It’s impossible to be in transport circles for too long before the good old arguments of ‘buses versus trains’ or ‘BRT versus LRT’ and so forth come along. I suppose that it’s inevitable to end up in such technological debates – as while each technology tends to be best suited for a particular type of job, there’s an enormous level of crossover. For improving Dominion Road’s public transport priority measures, for example, we could focus on better bus lanes or we could look at light-rail. Both options have their pros and cons, there’s not an obvious solution – at least not at first glance.
The debate is also complicated by romanticism to some extent. Trains and trams have a history, or perhaps better put – they have a culture associated with them that buses seem not to have. This may mean that a rail-based solution can sometimes be promoted when buses would do just fine; but the on the other hand evidence seems to show that trains and trams do attract ridership that buses simply don’t attract. So there is a level of attractiveness to rail-based solutions that perhaps cannot be explained through the analysis of numbers.
That said, public transport advocates who are blind to the benefits of buses do themselves no favours by being fixated on technology: buses are cheap, flexible, fast-to-implement and so forth. As I have said on numerous occasions before, if we wanted to do one thing cheaply and quickly to dramatically improve Auckland’s public transport system – the best thing we could do is drastically expand the city’s bus lane network.
But what’s important here is a ‘horses for courses’ approach. In some situations our existing infrastructure, the type of demand along the corridor, the land-use patterns and whatever other relevant factors there are will combine to tell us that buses are the best solution. In other situations they will combine to tell us that rail will work best. There are likely to be further subdivisions too: light-rail or heavy-rail? What level of bus priority do we need? Is demand so low that what we really need is some sort of demand-responsive van system?
A common criticism made by bus advocates is that those promoting rail solutions are only doing so because of the romanticism associated with rail. Bus advocates like Australian David Hensher (who commented on this blog once, though unfortunately he never replied to a few questions I put to him) can sometimes fall into this trap. Mr Hensher puts together excellent critiques of how we cannot continue to have such roads-centric transport policies, and he calls for a balanced approach to PT that focuses on cost-effective improvements and a blind-eye to technology. But then he goes and throw away all that good work by forever focusing on how fantastic bus rapid transit (BRT) is, and slamming any other technology.
For example, in this document even the abstract shows quite clearly his contrary approach:
Right up until the last sentence I could not agree more with what he says. An integrated multi-modal system based around frequency and connectivity – I absolutely agree. And BRT, with feeder buses may well provide what we’re after in some circumstances. It seems to be working pretty well on the North Shore, and I’m supportive of Auckland Transport’s plans to create a busway between Panmure and Botany as part of the AMETI project. In the right situations, BRT will be the solution.
But that doesn’t necessarily mean buses will always be the right technological solution to a transport issue. And this is where Mr Hensher falls into his own trap, decrying others for focusing on a particular technology but then doing so himself. It is a pity actually, because Mr Hensher’s analysis of the need to find cost-effective public transport solutions is very sound. This is outlined in another paper of his:
There is growing support for an attractive alternative means of transportation to the car in cities. If increased public transport capacity is the way to proceed, it is very important that the investment in such systems is made in a rational way. There is a need for sensible selection and funding of technology and consideration of appropriate ways of addressing the problems attributed to the automobile.
I couldn’t agree more. Especially in New Zealand where our central government is focused on building crazy motorways everywhere we will need to make sure that the scraps available for public transport are spent carefully. He also says (from this article again) good things about the need for a multi-modal approach:
Public transport investment is being touted as a key springboard for a sustainable future, especially in large metropolitan areas with growing populations. Public transport, however, is very much multi-modal and should not be seen as a single mode solution as is so often the case with many ideologues. Hence, any commitment to improve public transport has a growing number of options to pursue.
The problem is he then goes on to argue that buses are what you need in pretty much every situation, that rail is too expensive and so on. This from his other article:
There is growing evidence around the world, in origin–destination density contexts similar to the locations proposed for light rail, that a dedicated BRT system (i.e. road infrastructure dedicated exclusively to buses as in Brisbane, Curitiba, Bogota, Pittsburgh, Ottawa etc.) can carry the same number of people as light rail for a typical cost 4–20 times less than a LRT system and 10–100 times less than a heavy rail system (USA General Accounting Office, 2001). It is flexible, it is as permanent as light rail and it can have the image of light rail (rather than the image of boring buses) if planned properly. The USA General Accounting Office (2001) audit of BRT and light rail in six US cities found that the capital cost per mile for LRT compared to BRT in its own lane was 260% more costly. Comparisons with BRT on street or on an HOV lane are not useful and have been excluded. When one also notes BRT’s lower operating costs for both institutional and maintenance reasons, the case is clear.
