The CRL tunnel is the easy part: What Sydney, London and Leipzig tell us about the nine years after the ribbon
We are back in New Zealand this week and came home to the opening of the CRL...exciting times! This week's article focus on CRL and what other examples around the world can teach us for the city of sails.
Auckland's City Rail Link opened on Sunday. By 11am on Monday around 32,000 people had gone through the gates, up 38 per cent on the Monday before, and Sunday was the busiest day the rail network has ever had on a Sunday.
Every one of those numbers will be quoted this week, but it is important to remember one thing...
Opening weeks are novelty.
The interesting question is what the number looks like on an ordinary Tuesday in March 2028, and for that we have to look at places that are further down the same road.
In this article we review three of them.
Leipzig: the tunnel is not the product
Leipzig built a city tunnel almost exactly the length of ours. 3.6 km against Auckland's 3.45 km. It opened in December 2013, four underground stations, and it cost around 960 million euros against an initial estimate of 572 million, so the overrun story is familiar too.
What Leipzig actually bought was not a tunnel. Leipzig Hauptbahnhof was a terminus. Trains came in, stopped, and went back out the way they came. The tunnel turned that terminus into a through station, and the regional rail network reorganised itself around the fact. S-Bahn Mitteldeutschland is now the largest S-Bahn network in Germany, 802 route kilometres, ten lines, 145 stations, carrying about 90,000 people a day across a metropolitan area smaller than Auckland's.
That is the same trade Auckland has just made. Britomart was a dead end. Waitematā is now a through station. The tunnel is three and a half kilometres of concrete. The product is that every line can now reach every other line without reversing, and the network can be redrawn accordingly.
The lesson is that the value does not arrive with the tunnel. It arrives when the timetable, the bus network and the land use are rebuilt to exploit it. That work is cheaper than the tunnel and it is where the benefit actually sits.
London: forecasts are not built for step changes
The Elizabeth line opened in May 2022. Transport for London's post-pandemic forecast was 104 to 136 million journeys a year once fully open, rising to 130 to 170 million by 2026.
It passed 500 million journeys in under three years and has run above 800,000 on its busiest days. It is now the busiest railway line in Britain and it is running close to the pre-pandemic forecast that everyone had quietly written off. The benefit cost ratio is holding at about 1.9.
There is a reason for the miss, and it is not that the modellers were careless. Demand models are calibrated on the network that exists. They are good at predicting what happens when you add a lane or a bus route. They are poor at predicting what happens when you change the shape of the network itself, because the behaviour they need to forecast has no precedent in the data they are fitted to.
New Zealand and Australia can consider this.
We routinely kill rapid transit proposals on a modelled BCR, then pat ourselves on the back for prudence.
The Elizabeth line's own numbers say that a model fitted to a radial, terminating network will systematically under-read a project that stops the network terminating.
Sydney: the tunnel gets built, then the constraint moves
The most useful comparison is the closest one. Sydney Metro City opened in August 2024 and has just published its first full year: 66.8 million trips, 210,000 on an average weekday, 99.4 per cent on time.
Look past those.
The key numbers are at the interchanges. Around 940,000 transfers a month from Sydney Trains onto the metro, and 390,000 a month from buses. At the Northwest stations, weekday bus to metro interchange rose about 40 per cent.
In other words, the metro's patronage is substantially made of people who arrived on something else. The tunnel did not generate those trips. It gave them somewhere worth going, and the feeder network delivered them. Where the feeder was good, the metro filled. The capacity constraint moved from the corridor to the interchange and to the last kilometre of footpath.
Sydney also reports 3,200 fewer southbound vehicles a day across the Harbour Bridge. A tunnel under the harbour took cars off the bridge above it, which is the network effect stated as plainly as it can be.
What this means for Auckland
Three things in our mind, and none of them relate to the actual tunnel itself:
-
The bus network is now the variable. CRL's tunnel is designed for 36,000 passengers an hour and is not expected to reach that until about 2035. Nothing in the tunnel will constrain us for a decade. The bus timetable, the interchange design and the walk to the station will.
-
Auckland rail carried more than 20 million boardings in 2017 and 13.6 million in 2025. Getting back past 20 million is not a tunnel question any more. It is a frequency and reliability question.
-
And we should publish the counts. Monthly, by station, without an OIA. We heard the cost figure for nine years. The public is owed the patronage figure for the next nine.
My honest position is that the hard engineering is behind us and the easy decisions are ahead.
Unfortunately, that is usually where these projects are lost!
Til next time,
Andrew
#cityraillink #CRL #AucklandCouncil #AucklandTransport
Transit Trends · Andrew Collings, KA Advisory. Read and comment on the original on LinkedIn.
Planning a development? Find out in minutes whether you need a transport assessment with KA PreCheck™, or estimate your trips with the free KA Traffic Estimator™.