Auckland Transport Blog

The transport section of the recently released Draft Auckland Plan makes for very encouraging reading, with the main priority being the development of Auckland’s transport infrastructure into a single cohesive network integrated with land use and development. The main ‘principle’ for achieving this (apart from a much needed look at revised and new transport funding mechanisms) is the development of Auckland’s railways into a true rapid transit system. The plan is to build the city rail link at its core and new suburban extensions at its periphery, to unleash the existing demand while promoting intensification of development in the right places to create a longer term mode shift.

Right now the City Rail Link is gathering momentum and I am confident it will be incontestable once Auckland experiences the patronage explosion that will undoubtedly accompany the new electric rail fleet. Once we have addressed the capacity and integration issues at the core, the easiest next step would be to extend the Onehunga branch via the residences and jobs of Auckland’s southwestern suburbs and the airport zone, forming a fourth main line linking from the CBD to Manukau. At this point we would be looking at a very functional rapid rail system, with new electric trains gliding seamlessly from the one side of the region, through the CBD then across the other. A network of four integrated lines sharing a tunnel at the centre carrying commuters in comfort, speed and reliability across most of the region, while providing a massive boost of access to the central city without any impact on the on the existing urban fabric. Fanstastic!

However, the question then comes: where do we go to from there? A quick glance back at the Auckland Plan shows several more rapid transit corridors, in particular routes across the North Shore via a new harbour crossing, through the outer eastern suburbs to service the growth zones of Botany and Flat Bush, plus across the upper harbour and along parts of the Northwestern motorway to provide much needed rapid transit there. To complete all of these rapid transit corridors would be the best solution we have for Auckland’s transport problems, but how to go about it?

Where to next for Auckland rapid transit, more buses and trains?

The simplest and most immediate answer is to build a series of busways, starting where they are needed the most and then expanding into longer contiguous corridors. Buses have the ability to climb just about any hill and take any corner, and can easily run on local roads where appropriate. The Northern busway is a good example of how we can pick the low hanging fruit and get some huge gains from our public transport without tackling the big and expensive issues (like a new harbour crossing) immediately.

However as we have also seen with the busway this approach is somewhat limited by it’s relatively low people carrying capacity, while the dispersed nature doesn’t promote much change in land use. It also has the unfortunate side effect of pumping tons more buses onto already congested city streets. Not exactly ideal when the goal is to decongest streets in order to work more efficiently and make them livable urban spaces. Furthermore operating rapid transit with buses can have surprisingly high staffing costs, particularly because each bus and driver can only move around 40 to 50 people at a time. This leads to high operating costs on busy peak routes, plus a tendency to cut frequencies in the off peak to avoid losing money on less busy routes.

It seems that buses are probably the best way to get the ball rolling in the short term, and we should strive for busways and bus lanes to be introduced in all major corridors as soon as possible. But to effect a significant mode shift and create a real change in land use bus based corridors can only go so far, so we need to look to the next step also.

Having discounted buses as a very effective long term solution, then perhaps the best idea to simply to expand the network through new electrified railways using the same track standards and trains as we will have already. This approach definitely has its appeal: modern electric suburban trains are fast, capacious and comfortable, they have low operating costs per passenger on busy routes, they are reliably run on their own tracks free from road congestion, and can be tunnelled under sensitive or highly developed areas. Overall rail based rapid transit is what Auckland needs to really get changes in land use and make a significant mode shift. A new rail station linked to the rest of the network by a modern train every few minutes is likely to allow people to change their travel habits, and encourage residential development and new businesses to set up shop nearby. I’m not sure if the same can be said for a bus stop on a route that leads to a busway somewhere down the road.

But railways have a critical Achilles Heel. While upgrading and integrating our existing rail lines is a very cost effective way to realise the capacity inherent in the corridors we already have, building brand new ones can be eye-wateringly expensive.

Main line railways must have particularly gentle grades and curves in order to operate at high capacity, high speed and high frequency. For example the city rail link tunnel will be at the limit of what regular trains can handle,  just to make up the rise in terrain from Britomart to Mt Eden. Auckland has had to specify extra powerful EMU trains to handle the grades of 1 in 33 in the tunnel, yet over at the harbour bridge and along the busway grades of 1 in 20 are not uncommon. At the end of the day suburban rail is built to the same basic characteristics as freight trains and intercity railways.

