One of the key public transport projects on the books – and in our Congestion Free Network – is an extension of the Northern Busway to the north. Thanks to an Official Information Act request by reader Hamish O we now have a whole lot more information about the project. A study on the extension was completed for the NZTA early last year to look at the preferred route for the extension all the way from Constellation Dr to Silverdale. As well as the total route, for practical purposes the report also broke down the sections in Constellation to Albany and Albany to Silverdale. You can read the full report here (8.9 MB) however here is the executive summary.

The New Zealand Transport Agency (the NZTA) has engaged Beca Infrastructure Limited (Beca) to investigate an extension of the existing Northern Busway from Constellation Station to a future Hibiscus Coast Busway Station at the Silverdale Interchange (the NBE). The investigation by Beca involved: review and development of previously investigated and new route options for the NBE, to identify a preferred route; identification of land required to accommodate the preferred route; and consideration of station location and operational issues.

The investigation responds to the strategic objective of the NZTA to deliver an integrated, safe, responsive, affordable and sustainable public transport solution for North Auckland. The extension has been considered as a dedicated facility, separate to SH1 with associated stations being the responsibility of Auckland Transport and Auckland Council. These parties as well as local Iwi have been engaged throughout the project.

Increased population and employment growth is forecast for North Auckland and this will place increasing pressure on the transport network and available land. Current predictions to 2041 show increased SH1 traffic, resulting in congestion and varied travel times for people and goods moving through North Auckland. The need for additional dwellings and places of employment associated with the predicted population growth will require land which may also be necessary for an extension to the Northern Busway.

An extension to the Northern Busway would provide a public transport solution to accommodate some of the transport needs of a growing population within North Auckland. The Busway extension is not predicted to result in a significant decrease in traffic on SH1 because the number of people expected to use the bus rather than travelling by car is small in comparison to the overall number of vehicles using SH1. However, the NBE would improve travel times for people travelling by bus in the future.

To ensure an extension to the Northern Busway can be built in the future there is a need to allocate land for this purpose. This can best be achieved by introducing a new designation for Busway purposes and purchasing required land. It is recommended that the land purchase initially focus on properties where there is likely to be an increase in land value (as a result of population growth placing pressure on available land) and/or where negotiations with land owners may require significant time.

Through investigation and evaluation an eastern aligned option has been identified, and is recommended for the future extension of the Northern Busway. Being aligned to the east of SH1 the option:

  • avoids a site of ecological significance at the Lucas Creek West Bush (located just north of the Oteha Valley Road Interchange to the west of SH1) providing for the protection of the environment;
  • provides the greatest flexibility for future State Highway improvement projects;
  • provides a bus only road link across SH1 to serve the Albany Station, enabling this station to support the future growth of the Albany Centre; and
  • is cheaper to construct as it avoids the need to construct one or more major structures across SH1.

Based on economic investigations, the full NBE would likely be economically justified as early as 2019. Should the NBE be constructed in stages, a first stage from Constellation Station to Albany Station could be economically justified as early as 2015. The construction of the Busway during the indicated years would support the growth of the Albany Metropolitan Centre, Silverdale and Orewa in accordance with the emerging strategic direction for growth in Auckland.

The analysis undertaken as part of this project demonstrates that there is little benefit in providing bus shoulder lanes to Silverdale or incrementally. However, the case for bus shoulder lanes should be considered further when the project proceeds to preliminary design, and better information is available as to the associated costs and the effects on the network following completion of other projects (i.e. SH1 to State Highway 18 connection, Constellation to Greville improvements, SH1 to State Highway 17 connection and Penlink).

Prior to confirming the preferred option for a future extension of the Northern Busway and setting aside land for this purpose, it is recommended that NZTA undertake further consultation with Auckland Council, Auckland Transport and Hokai Nuku, in addition to initiating consultation with other Stakeholders, such as Watercare, and the community.

There are two really interesting outcomes to the study that are mentioned in the summary above. The first and most significant is that the preferred alignment is not to the west of the motorway like many people have long assumed but to the east. The second is that extension from Constellation to Albany could be economically by 2015.

So let’s look at the alignment options. The study into alignments needed to take into account the following potential projects.