If one was building a city from scratch and deciding what transport technology would offer the best bang for your buck, then all of what is said above would probably be incredibly important to consider – and chances are you would think very strongly about the credentials of BRT as a ‘system-wide’ solution. However, in the real world that isn’t the case, we have to deal with issues like the following:
- What existing infrastructure do we have and how well utilised is it?
- How much physical space do we have to upgrade this corridor?
- What are the existing land-use patterns and how might they be affected (positive and negatively) by the different technology chosen?
- How do operating costs compare at different demand levels, and in different countries?
If we take the CBD Rail Tunnel project for example, the ability to generate its benefits from bus improvements seems ludicrous. One of the main benefits of the rail tunnel is to unlock the capacity of the existing rail system – enabling it to carry far more people and therefore take pressure off the roading system. Any bus solution that would increase capacity to the CBD would probably increase congestion (not only of buses, but also of cars as many streets would probably need to become bus only) and would only add pressure onto parts of the transport network with no spare capacity.
It’s also underground, reflecting the lack of physical space (but cheaper than a bus tunnel) and supports existing land-use patterns by encouraging the concentration of economic activity in the city centre, and the intensification of suburbs along the rail corridors. Finally, one would imagine that the operating costs of running electric trains that carry up to 1000 passengers with only one driver (here’s hoping) is likely to be lower than carrying those 1000 people on about 15 diesel buses – especially if fuel prices increase further in the future.
Another example where the bus versus train debate often flares up is how to provide rapid transit to Auckland Airport. The current PT situation for the airport largely involves the Air-Bus: which is a reasonably decent bus service that runs along the motorway for much of its trip, before travelling along Mt Eden Road (soon to be Dominion Road as well) on its way to the city. With the opening of the Mangere Bridge duplication project last year, the motorway part of the trip is probably fairly congestion free (at least for now, induced demand will probably eat away at this within the next 2-3 years), with the route along the arterial roads likely to be slowest. This is relevant because there are effectively two “parts” to the airport to city route:
- between the city and Onehunga
- between Onehunga and the Airport
Assuming that we consider it highly important to offer people a one-seat ride between the city and the Airport at a rapid transit quality, that effectively means we either need to find a way of providing a busway, light-rail or heavy-rail line along that whole distance. And here’s where the importance of our existing infrastructure shines through. Anyone got a good idea about how to thread a busway or light-rail line between Onehunga and the city centre? Aside from taking two general traffic lanes away from the southern motorway (not that I would necessarily oppose such an idea) the proposition of an additional RTN between Onehunga and town, duplicating our existing railway line, is not only impractical, it’s also pretty stupid. This tilts the arguments towards rail as the preferred solution for rapid transit to the airport very significantly.
The excellent blog “The Transport Politic” looks at this argument in quite a bit of detail in a recent post – usefully pointing out the following:
What is clear is that for the majority of American cities — excluding only a few in the Northeast — buses will remain the predominant mode of public transit for most riders, even after major expansions in train networks planned for cities from Charlotte to Phoenix. So even cities that choose to invest in rail projects must also spend on the improvement of their bus lines.
Nor is the difference in costs between rail lines and BRT nearly as great as some would argue. The Journal article quotes Dennis Hinebaugh, head of a transportation center at the University of South Florida, saying “You can build up to 10 BRT lines for the cost of one light-rail line.” That might be true if you’re comparing a train operating entirely in its own right-of-way with a bus running in a lane painted on the street. But a streetcar is probably cheaper than a busway. Just ask Hartford, whose busway project will cost $60 million a mile to build.
One thing that can make bus solutions cheaper is that they’re easier to have short-cuts. If we think about the Northern Busway, it’s really only a true rapid transit corridor for part of its length. It shares a lane with general traffic across the Harbour Bridge, it doesn’t have full priority at intersections in the CBD and between the harbour bridge and once again it shares motorway lanes with general traffic between Constellation and Albany stations. While a railway line would have required a higher standard of alignment geometry and therefore more earthworks, the primary reason the busway is way cheaper than a rail option is due to the short-cuts it has taken.