This means in a hilly harbour city like Auckland any new line will be comprised mostly of expensive structures like cuttings, embankments, viaducts and tunnels in order to keep the line straight and even, while threading new lines into the existing urban fabric effectively means long sections of tunnel or long swathes of properties being purchased and demolished. The irony here is that the very qualities that make new suburban train lines almost essential for Auckland are the same ones that make them almost unattainable.

Now at this point I must say that new railways are still far more cost effective than trying to provide the same capacity with new motorway developments. Given a like-for-like comparison trying to build a new railway across Auckland would be expensive, but trying to build a brand new motorway would be masochistic. Yet to be realistic the cost of new urban railways is still going to be the largest stumbling block, especially with a government so intent on wasting most of our transport funds on an economically destructive fetish for boondoggle motorways.

Light metro as a third option

This leaves Auckland in something of a predicament. On one hand we need more rail based rapid transit to get the real step change in land use and mode share we need, yet we can -for now- barely secure funding for less than ideal bus based solutions. If only there was some sort of rail system that could be built and operated cheaply without the usual constraints of main line railways, but still give much the same level of superior performance we need from a rapid transit system.

Well there is. It’s not surprising to learn that Auckland isn’t the only city to have faced such a dilemma. There are many mid sized cities like ours than need a first rate transit system without spending first rate funds. Generally this has come in the form of ‘light metro’: metro style rail systems designed solely to move people around cities on dedicated corridors free of the constraints of other heavy metro or railway systems based on the demands of freight trains or intercity carriages. In this regard these metros are ‘light’ on cost and construction, but not necessarily light on capacity or performance.  Note that the term metro is used here to refer to the service model, it needn’t necessarily be built underground like the metros of Paris or New York.

Light metro may present just what Auckland needs to extend its rapid transit system once the core suburban rail network is completed.

Introducing ART: New technology light metro

One such light metro system in the Bombardier Advanced Rapid Transit (ART), used most famously onVancouver’s Skytrain, but also found in various cities including Kuala Lumpur, New York, Beijing and Seoul. Although there are various other light metro systems across the globe (such as the Docklands Light Railway in London or the Copenhagen metro), I will use ART as the gold standard for light metro in this post. It is the most advanced and most common example worldwide and has the longest track record stretching back to the first line in Vancouver that has been in continuous use since 1985. One interesting point  is that this technology used to be known as the “Intermediate Capacity Transit System” (or ICTS), however they dropped the name once they realised it can actually provide more capacity that many regular metro systems!

So what differentiates this system from regular trains?

First of all let’s look at the main innovations of an ART type system and see why these innovations were introduced:

1)      Driverless operation

Yep that’s right, no drivers. Much like a giant horizontal elevator, the ART is controlled entirely by a central computer system during routine operation (there is a small lockable control panel that can be used during maintenance, testing and emergency situations). Because staffing is the number one cost in any transit system this has amazing benefits. Not only does it make the system far cheaper to operate, it means the marginal cost of putting on another train is low. This is basically just the cost of the electricity used, so suddenly you only need a small number of paying passengers on each train to make running it worthwhile. This means that running trains very frequently becomes affordable, and frequencies can be kept at peak-hour levels most of the time. With driven trains the tendency is to have one bigger train run less frequently to minimise the staffing costs, say a six-carriage set every fifteen minutes. With driverless trains the costs are basically by the carriage-kilometre, making it the same cost to run a two-carriage train every five minutes as a six-carriage train every fifteen. Same number of vehicles, same capacity but three times the frequency!

It also means that without needing actual people in charge, running a train at 3am on a Sunday morning or in the middle of the night on Christmas Eve is no harder or more expensive than running one on a weekday morning. Frequent operation all day long, even 24/7/365, becomes perfectly achievable. Also the lack of a drivers cab means space for more passengers in each train, not to mention a nice view out the front windshield!

 2)      Computer controlled system with rolling block signalling

On most train systems lines are split into a sequential series of ‘blocks’ to keep trains a safe distance apart from each other and prevent collisions. Generally a driver cannot enter a block until the train in front is perfectly clear of it and signals like traffic lights are used to alert drivers when they can go or when they have to stop. This works fine on main lines but can cause limitations at high frequencies, and generally if you have flat junctions on the line about the best you can get away with is a train every three minutes per track. Without human drivers traditional signalling is not needed, and the ART system uses ‘rolling-block’ signalling. Here there are no fixed blocks or signals, but the computer simply ensures sufficient stopping distance is maintained between trains at all times. It’s a bit like the ‘two second rule’ for keeping a safe following distance while driving. The end result is that ART can safely run trains every 75 seconds, including routing them through junctions.

