Back in early September the council’s transport committee looked at three important reports: the annual screenline study of trips into the isthmus and CBD, a study of congestion levels and a benchmarking study which compared Auckland to various cities overseas. Quite rightly, most of the focus was placed on the bench-marking study, which highlighted how comparatively poor our public transport system is (not that surprising) and also how expensive it is for users (a bit more surprising).

The screenline study, while based on a single day’s data (and therefore impacted on by one-off issues such as what the weather was like), provides some really useful longitudinal data of how Auckland’s PT system has worked over time – particularly in terms of the number of people entering the city at peak times. It allows us to create graphs like this one: Perhaps a bit more helpfully, it also enables us to look a bit deeper, comparing how each entrance point to the city centre has performed over time:
You can see that the most dramatic increase has come along Fanshawe Street, reflecting the Northern Busway making public transport travel for those on the North Shore much more attractive over the past five years. The decrease for Symonds Street is due to many buses being rerouted via Grafton Bridge.

What’s particularly interesting is to start combining passenger numbers with service numbers. Short of spending hours digging through timetable information to find the exact number of services entering the CBD along each route at peak times, the best way of getting a rough guide is using some very handy work that NZTA had done for them as part of the Waitemata Harbour crossing project – which tells us the number of services per peak hour (not the whole AM peak). A general rule of thumb is that the peak 60 minute hour contains around 60% of the whole AM peak numbers, so we can start to formulate some interesting data: What we can see is a surprisingly large difference between the average number of passengers per bus along the different main routes. Amazingly, on average buses that enter the city centre via Fanshawe Street have more than twice the number of passengers as those entering the city centre across Grafton Bridge!

The obvious reason behind the huge difference is to look at where buses that travel northbound along Albert Street and come in along Grafton Bridge have travelled for most of their journeys, which (generally) includes places where these bus routes duplicate railway lines. Grafton Bridge is the main entry point for buses from south Auckland, while Albert Street (northbound) is generally used by buses from west Auckland. By contrast, there is obviously no rail on the North Shore, while many buses that come in along Upper Symonds Street have come from routes like Mt Eden Rd, Dominion Road and Sandringham Road – which generally don’t duplicate the rail corridor (New North Road being the obvious exception).

This is clearly a legacy of us not adapting our bus network to the fact that we now have a rail system. It means that we don’t have enough buses on North Shore routes (meaning they’re overcrowded) while providing far too many buses from the south and west, even though people aren’t bothering to use them anymore because the train is so much faster. Fixing this wouldn’t cost any money, and in fact could save money and provide a much better service level where it’s most needed.

Share this


  1. It means that we don’t have enough buses on North Shore routes (meaning they’re overcrowded) while providing far too many buses from the south and west, even though people aren’t bothering to use them anymore because the train is so much faster.

    Not that I doubt what you say but you’d should have the train numbers there as well so that it can be compared by your readers.

      1. Interesting, making some guesstimates of the numbers in that graph we get that the average of about 180 per train. The average four car train has about 266 seats so decent utilisation but space for growth (although there will obviously be some services that are busier than others and not all trains are 4 cars long)

  2. The peak rail services have no free seats between Kingsland and Grafton, or between Ellerslie and Newmarket. Plenty of capacity between Newmarket and Britomart, but that’s almost irrelevant. Teh EMUs wil help some but only the loop can really help.

    1. those passengers could be well served by cross town (southern to western line) services. I think they could really save us over the next 7-10 years while we have no loop

  3. “The peak rail services have no free seats between Kingsland and Grafton, or between Ellerslie and Newmarket. Plenty of capacity between Newmarket and Britomart, but that’s almost irrelevant”

    What is the current situation for the peak loadings between Newmarket and Britomart? If there is ‘plenty of capacity’ then it means that we can add extra trains, but not direct them into Britomart. Instead:

    * Western Line-Newmarket-Otahuhu
    * Otahuhu-Newmarket-The Strand-Otahuhu

    come to mind. It would simply mean ‘training’ the passengers to change at Newmarket, but if the timetabling is tight enough it could work.

    I know I have asked this before, but what are the problems with doing this?

    1. Main issue with adding extra trains is that there aren’t any to add – even the opening of Manukau next year looks like it will be delayed because AT have no spare trains to run the service with.

      1. Really? that is a detail that I would have hoped someone would have thought of before now.

        Could we bring the silver ferns back into commuter service? Also, could we not make trains shorter/ enact the south-west services? Overall- shorter more frequent trains would provide the same number of seats in the critical Ellerslie to Newmarket section.

        I thought we had no new carriages between early 2012 and the electric fleet anyway? surely they wouldn’t delay it that long would they?

        1. Making trains shorter, you have to be kidding me and you wouldn’t say that if you caught peak services as even the ones we have now are struggling in many places because they aren’t big enough at certain times. Further there are no more SD cars (the ones with the driving cab on them) to make up new trains even if we wanted to so the only other option would be to hook up a loco at each end which is probably a pretty big waste of resources.

          I would recommend that you jump on a peak train during March next year to see if we should be cutting seats.

    2. The service I catch is generally pretty full between Newmarket and Britomart even after a large number hop off at Grafton and Newmarket. Really it depends on the service as some are busier than pointers others.

      Just the other day I caught a later service that didn’t get to town till just before 10and it would have had 150 people on it at one point

  4. That might work, but would be annoying because of the way Newmarket station currently operates. If the trains from the Western Line continue to pull into platform 1 (I think it’s platform 1 – I don’t have the numbers memorised!) as they currently do, they have to go up the escalators and down onto the other island so as to get to the Southern Line trains. This problem could be pretty easily resolved by having the connecting trains use platform 2, which is not currently used. That would let passengers go directly from one train to the other, a simple matter of crossing the island.

  5. I meant capacity in seats, not schedule slots.

    But anyway, forcing transfer at Newmarket doesn’t make sense in terms of passenger destinations. More than 50% of traffic through Newmarket is to/from Britomart. South transfers are quite a small percentage. The Strand idea was tried when Britomart opened and at that time there was no patronage for it whatsoever – single digits as I recall.

    1. If there is spare seating capacity between Newmarket and Britomart, then we have to know how to use it effectively. As Ernest Rutherford put it, “we haven’t got any money, so we are going to have to think“. Some considerations:

      * If more than half the Newmarket traffic is for Britomart, that’s still a lot that isn’t.

      * Britomart proper handles half (?) the peak traffic in the system. This is in contrast to Wellington, in which 90 percent of the traffic is from Wellington or to it.

      * There’s a lot more activity around the Strand than there used to be.

      A perfect solution? Hardly. But we may not have a perfect solution for a while to come.

      1. Ross, I think that if there were the trains the next best option is west/south. Especially once Manukau is operational, and even more so when MIT opens its campus on top of that station. The likelihood of students living in the western line catchment enrolling in courses there could be improved by the existence of a good direct service or at the very least some co-ordination of transfer services at Newmarket.

        You’d hope MIT and AT might even meet and discuss these things…?

        The Strand?; it’s nowhere.

  6. Regarding the not-so-full buses coming in across Grafton Bridge, could that be because inbound passengers are getting off at or around Newmarket?

    It would be useful to have stats for patronage entering the wider centre city area, including Newmarket, to get an idea of how many people are travelling there.

    Roll on HOP statistics.

Leave a Reply

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