  • Three Laning of SH1 – the proposed NBE design and designation footprint has been developed in a manner that would enable three lanes in each direction to be provided continuously on SH1 as far north as Silverdale without disruption to the Busway in the future.
  • SH1 to SH18 Motorway to Motorway Connection – the proposed NBE design and designation footprint has been developed to accommodate a future State Highway 18 (SH18) to SH1 motorway to motorway connection upon completion of the Auckland Western Ring Route (based on current design information).
  • SH1 Greville Road Interchange – the proposed NBE design and designation footprint has been developed so as not to preclude improvements proposed at the Greville Road Interchange in the future (based on current design information).
  • Penlink (or Weiti Crossing scheme) – the proposed NBE design and designation footprint has been developed so as not to preclude a proposed connection between the Whangaparaoa Peninsula and SH1 (south of Silverdale) in the future (based on current design information).
  • Weigh Station – the proposed NBE design and designation footprint has been developed toaccommodate a future weigh station (compliance checking site) located to the east of SH1 and north of Bawden Road, which would enable overweight vehicles to be diverted onto the Western Ring Route away from SH1 and the Auckland Harbour Bridge (AHB) should this be required in the future.
  • Hibiscus Coast Busway Station – the proposed NBE design has been developed to accommodate and connect with the Hibiscus Coast Busway Station.

The team investigating this then came up with 5 potential options.

  • Option 1: Offline facility primarily on the western side of the existing motorway corridor (crossing north of the Rosedale Oxidation Ponds);
  • Option 2: Offline facility wholly on the eastern side of the existing motorway corridor;
  • Option 3: Offline facility crossing from east to west beneath SH1 in a covered trench or tunnel and returning to the east by way of a bridge to the north of Lonely Track Road;
  • Option 4: Online facility comprising bus shoulder lanes in both directions, accessed via the existing motorway on and off ramps; and
  • Option 5: Central median Busway, accessed at Silverdale interchange and Constellation Drive.

After initial screening, options 1, 2 and 3 were considered the best to take forward for more detailed study which assessed them based on integration, social, environmental and economic criteria. I won’t bother going through details so feel free to read the report if you want more info but as mentioned earlier, option 2 was considered the best. Probably the most interesting part of it is how it would access the existing Albany Busway Station. The answer is it would be done by way of a bridge across the motorway which would able to be used by both buses from the south of Albany and those from the North. Essentially it means that only one crossing of the motorway needs to be made.

Northern Busway Extension - Eastern access to Albany

As for costs it is suggested that the section from Constellation to Albany would cost just over $200 million to construct while the section from Albany to Silverdale would cost just over $300 million although the report does note that the figures have been rounded to the nearest $100 million. A separate report from 2011 also released with the OIA request suggests the costs would be $249 million for the Constellation to Albany section while the Albany to Silverdale section would cost $304 million.

Northern Busway Extension - costs

To put things in perspective the original busway cost around $220 to build the roughly 6.5km of busway from Constellation to Akoranga. The study – as well as a separate one done at the same time by Auckland Transport – also considered whether any new stations could be justified along the route. The only one that was considered to be potentially viable was one at Greville Rd

One other comment really caught my attention in the report. It is this from page 18 and 19 and it explains quite nicely just how successful the existing busway and associated improvements have been.

Over the past few years investment in the Northern Busway, and efforts to improve bus and transit lanes in other parts of the North Shore, have resulted in a significant increase in the proportion of trips made by bus. Not only has the number of bus users across the Harbour Bridge improved significantly during this time, but there has been a decline in the number of cars crossing the bridge: freeing up space so everyone’s trip is faster and more reliable.

Recent figures indicate that almost 12,000 out of the 29,000 people crossing the bridge in the morning peak period are now travelling by bus (i.e. almost 41 percent of all people use the bus). This figure represents a significant increase in bus mode split compared to 2004 (which had roughly 5,000 out of 27,000 (18.5 percent)) of people crossing the bridge at peak times by bus.

Personally I think that the extension from Constellation to Albany needs to be built as soon as possible and is far more important than the works the government is proposing in the area with the motorway upgrades (which to be fair do mention Northern Busway improvements).

Lastly thanks to Hamish O for putting through the OIA request for this.

Share this


  1. It seems remarkable that getting a busway between the existing stations at Constellation and Albany will cost as much as the whole busway, including the stations and the bridge over the Tristram intersection.

    I think they seriously need to look at putting in multi-level parking at Albany. Even opening up several new rows of parking only pushed back the time the whole thing is full by a few minutes.

    1. What about better feeder bus services instead? And better walk & cycle accessibility of the busway stations? Albany may be (compared to other locations further south) and better place for park & ride than others. But it still rubs me wrong to spend enormous amounts of money on car parking for a public transport service.

    2. The problem with multi-level parking is that it costs around $20,000 per space to build (assuming you own the land already: see here 680 parks for $14 million

      At that price the cost of capital alone works out to be about $5 per weekday. Add in operations and maintenance and you’d have to charge commuters about $8 a day for parking just to break even. If you provide that free then you are subsidising each trip by $8, which is enough to subsidise three bus trips instead.

      Or put it another way, $14 million buys us 680 car parks, which we then have to pay more to maintain and operate. That gets us a bit more than 680 extra commuters in total all day. Put that same $14m in the bank and we get $840,000 a year in interest. That’s enough to fund about eight buses and drivers running collector routes all day, or enough to deliver about 800 commuters *every hour* to the busway station instead.

      1. That works out to be about $1k per square meter for building a car park which sounds about right.

        Not on topic but for future reference. An apartment would go at around 4 to 5 times that amount.

      2. In terms of cost, assuming a 5% interest rate you only need to charge $2.70 a day to cover finance costs. Adding on rates and other costs you should still get change out of $8 a day.

        It’s still a subsidy however if you provide it for free. Little different to what you get with other modes of PT.

        1. I’ve used a 6% discount rate (capital risk and all that), and only divided the annual cost by working weekdays (nominally five days a week, less twenty a year for holiday periods), but either way the conclusion is the same.

          Is it actually worthwhile to subsidise those park n riders when we could probably get much better outcomes subsidising collector buses. On another level, you have to ask why we would ever let park n ride happen for free (100% subsidised) when we charge fares for the the collector alternative (~50% subsidised).

          You make a good point on the opportunity cost of the land, why build a bunch of carparks next to the station when we could build a bunch of apartments instead. Surely that would be a better use of the land, and might generate more trips than the parking would.

        2. Effectively parking at a park n ride is a luxury of conveniance and I suspect a number of people would be happy paying a $4 fee. I guess the risk is that these direct costs can scare some folk away and back to driving the full length.

      3. It’s also a little pointless, since the busway station is surrounded by acres of vacant land. Why build an expensive multi-storey structure when the landbanking neighbours can provide pay-and-display parking for pretty much the cost of asphalting a bit of a paddock?

        1. Yes by all means, if the land owners want to charge market rates for parking while waiting to develop then let them, but lets not have AT buying land and building structures to provide it.

    3. As well as multi level parking at Constellation, and a parking facility at or near Akoranga. The streets along Esmonde Road are filled with the parked cars of commuters catching the bus into town.
      I hear regularly that the busway is nowhere near capacity; but from the times I’ve attempted to use Northern Busway park-and-ride facilities, the carparks are certainly at capacity and the stations frequently have buses queuing to get to the platform (the last time I used the expressway was prior to the introduction of the double decker buses, however).
      As I see it, the busway is such a success that it should be one of the first priorities to extend it further north and upgrade the current facilities, which were inadequate within months of the busway opening.

    4. The reason the Albany carpark fills up so quickly is that it’s the northern most access point to most of the bus services. By creating a park and ride at Silverdale, served by frequent buses that tie in properly to the rest of the network, a lot of the parking pressure at Albany should disappear. A lot of the cars I used to follow into that carpark came straight off the motorway.

      1. Relatedly, it’s also the start point for most of the City-bound buses. It may be better now, but when I was commuting, it was usually standing room only by the time we got to Constellation. I imagine quite a few people drive *north* to Albany just so they can sit down the whole way.

  2. Is it just me or am I not alone in thinking that the little branch to Albany is silly? Why not just put the busway right next to the, very busy, Albany interchange? What if, in the future we do want to run Light Metro rail along the busway? Yes, it’s probably cheaper (and I have not had time to read the document yet) but is the cost saving a stitch up for later on?

    1. Not sure what you mean by right next to the very busy Albany Interchange, but I think you’ll find that that ‘little branch’ probably ends up being the end of the busway for several decades. Personally I can’t see them needing to extend the busway proper north of Albany for a very long time (buses running on the motorway with some bus lanes in congested places would do very well).

    2. I’ve had a skim through the document now and I think the Option 3, while being more expensive, would be a better long term proposition.

      1. I, too, have only skimmed it. Do they say why they considered only a tunnel for option 3, and if so, where? The obvious place would seem to under the hill between Greville and McClymonts Roads – get under the motorway and ease the grade a bit for the buses at the same time.

        1. That sounds very $$$. What is the advantage of a tunnel at that point over a bridge next to the station. I just can’t see any material difference for bus service but its probably an order of magnitude different in cost.

  3. “The Busway extension is not predicted to result in a significant decrease in traffic on SH1 because the number of people expected to use the bus rather than travelling by car is small in comparison to the overall number of vehicles using SH1”.

    Surely the point here is peak traffic (since that is what determines the build of SH1 in any one place) and I imagine a comparison of peak travel ratios would be way higher than ‘small’.

    1. Thats the thing right, the real value comes at the margin at peak times, at the point where and when it is actually congested. I don’t see 40% of trips over the bridge being ‘small’.

  4. This alignment looks good for a future DLM extension across to Massey University/ the Mega Centre. Looks very difficult for an extension to Silverdale though.

