Energy & GHG

Dec 4, 2012

Is public transport as green as you think?

It's true emissions from public transport aren't as low relative to cars as is popularly imagined, but the savings are still significant. And there are other good reasons to invest in worthwhile transit services.

Alan Davies — Editor of The Urbanist

Alan Davies

Editor of The Urbanist

US energy use per passenger mile (Btu) Source: DoE Transportation Energy Data Book, 2012

There was one of those intense internet fracas last month that generate lots of heat and light yet pass by completely unseen by the non-connected world.

It started when Eric Morris argued that, contrary to the conventional wisdom, public transport’s emissions aren’t markedly lower than those from cars (see Save the earth, drive your car? and Can mass transit save the environment?).

He says that’s because trains and buses are heavy and use a lot of energy, but have low passenger loads most of the time.

There were a number of critical responses like this one from Angie Schmitt and this one by Michael Lewyn. The author of Human Transit, Jarrett Walker, also weighed in with a couple of fighting articles, The same “empty buses” fallacy, over and over and The “transit isn’t green because it runs empty” line.

Jarrett Walker’s counter-argument is essentially that the mission of transit agencies isn’t to maximise ridership. Equity objectives oblige them to provide services in areas that can’t be serviced efficiently. Such areas have “predictably low ridership” which increases emissions per passenger kilometre.

By and large transit systems perform considerably better on emissions than cars, but the gap isn’t as big as is often imagined. The US Department of Energy estimates transit rail consumes 37% less energy per passenger mile than cars, but buses consume more (see exhibit).

In 2008, Victoria’s former Commissioner for Sustainability said modes that rely on coal-fired electricity, such as Victoria’s trams and trains,

have GHG full fuel cycle intensity levels on an average per person-kilometre basis that are comparable to motor vehicles.

This report prepared for the 2008 Victorian Climate Change Summit shows the GHG intensity of cars in Victoria is only 13% higher per passenger kilometre than either trains or trams. Emissions from buses are 27% higher than those from cars.

The explanation for the relatively modest difference between cars and public transport is partly about factors like how the electricity is generated (e.g. Victoria uses brown coal), but is mainly the result of low load factors.

Those celebrated photographs comparing how many cars, buses and bicycles are needed to transport a fixed number of people give a misleading impression because they assume the bus is full (and the car is at average occupancy).

That’s largely true of trains and trams in the peaks, but not during the middle of the day, on weekends and at night.

A train that starts out full from the CBD in the evening will also progressively shed passengers as it approaches its destination. