One of the items on the agenda for a decision at tomorrow’s Auckland Transport board meeting is about the intersection of Tamaki Dr and Ngapipi Rd. This intersection has been discussed for quite some time as is quite dangerous with it considered by the NZTA the 10th highest risk intersection in NZ. AT say 31 crashes have been recorded at the intersection over the past five years (possibly more now as there was another one recently) with 25 of them resulting in injury.

To improve the intersection AT want to up-scale the intersection and signalise it which they say will provide the best outcomes in terms of improving safety and improving traffic flow. As part of the up-scaling it will see the sea wall pushed out quite a bit to create more space. AT’s preferred option is shown below

Tamaki -Ngapipi Signalised Option

However as I understand it the Orakei Local Board have continued to push for an alternative roundabout option. I’m not entirely sure of their reasoning for this although I wonder if it isn’t about some kind of animosity towards stopping at lights. This option is shown below.

Tamaki -Ngapipi Roundabout option

For me one of the big issues with the roundabout option is how it works for pedestrians and those on bikes. For example if you were coming from Ngapipi and wanted to walk anywhere on the northern side of the road you will need to

  • be capable of dashing across two lanes of traffic approaching the roundabout to the traffic island
  • cross a lane of traffic exiting from the roundabout including or turning left around a blind corner
  • cross Tamaki Dr east of the intersection

At least with the signalised option this is controlled and you can cross in any direction with a wait for lights. This isn’t to say I think the signalised option is ideal. For example why go to the cost of extending out the seawall and not at least provide a protected cycle option that isn’t a shared path.

Other than safety there’s another area where AT say the signalised option works better, traffic flow. They have put online the results of their modelling in the AM and PM peak showing how long they think the traffic queues will be with either option vs the current layout. As you can see in the AM peak AT think the traffic queues will be almost non-existent with the signalised option – which seems almost too good to be true. The roundabout option is suggested to have traffic queued up way back down Kepa Rd.

Tamaki -Ngapipi AM queues

In the PM peak the roundabout performs a little better for those heading away from the city but not but too much.

Tamaki -Ngapipi PM queues

Based on this information it seems the signalised option is a far preferable solution. I’ll be watching with interest to see the outcome of the board meeting.

Share this


  1. Sharrowsing your way around a roundabout is hardly safe, and it never feels hospitable either. This could be one of the best places to cycle anywhere in Auckland, giving people recreation and health. Right now it’s not safe, but a roundabout would have the effect of making it even less safe.

  2. After reviewing numerous modelling reports, I have come to the conclusion they are all BS and are hardly worth the paper they are printed on. There are so many assumptions put into them that the outcome has little basis in reality. You can get them to show whatever you want them to show depending on who is paying the modeller. I think the roundabout option would perform better than signals. For cars at least. Signals will be better for pedestrians and cyclists. Either option would be hideously expensive.

    1. Im sure the outcome will be a form of non-decision, and they will just let people crash and get injured some more. It’s been what has been happening for years now, with no sense of urgency.

    2. How can the roundabout perform worse? The design shows in the AM peak 2 exit lanes towards the city with one approach lanes from each direction. This is a free flow solution given the light right turn demand.
      The modelling as Ari suggests is rubbish.

      Go with the signals, roundabouts are rubbish to cyclists and this is the premium route in the city. Why is a model dictating a strategic choice?

      1. A roundabout depends on somewhat balanced flows from all arms to function well at high capacities. A roundabout like this means that a steady westbound flow of traffic from Tamaki Drive into the city means that those drivers do not have to give way to anyone (as hardly anyone right turns into Ngapipi in the morning). Therefore, there will be almost no gaps in the westbound traffic, and the only way Ngapipi traffic can enter the roundabout is after a LOOOOOONG wait, or by chancing it and taking too-short gaps. Recipe for long queues and dangerous behaviour, and even more crashes of cars& cars and cars&cyclists.

        The model may not show exactly where the queues will end up – as Ari noted, it kinda depends on lots of inputs. But a roundabout is still crud for this intersection, for various reasons. Part of me wants to see it built so the Local Board falls painfully onto their faces. But all they’d ask for in that case is to make it a three-lane roundabout!

        1. “Part of me wants to see it built so the Local Board falls painfully onto their faces. But all they’d ask for in that case is to make it a three-lane roundabout!”

          Road expansions can never fail: they can only be failed.

        2. Nope. Look at the layout. Both movements into the city move have their own lane.
          Anyway, the design is as poor as the modelling. The horizontal alignment is appalling. Who was the consultant who drew this crap up?