 3)      Linear Induction Traction Motors

This sounds a bit like something out of Star Trek, but the concept is very simple. Regular electric trains have motors attached to the wheels to provide motive power. Electric motors are very elegant machines, far more simple and powerful than diesel engines. They are basically comprised of a coil of wire wrapped around a magnet attached to a driveshaft (called a rotor). If you put current through the wire coil it creates a magnetic field, this pushes against the rotor causing  it to turn and providing the force to drive the wheels. If you cut the current to a moving motor the process works in reverse in a process known as ‘regenerative braking’: moving the rotor induces a current in the coil which provides resistance for braking and converts the momentum of the train back into electricity.

A linear induction motor takes this simple concept and refines it even further. Instead of having a ring of wire the linear induction motor has it’s ‘coil’ stretched out along the underside of carriage, while the rotor takes the form of a metal plate affixed to the track between the rails. Apply current to the ‘coil’ fixed to the train and it pushes against the track itself for propulsion. The benefit here? Well firstly it means the motor has zero moving parts, thus increasing the lifetime of the equipment and reducing the cost of maintenance. But more importantly,  propulsion and braking are not limited by how much traction you can get between the steel wheel and the steel rail because the train pushes magnetically against the track itself. This means that ART trains can climb and descend grades over twice as steep as conventional trains.

 4)      Steerable bogies

On a regular train the wheels are fixed to the bogies because the rails do all the steering. This is usually quite fine, except on tight curves where the pressure of the wheel flange against the rail can result in a nasty screeching noise and cause excessive wear on the rails. Anyone who has ridden a train into Britomart will have experienced this on the tight curve around the Vector Arena. The ART design overcomes this by having wheels that can turn into the corner much like a road vehicle. The end result is a vehicle that can take even tighter curves than normal trains without any of the noise, and less track maintenance to boot.

 5)      Compact body design with third rail power.

Like most light metro designs, the ART has a relatively compact body shell, lightweight aluminium construction and third rail power supply rather than overhead line. This creates a lightweight train that can climb those steep grades, yet requires only minimal amounts of clearance in tunnels or under bridges.

 So how do these innovations translate into benefits for the Auckland context?

In many ways as it happens. Firstly the lighter vehicles and ability to take much steeper grades and tighter curves makes it easy to construct new routes over and around Auckland’s hilly, harbour side terrain. Ground level tracks can follow the contours of the land to a great extent; underpasses of roads need not be very deep, while elevated structures and viaducts could be much lighter and lower profile. Now nobody wants to see an elevated line blocking out the sky on Queen St or ruining the Domain, but in a place like Albany or Westgate it might be the perfect way to get stations right where they need to be.

The relatively small cross section and agility of the ART would make tunnelling lines a much cheaper prospect also. For example a line through the CBD could be built just below the surface using the cheap cut-and-cover tunnelling method, as the line could easily follow the contours and curves of city streets. The factors would also make it simple to upgrade existing and future busways. For example the Northern Busway would need massive reconstruction and modification to support a regular rail line, but only a simple refit with rails instead of tarmac to take an ART light metro. There is also the tantilising prospect of running a metro line over the harbour bridge, as the ART could handle the grade. This could also prove to be an effective model for other busways, such as the ones mooted for the Northwestern Motorway or  AMETI corridor. We can start with a busway at the core, then after ten or fifteen years upgrade and extend the corridor with light metro.

This has the potential to shave billions of the cost of building brand new rail lines (to the North Shore or the Botany-Flatbush area, for example), and makes linking them together with another tunnel through the city an economically feasible idea. While the ‘smart’ vehicles and track systems are likely to be somewhat more expensive than regular ‘dumb’ trains, the capital costs of constructing new light-metro alignments would be far far lower than the heavy rail alternative.

A second benefit is that driverless operation means they can be cheaply run at high frequency all day and night, without always needing high occupancy to offset costs. High frequency  means great ‘turn-up-and-go’ accessibility, so we could design bus feeder routes around bus-to-bus interconnections without having to consider connecting to any one particular train.  This high frequency also translates into high capacity. In Vancouver the Skytrain lines are usually run with just four-carriage trains, but because the come so often the peak capacity in each direction is around 25,000 people per hour. That is more than even our new EMU trains could ever achieve. Extra trains can be bought into play where and when they are needed for special events without rostering staff or paying overtime. Overall this means very affordable operating costs, which is important politically and economically. In Vancouver, a city that has lesser population density and centralised employment than our own, Skytrain actually makes an operational profit.