The Transport Politic’s post also points out the primary situations where rail offers something that a bus solution simply cannot (while also pointing out that these scenarios are relatively rare):
The best argument for rail is that it has the ability to provide massive rush-hour passenger-carrying capacity without destroying the city through which it runs. Whether buried in a subway or operating quietly along in grassy medians, trains can be integrated into the public realm without diminishing the pedestrians-friendly qualities all urbanists should hope to encourage. BRT boosters often argue that their mode of choice can carry a similar number of riders, but neglect to mention that this is only possible when buses arrive every 10 seconds along highway-like four-lane corridors. These are conditions that destroy the walking environment.
Fortunately for American cities looking to invest in new public transportation infrastructure, there are few places that demand the passenger-carrying capacity provided by those freeway-based BRT lines in places like Bogotá. In most metropolitan areas, a two-lane busway inserted on an arterial is perfectly appropriate and sometimes even beneficial for a city. Indeed, as we all know, the story that is too complicated for any mainstream paper to explain is that BRT can mean any number of things. The most rudimentary elements of BRT — the nice buses, the well-articulated stops, the traffic signal priority — are basics we should expect from all of our bus lines. Pushing for their implementation along certain corridors shouldn’t arouse much controversy.
Neither buses, trains nor trams are inherently better than the other. Each technology has an important place in developing a proper public transport system – especially in a city like Auckland. When deciding on our preferred technology cost-effectiveness is obviously going to be a key factor, but cost-effectiveness does not simply mean “which option is cheapest to build?” It also means which option best utilises our existing infrastructure, which option will integrate best with the surrounding environment and which option will make economic sense in the long-run. In some situations that will be a bus, some a train and some a tram.
Anyone who thinks that one particular technology is suitable for all situations – while criticising others for being obsessed with their favoured technology, is really falling for their own trap.
I caught the Brisbane Busway this morning and I can tell you that this is NOT true. The SE busway cost approximately 50 million/km to construct, however the newer sections of the busway are now even more costlier than this coming in at around $100 million per kilometre. There is a section of busway at Buranda which is tunnel & bridge which is costing… wait for it… $465 million dollars (so half a Billion almost!) for ONE KILOMETRE. Is is costly because one vehicle has steel wheels and the other one rubber tyres? Of course not- it is costly because the right of way is Class A.
Also, to reach high capacities say beyond say 5000 – 6000 pphd (a 200 seat bus every 2 minutes or so) , I think busways become inefficient relative to rail because you have to employ a lot of people to operate the system.
The Gold Coast Light rail system, over the life cycle, also came in slightly cheaper than BRT.
There is no universal magical mode. Broad sweeping statements should be carefully looked at. The SE Busway really should be upgraded to metro.
I think lower initial costs will always be more important to politicians as the average polticians career is likely to be far shorter than life cycle of the asset.
I think there needs to be both short term (stop-gap) and longer term plans. The way I see it is like renting a house vs buying a house.
Not everyone has almost a million dollars to buy a house outright from day #1, and that is understandable.
The way I see it, BRT is great because even at low demand you can have really excellent frequency. And good frequency is really important. As the demand grows, the frequency gets closer and closer to a point there you can switch modes to something higher capacity like light rail or trains without a large drop in frequency (if you ever need to- some places will never need to do this). I think this swich-over point is around a high capacity bus every 2-3 minutes.
I still think the SE busway was one of the best and greatest investments in PT that Brisbane came up with in a long time and was the correct mode of choice at the time. That said, there comes a time to move on, and with a subway system we could double the capacity of that busway. And we need that capacity- That busway carries 88% of the entire daily patronage of our 149 station train network over the whole region carries. During peak hour the demand comes close to pushing the lower bounds of capacity of metro systems.
The SE Busway is so pretty and lovely to ride though. It’s like a dream busway (seriously, it would have to be one of the best of its type in the world).
It’s comparable to rail in a great number of respects.