        3. Amazing. The roundabout queue is somehow going to tailback through the existing queue that is already on Kepa Road each morning. How do they get two cars to occupy the same traffic in gridlock? It could solve all our problems.

      2. Except right turning traffic would have to give way to the endless stream of city bound traffic. There is only space for about two or three right turn cars to stack until they block Ngapipi. So unless you are widening the sea wall all the way down Ngapipi, I can see why it performs worse b

        1. How is that different to the existing?

          An objective of the scheme is to address right tirn to the east movements? Rubbish.

          I am for the roundabout, but just saying if this is challenged AT are going to have a credibility issue with in house modelling going on.

          This article implies AT are using capacity to justify a signals when the issue is safety. Bonkers.

        2. It’s not different to the existing, which is presumably why it performs worse than the signalised option.

  3. I don’t see why a roundabout would be that bad for cyclists.
    It’s not not like either option is going to have a dedicated cycle lane through the intersection.
    A signaled intersection takes longer to clear, even with low traffic volumes.

    1. Roundabouts – especially large multi-lanes ones like this, are horrible in terms of crash record for cyclists. Even our cycle safety panel last year said so. It’s because drivers concentrate on other drivers, because cyclists are in a worse spot (edge of vision) compared to cars, and because such multi-lane roundabouts dont achieve sufficient speed reduction. Plus, it means pedestrians need to run multiple lanes to cross. It is crazy that in a location where even the normal benefit of a roundabout – massive car throughput – doesn’t really work, we still consider putting one in anyway! Crazy.

        1. Did you look at the diagram of the roundabout layout above? In particular the part labelled “new cycle ramp” on the slip lane?

  4. If this route is to be used by the Maxi sized trucks then I think that the roundabout is not a possibility.
    Maxi sized trucks seem to think that they can drive anywhere and I do not know where to find anything that limits their use of any routes/streets.
    My experience as an older cyclist is that roundabouts are very stressful and you have to take the lane. Which takes a lot of confidence and guts in the approach.

    1. trucks are actually not allowed at night east of the Port. There are few very small signs saying this. Not that the truckies repect it anyway.

  5. Seems like AT has to beg everyone for permission before they do anything… Imagine if every set of lights in Auckland went through this process before they were signalised – we would hardly have any signalised intersections
    AT should be the experts on this, they shouldn’t have to listen to the local board’s ideas on traffic flow.

  6. If traffic lights are put on that intersection you can guarantee that there will be traffic jams on Tamaki Drive far worse than now. And I would hate to think what the salt water that tends to wash around that area will do to them.

    Auckland cannot seem do a decent job with traffic lights. Quay Street lights, for example, regularly clash with each other and traffic starts and stops at each intersection. They are set up seemingly to short phase and then remain green on empty lanes. Its like they opt for the cheapest nastiest version although I doubt they are cheap to buy. And then its a very big wait to have the problem rectified. I think anything but traffic lights has to be better.

  7. Or they could actually fix either of the two cycle shared paths that are already there and make it actually usable.

    But yea, another set of lights to cause tailbacks on Tamaki Drive, while at the same time not putting in any eastbound bus lanes? Can’t see that going wrong at all.

    1. Also the intersection retains the option for bikes to proceed straight and then mentally requires them to phase across the left turning lane from Ngapipi and up onto the footpath.

      Which, if I’m not mistaken, is one of the current design’s most glaring flaws.

  8. People on here claiming roundabouts would be better for flow makes me laugh. First day as a transport engineer by supervisor said in passing, unprompted how bad roundabouts are in a situation like this

    1. Well given there is only one intersection with traffic lights on Tamaki Drive proper and it causes massive tailbacks, I’m not sure I blame people for being apprehensive.

      1. The traffic lights dont cause tailbacks, the amount of people chosing to drive at any particular time do. The best thing for Tamaki Drive would be to improve bus/ cycling and walking infrastructure to help reduce demand for private vehicle travel along the route.

  9. Actually i think the signalised intersection looks quite good all things considered and by auckland standards. I note that it has:
    1. Pedestrian crossings on *all* approaches
    2. Marked bus lanes and cycle lanes.

    While the latter has to cross the left turn i can’t see much alternative, except for an extra signal phase to hold up cycles when left turning vehicles.

    I do tend to agree with others on the need for bus lanes in east bound direction: if you’re moving the sea wall anyway is now the time do it?

  10. I hate cycling through the existing intersection westbound and multilane roundabouts. Has the local board got a vendetta against us?
    As a rate payer, the signalised option looks expensive and why isn‘t tamaki drive left straight?

Leave a Reply

Your email address will not be published. Required fields are marked *