Thirdly, with the very fast headways and rolling block signalling made possible with computer control, flat junctions can be switched very frequently and many trains can share the same section of track over a short period. Furthermore the driverless operation means that it a terminating train takes no longer to change direction than it does to make any other stop, making it simple to operate branch lines frequently. This all provides a lot of flexibility in terms of having many lines on the map, despite only a little infrastructure on the ground. The London DLR is a good example of this benefit: this has two main lines and two branch lines linked at three junctions, but the services are typically operated along seven different patterns between various points on the network. Look at a track diagram and you see two main tracks, look at the route map on the station wall and you see seven different coloured lines each representing a separate passenger line.

Fourth, the driverless operation means that long crosstown lines become possible without concern for rotating crews or factoring in meal and rest breaks. This means we could have, for example, a line running from Orewa to Manukau all day long with it only ever stopping just long enough to let passengers on and off. That means no lengthy delays in the middle while drivers swap in and out (Melbourne is plagued by this on its City Loop), and making intermediate trips between suburbs are just as time-reliable as those to the CBD. Furthermore it almost eliminates wasted time or wasted vehicle trips, so we need less trains overall to provide the same level of passenger service.

Fifthly, the quiet motors and screech free steering make for very smooth and quiet operation, while the flexible grade and curve characteristics would make it simple to duck underground at sensitive areas. This would allow us to get stations right in close to residences and workplaces without creating noise and vibration problems, and to get routes through the city and suburbs without major impacts upon urban or natural features.

 Is all this techo mumbo jumbo really realistic, what are the pitfalls?

In short, the answer is yes. These systems have been in daily operation in Canada for twenty-five years with an exceptional track record: over 1 billion passengers carried with six extensions since 1989 and no full suspension of service for construction or commissioning. The two main lines carry over 240,000 passengers a day. The linear induction motor is extremely reliable; many of Vancouver’s original 1985 Mk I trains have accumulated over 3.8 million kms with only one minor overhaul of the motor and are still going strong.

There would no doubt be various objections to introducing new light metro line to Auckland, even if the initial hurdle of political and public scepticism could be overcome. The main issue is perhaps the lack of interoperability, for example a line on the North Shore  could not run into the city rail tunnel, nor could it take freight or intercity trains to the north of the country. In a way this is actually something of a benefit, the single urban-transit mode would ensure regular high frequency operation could not be disturbed by other transport uses. In the first instance connection to the other lines using the city rail tunnel should be made in the CBD and wherever else possible, but this should only be by passenger connection rather than by trying to run everything through the same set of tracks. In the second instance there already exists heavy rail lines heading north and south out of the city, and maintaining these for freight and long distance passenger access is no doubt the best idea. A new metro line would need a new stabling yard and maintenance facility, however this is likely to be the case too with any suburban rail extension.

Perhaps the best way to frame this issue is to consider a heirarchy of rail and public transport, each stage being ‘sectorised’ from each other. The first level is that of freight, regional and intercity trains, these would operate from the freight yards and Britomart terminal, using the main trunk lines to head north and south of the city. The second level is that of the suburban rail, using the existing and proposed suburban rail network and operating through the city rail tunnel very frequently at peak hours and approximating a metro system at the centre. The third level is that of our light metro, providing urban passenger-only services separate from the suburban lines. but directly interconnected with them into a wider rapid transit network. The fourth level would be street level bus and tram services, providing local access and feeding into the higher levels.

In summary

A light metro system such as Bombadier’s Advanced Rapid Transit could represent a way to establish high quality metro style rail routes across Auckland at a fraction of the capital or operating costs of conventional heavy rail or underground metro systems, meaning more lines could be built to more areas in a shorter time frame given the same amount of funding. Lines with low capital and operating costs yet frequent high quality service would no doubt perform well on any benefit-cost analysis, potentially making it much more feasible to secure funding for them.

After the essential City Rail Link is built and our existing rail lines are being used to their maximum potential, we will need to ask ourselves “where to next?” Do we look at developing the next suburban heavy rail line in Auckland, or the first metro line instead?