    1. Why not start with a DLM from Massey University to Constellation. That way it is done once and it is done properly, no retro fitting to something that wasn’t designed for it; or worse perhaps,something that is over designed in the first instance because it later has to accommodate DLM.

      If the Shore is to have rail, and many see that as a reality, then starting with a short link doesn’t seem unrealistic. Compare the holiday highway where it started with a small chunk. People said that this was silly, and still say this, but for an entirely different reason than it is only a small bite of what is planned.

      It will be inconvenient, but changing from DLM to bus at Constellation is not impractical. The line can be extended south as funds allow. Arguably building from the north first fixes the area of greatest congestion as beyond Akoranga the road is four and five lanes.

  5. This report uses 2011 data. I have made comments below related to section 3 of the report on volumes with 2013 updates. Based on own experience and timings as travel Albany to CBD each workday in 2013.

    Section 3.1 Overview pg. 18
    “In general, traffic currently flows satisfactorily on SH1 north of the Greville Road Interchange.”
    2013—Southbound Traffic now is slowed by congestion North of the Greville road onramp at 0640 on university days.

    3.3 Traffic speed and delays pg. 20
    “Based on (2011) surveyed traffic speeds along SH1 between Oteha Valley Road and Fanshawe Street (for car traffic in the AM peak, PM peak and inter-peak periods (by direction)) there are currently few delays experienced along the SH1 corridor north of Greville Road.”
    2013—There now are delays North of Greville road at the AM peak on University days.

    3.7 Bus operations Page 22 (i)
    “Services between the Hibiscus Coast and Auckland (hourly off-peak and via Flat Bush, more frequent at peak times and some via the motorway) – approximately 50 buses per day (i.e. services 893/895/896/898/899).”
    2013—From Timetable 68 buses running from Silverdale to at least Takapuna from 2012

    3.7 Bus operations Page 22 (ii)
    “Weekday peak period only services between Long Bay and Auckland / Newmarket – approximately 20 buses per day (i.e. service 86X/881).”
    2013—On 5 March 2012 the 881 increased to 47 buses per day.

    3.7 Bus operations Page 22 (iii)
    “At peak times (Monday to Friday) a number of additional services use the Busway. The majority of these additional services are north-south services to/from Auckland CBD (e.g. services between Greenhithe / Browns Bay and Auckland CBD (Mayoral Drive / Wellesley Street / Albert Street)).”
    2013—The 76X, 85X, 86X, 87X, 863X and 866X run are express services on weekdays only. There are 66 of these per day.

    3.7 Bus operations Page 22 (iv)
    “Bus services operating on SH1 also experience some delays at peak times between the Oteha Valley Road and Upper Harbour Highway Interchanges due to traffic congestion.”
    2013—On university days the congestion delay for buses going from Albany station to Constellation station is about 2 minutes 40 seconds at 0635 in the morning (My own recordings)

  6. I have just started catching the Northern Bus from Sunnynook to Akoranga, and it is very good, except that I always have to stand, usually right up the front next to the window. So I think they need more buses on. Secondly, why aren’t there more park and ride facilities? Sunnynook and Totoravale turn into car parks during the day, and I know one street is literally one way only because it has to accommodate so many parked cars for the Sunnynook bus station. Why didn’t they create more car parks? It seems that the system would work really well if there were more car parks and more buses.

    1. great to see NEX being used for non CBD trips. Hopefully more double deckers coming soon, that will help a lot, but patronage growing fast so hard to handle increases.
      Most stations surrounded by very valuable land, not really any space for more parking at Sunnynook. Need much better connecting bus services, and these should be free, so people who can catch the bus do this. Will free up spaces for those that live more distant from the bus network.

    2. The biggest problem with carparking is they require about 10 square meters of tarmac for your car to sit on all day, and another ~8 square meters of shared tarmac for you to get your car in and out of that parking space – so about 18sqm per car (and I’m being conservative). Auckland average commuter car occupancy is 1.2 people per car, so that’s about 15 square metres of land per person catered for at a park and ride. That’s a lot of land – one park-and-ride busload (60 people) needs 900sqm of land for carparking, or an area 30m long by 30m wide.

      For context the average lot size for a North Shore house is 176 sqm. So that’s 6 housing lots worth of land required to park the cars of just one busload of commuters.

      Pam do you still think bigger park-and-rides are a good idea and will meet your expectations given those numbers?

  7. From 5,000/27,000 to 12,000/29,000 … that is impressive. If you extract out the share of this bridge traffic which is *not* going into the CBD, say, 10,000 or so – because buses would have a very small share of this market – then the ratios go from ~ 4,500/17,000 to 11,500/19,000. That represents an increase in market share of ~27 to ~60 percent of the CBD market proper. Could someone provide a more accurate value for cross-bridge am peak movements into the CBD proper?

Leave a Reply

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