I don’t understand the point about existing infrastructure. Seems like you are proposing painting some bus lanes on the streets there now, versus building a separate rail line. If you have the right of way to build a rail line, why couldn’t you build a BRT route in that right of way? Or is it that there is already a rail line between the airport and the city? For the record, I’ve heard airport to city rail links are often quite costly to build and maintain with limited ability to change the private auto vs. PT mode split. An airport-city link in NZ might be the worst example I can think of to use to claim ‘horses for courses’ favours rail.
Costeffective, there is an operating rail line half way to the airport, CBD to Onehunga, so a direct route in this mode is already halfway there, cheaper to continue this on than to somehow ram a dedicated BRT to the CBD. But also a rail line through Mangere is not just for the airport it would provide this whole area of Auckland it’s first direct links to both the CBD and south east to the Manukau city centre free from congestion. This line will effectively move not only the airport but also the employment and dormitory suburbs of this area closer to both the CBD and Manukau city centre, whatever the state of the traffic, and, of course, the price of petrol.
Not to mention allowing Puhinui to be infilled with housing or industrial and not be totally reliant on the motorway. I wonder if that could be a way of helping to pay for it. Charging development levies on new developments along the corridor.
One further issue which isn’t raised by this post is that electric rail is more established technology by a long way than electric high-speed buses. I’m informed that trolley buses can’t run much quicker than 50km/h, and the question of running on renewable energy must be a very important one for sustainable long term PT.
Some quick thoughts:
1. Average operating speeds on Auckland’s bus network are about 20km/hr, so I don’t think the 50km/hr limit is particularly crucial. Even on the NEX buses probably only exceed 50km/hr for very short time periods.
2. I’m fairly sure this can be resolved through better surface engineering, as I imagine the issue arises when the buses lose contact with the overhead wires. There’s no reason why a BRT surface could not be identical to LRT …
3. Electric trolley buses could be fitted with batteries to overcome the effects of jumping “off line” (which by the way would be useful in places where they need to make short deviations “off line” (not so good for LRT, because the weight of the units creates a high load)
Oh yeah, I see the part about the CBD to Onehunga rail link in the article too now. Thanks.
I guess the issue now would be what’s the cost of going from Onehunga to the airport and what’s the benefit upgrading from the current AirBus to the new rail link. I don’t think it makes sense to automatically assume “that we consider it highly important to offer people a one-seat ride between the city and the Airport at a rapid transit quality”. That, coupled with the argument that it’s almost impossible to build a BRT on the CBD to Onehunga section, seems to require a rail link. At the risk of angering everyone, I wouldn’t be surprised if it didn’t make economic sense to build a rail link between Onehunga and the airport. Sorry, I don’t mean to be difficult here. And I’d certainly take the link if it was there.
I don’t think anyone would necessarily wail on you for suggesting that it needs to be looked at carefully economically as I think everyone here would agree that all projects need to pass the sniff test.
Sadly our Minister of Roading doesn’t seem to apply the same principle to his projects.
It’s a justifiable debate as to whether we want an RTN quality one seat ride from the Airport to the CBD. I would argue that anything but that isn’t much improvement over what we have now, so it’d be dumb to spend a lot of money and basically not improve things greatly from the current situation.
Whether that’s cost-effective or not would probably need to wait for the business case.
All this talk is too airport focussed for me…. I see finishing the loop from Onehunga to Manukau City being much more about fixing the disconnection and severance of the greater Mangere area as well serving the travelers and workers at the airport. And amplifying the network itself. There is so much underutilized capacity on the rail network which means that until both frequency and reach of the network are extended we have an underperforming asset.
And, unlike the current government, I see switching to the plentiful and renewable electric fuel source as a matter of some urgency wherever possible. I see their complacency around resource supply and climate change issues as recklessly optimistic.
I see their complacency around resource supply and climate change issues as recklessly optimistic.
Agreed entirely, although I think Labour are almost as bad on this one issue (although obviously a little more supportive of PT initiatives.)
I agree Patrick, I see the value of RTN in that corridor as firstly providing a core rapid transit link to the south west, secondly creating integration between the existing parts of the system to create a true network (rather than just a series of Faisal commuter lines) and only thirdly providing RTN access to the airport and it’s environs. All three are worthwhile goals, but the third on it’s own wouldn’t stack up. All three together make a very compelling case.
I’ve been thinking … is the busway all that impressive in terms of how many people it is moving?
The Northern Express looks like it’s bursting at the seams, yet stats show that it carried 172,187 passengers in April – only 58% of the patronage of the Western Line. So is its capacity all that great?
How many people would we be able to move if that corridor could take light rail as well as buses?
I concede that that number doesn’t include other buses using the Busway, but I believe the NEX makes up the bulk of the Busway’s patronage.
Despite the dedicated corridor (for part of the way) the NEX is not really run like a RTN. If we ran it like a true RTN we could get a lot more capacity out of the busway.
Upgrades I would recommend:
– RTN style ticketing (fare paid zones controlled with electronic gates)
– All door loading/unloading
– High floor (multi?) articulated buses with level boarding (from platforms), and many large doors (4+), also a greater proportion of space dedicated to standing rather than seating.
– Aggressive bus priority measures on non bus-way sections of the route.
These upgrades would give much higher capacity without dramatic changes to the fixed infrastructure.
I was quite surprised to see low floor buses running the NEX, I think Medium or high floor busses would do the job better at lower cost. The only disadvantage is that they would only be useful for the RTN, limiting end of live (i.e. schoolbus) applications.
Don’t disagree with your suggestions (apart from one), but we can upgrade the capacity of the Northern Express all we want, but if more people are actually going to use it there need to be frequent connecting services and intensified development around the stations. In my experience the majority of people using the NEX at the moment seem to be being picked up and dropped off at the stations.
It would be rather strange to move to high-floor buses for the NEX when all new buses in Auckland are moving to low floor. High-floor buses would slow boarding and disembarking and reduce accesibility.
“High-floor buses would slow boarding and disembarking and reduce accesibility.” Why? – with bus platforms allowing level boarding I think boarding would be faster and acceptability would improve.
Low floor articulated buses are complicated and expensive to build, Probiabily why we don’t have any in NZ. also with 4+ doors some would need to be in the high floor area anyway (got to fit the engine somewhere). The NEX doesn’t have many stops, why not build bus platforms to match the floor height of the buses?
sorry it should be accessibility in my first paragraph. Look at the platform height in the following photo:
The only high-floor buses I’ve seen here all have steps, because there’s no high platform bus infrastructure in Auckland.
The busway stations already have low floor platforms and waiting rooms, so all of the stations would need to be reconstructed, and the on-street stops would need to be raised, and somehow integrate with the surrounding pavement.
I think all that building would negate any cost advantage of high floor buses. Low floor articulated buses are common overseas, there’s no reason we can’t build them here.
A couple of weeks ago I was in Yogyakarta Indonesia, there they have two bus systems, one is the regular slow side of the road stop system, the second is an express system that uses flat high floored buses and elevated platform stops. Neither has so much as a bus lane, let alone a dedicated right of way, but the high floor buses did benefit from very quick loading times and offline fare collection. Interestingly every high level mini station was only accessible by steps, so they missed a great opportunity to provide accessibility to the mobility impaired.
Not exactly sure if the benefit of high level is worth it in Auckland, even to covert just the NEX would require every station and stop it uses to be raised, plus you would be left with the awkward situation of having two levels of platform at each stop and station as they are all used by NEX and local routes too. Really I can’t see what is wrong with the current arrangement of low floor buses and low platforms where two thirds of the bus is level boarding anyway. More important would be looking offline are collection at stations and major stops (easy to do with Hop I assume), and moving to buses designed with double doors front and back (to allow up to four people to board or alight simultaneously) and perhaps a more ‘metro’ style seating arrangement with greater focus on standing and circulation room in the front half of the bus.
As far as I know, the NEX outperforms the rail network on almost every measure. Patronage is mainly a factor of the length of the line and the number of stops. So because the western rail line is longer and has more stops it will generate more patronage, so total patronage is not really a good measure of effectiveness.
Effectiveness is hard to measure. If we look at peak hour trips, we see that buses carry about 6,000 people over the harbour bridge in the peak. That’s approaching a 40-50% mode share, which I bet is a lot more than the western rail line.
Much more in fact. In terms of capital costs, the busway was built for about the same cost as the cost of the station upgrades on the western line. So that excludes costs of double-tracking, electrification etc – which we need to get to do to even 10 minute peak rail frequencies. Meanwhile the NEX chugs along at 3min frequencies with no major probs.
In terms of operating costs, I don’t have access to good data. But the last time I looked the NEX required about half the operating subsidy per passenger as the rail network. Since then patronage on NEX has grown faster than the rail line so I expect the gap would be even wider now, although the rail network has recently been catching up.
So the short answer to your question is “yes” 😉
Personally I often wish there was more objective information out there relating to the different transport modes. You have all these self proclaimed experts arguing in favour of a particular mode and it is often hard to tell which is really the best solution for a particular problem. It woudl be nice if tranpsort agencies would do (and publish) more evaluations of this type of thing…
Lucy, the Arta report on the south west RTN corridor (i.e the airport line) is currently published on the Auckland Transport website. This includes a fairly thorough comparison of BRT, light rail and heavy rail options for the route.
Agree completely. I’m doing some academic research at the moment on this, which will be presented in October at a research conference. Some initial thoughts:
* BRT can handle up to 3,000 passengers/hour/direction without significant grade-separation – my sample size does not have too many projects busier than this – and where your market does not want to transfer within the journey. It is also recommended where existing road space can be taken over and dedicated to BRT – I have a couple of examples where whole motorway lanes were turned into dedicated median bus lanes, and these systems have very high capacity (>8,000 pax/hour/ direction). Buslanes down road medians appear to be very promising, although this is something of a culture change.
* LRT is recommended, all other things held equal, where: there is existing rail infrastructure you can press into service; above 4,000 pax/direction/hour; where localised environmental conditions, like tunnels, require an electric mode; and where the middle class will simply not use the bus. This is not an issue in the UK and possibly not in New Zealand, but it certainly is in the USA.
In terms of costs, across my sample: LRT is over three times as expensive per mile as BRT, and that seems to be consistent when you control for project scope.
Ross, I agree with your analysis in general, except on localised conditions favouring LRT. In my limited experience localised environmental conditions often favour BRT rather than LRT, e.g. areas with steep climbs and sharp turns. In fact these crop up far more often than the tunnel example you mention. For example, the gradient on the harbour bridge, for example, is one factor why BRT was more suited to the Northern Express.
Go PUBLIC HELICOPTER as my pet mode !!!
Great for low density and poorly connected car dependant streets. Just walk or cycle or catch the bus to your nearest vertiport. Public Helicopter page —> http://www.britishhelicopterassociation.org/public_transport.aspx
(PS: this is a joke by the way)
@Ross seem to be over-simplifying things a bit there. For me LRT and BRT would be used for totally different corridors. Wouldn’t want BRT down busy urban streets like Dominion Road or Queen St, but these would be great for LRT.
Also with BRT have to think about bus congestion in the CBD, may work for a small portion of city – CBD trips; but if its scope its too big becomes a mess.
I also wouldn’t build LRT in low density suburbia but BRT and Heavy Rail can work well here. heavy rail is better for corridor cities like Auckland and Wellington, but for very spread out cities BRT would work better.
BRT can be staged much better than rail, even short sections like the Panmure section can deliver big benefits.
However on longer journeys with high patronage the system would become very confusing, and as real mess.
This is where heavy rail has an advantage.
Also where heavy rail exists building/upgrading a heavy rail system is far preferable.
I think there’s an element of correlation not causation creeping into your arguments here, which is exactly what Josh was warning against in his post. There’s no fundamental reason why BRT could not be operated as simply as LRT or heavy rail. The reason BRT networks can become complicated is because the technology is flexible (which is actually an advantage) – which means that service patterns are “tweaked” in a way that makes them complicated. The NEX extension to Massey University in Albany is a good example. But it must be recognised that these “tweaks” are an active choice made by people rather than being inherent to BRT technology itself, ie NEX could just have easily terminated at Albany Station.
Exactly. Nobody would try and drive down train tracks. But because buses have tyres and look like things people are used to sharing roads with, they are interpreted in quite a different way.
Hi Stu, The NEX extensin to Massey never happened, they do terminate at Albany Station.
Not but they run NEX services there, which somewhat complicates the route structure and timetables.
Luke, Stu – thanx for your comments – I’ll be back on when my thoughts on the matter are better organised.
*currently working on the written paper – now understands why editing your own work is known as “murdering your darlings”*
I really liked this post on the Human Transit blog a little while ago. There are differences between bus and rail but often they get misidentified or are a result of communitys attitude. Its important to make the distinction.