route, we really do not see the value in this or any kind of looping rail pattern; rail can only be justified where there is the highest demand and a rational and efficient running pattern. Currently we tend to agree with AT that parts of this RoW could be very useful for LR, as in the maps above.

      1. But Patrick, no consideration about moving freight through the isthmus. The North South corridor for freight should be considered a priority to get trucks of the roads.
        And this would tie in with NorthPort and Tauranga becoming the major ports to supply Auckland if further Waitemata Port development is curtailed.
        Without freight Auckland as an urban centre is stuffed.

        1. The freight demand north is tiny. Anyway no point in expanding this part of the route till you have a solution for the rest of it too, like New Lynn for example. How’s that going to work?

          Sadly, like at Panmure even more recently and on an even more important route, we just seem to close off opportunities for rail again and again.

          Just because there is an ancient designation through the Onehunga suburbs [yes that’s where it goes] it doesn’t mean we ought to built it, or that there is a viable running pattern or freight demand to use it.

          Third and fourth main to Metro Port [Southern Line to just north of Westfield] and a third and fourth to PoAL [Eastern Line] are more important for freight.

          1. Rail freight demand north may be tiny, but as Winston Peters often points out, this is because the line has seen no serious investment for decades. It is in “managed decline”.

            What are road freight volumes north like? All those trucks that they’re building the Holiday Highway for?
            Our vision should be to see this on rail sooner rather than later, and that might just mean that the Southdown-Avondale rail link has to happen sooner rather than later.

            Either that or we just give away rail freight north, as the National Govt obviously thinks we should. Vote National next election if you want to help them rid the country of this annoying problem that is the rail line to Northland! Then there will be no need for the S-A rail link and we can build a few more Kms of m’way instead. You’ll be Steven Joyce’s pal for life.

          2. I am all for increased rail [and coastal shipping] freight from Northland, but am just mindful of the constraints all along the route, particularly now at New Lynn. I’m sure many more freighters could go through at night, I just don’t know how many nor how viable that is. Would love to travel this route as a passenger too…

    3. David – a few things

      1. The CFN was never intended as a dream or even a complete system, it was designed to show what was possible by 2030 if we changed our focus. Getting an orbital RTN route in that time is too far fetched.
      2. Demands for a full RTN down SH20 are likely to be low compared to other routes and as Patrick says, there will also be cross town frequent routes which could work even better with some bus priority.
      3. In my view, a polycentric model doesn’t really do all that much to help reduce travel demand, in fact the opposite so can make things worse resulting in us needing much more infrastructure overall. It’s not uncommon for couples to work in different far flung locations meaning the only rational way to get around is for both to have a car simply because it’s so expensive to serve all areas with a high quality service that way. The idea that everyone will live and work close together is a fallacy and we can see from the real world it isn’t happening.

  11. So the “official plan” is not to have rapid transit to the Airport until ~2045??
    Heavens – I’ll be pushing 90 by then!
    And we will probably have fried the planet long before that milestone is reached, as meanwhile, we keep building moar roads. And moar aeroplanes.

  12. An observation from the Northwestern motorway this morning.

    Busy morning again in traffic (3 occupants), took 50 minutes from Hobsonville to CBD. I observed one bus this morning on the motorway. Once the roadworks are finished and the bus shoulder lanes are finished, I think there will be a big uptake in bus use in the North West – equivilent bus journey would be maybe 30 minutes from Westgate in a dedicated bus lane.

    I simply hope that PT uptake will spark the NW busway/LRT.

  13. I’m all for the re-introduction of our trams but please, for the love of Pete, make it heavy rail via Onehunga to the airport.

  14. LRT overseas is often placed on raised tracks. I guess these are cheaper than tunnels and would avoid many of the issues created by shoehorning rail into existing streets. Raised tracks can even go across existing urban blocks. The look futuristic (if done well) and would give people pleasant views across the harbour and city.

  15. AT thinking on light rail seems to be still at the “lines on maps” stage, without much thought of vehicle type, length or service pattern. Images of light rail vehicles in pedestrian malls look enticing, until the reader experiences it and find that (i) they travel very slowly, and (ii) most passengers are standing. It’s a great solution for distributing passengers within a downtown area, but painfully slow if crossing downtown.

    Most North American light rail is more like heavy rail in that it has long trains, passengers mostly seated, park-and-ride, and either travels in tunnels within downtown areas of larger cities (Los Angeles, San Francisco) or on the streets of suburbs (Long Beach, Santa Monica) and the downtowns of smaller cities (Salt Lake City). These are more like down-scaled heavy rail.

    The isthmus and East Auckland proposals seem to be more like the American streetcar or Melbourne’s trams, whereas the North Shore, North-West and Airport extensions are more like American light rail. A hybrid of these could introduce a lot of compromises such as seated vs standing space, vehicle length and downtown vs crosstown journey times. Perhaps it’s better to choose vehicle types suited to the route rather than mixing the two types on the same route.

  16. For Dominion Rd light rail to Onehunga, why not continue to use the Avondale-Southdown designation from Hillsborough Rd to Onehunga Mall? The alignment shown in the final diagram is likely to be significantly slower (there is a ~10% gradient along that part of SH20!) and less reliable due to traffic conflict. For all that it is only a 300 m shorter route. Using the existing designation would allow for stations with larger catchments as well.

    So why? Has Kiwirail prohibited its use by AT? Will NZTA chip in for a motorway alignment but Kiwirail won’t/can’t match funds on their designation?

    At first reckoning, it doesn’t even seem like it would be significantly cheaper. Easing that gradient is likely to involve significant earthworks/structures and the necessary upgrades to ~1.6 km of what are now narrow, winding, ‘high-friction’ residential roads on the old Onehunga foreshore will be costly as well. All for an inferior result.

  17. Congratulations all round as I feel we would not have got to this stage if it had not been for your advocacy. So thanks to Transport Blog, Generation Zero and Campaign for Better Transport for all you good work and keeping us informed and taking suggestions and developing them. This seems to be democracy in action and it is effective.

    1. + 1

      Hear hear. I agree, I think without Transport Blog, Generation Zero and Campaign for Better Transport we’d have seen the same old same old from the transport planners (if they can be called that).

  18. I question the assumption that the actual cost of setting up a light rail system would be much more expensive than busways. Buses have shown themselves to quite hard on their roads whereas light rail treads pretty lightly over the ground. This is basic physics. The very nature of rail spreads the impact of the weight on the wheels over a much larger area, whereas a bus’s weight is concentrated only on the spot beneath the tyre at any one point. Old construction techniques of laying tram tracks in concrete, as at MOTAT, have been superceded by new techniques of laying them on a solid bed and then holding them in place with hot mix asphalt, as per the Wynyard Quarter tramway.
    So the actual cost of laying the tracks could well be around about the same as a high grade motorway type surface, as per the Northern Busway, which just leaves the overhead, and it wouldn’t surprise me if that could be added and the cost of light rail would still be within the ball park figure of a Northern Busway type of facility

Leave a Reply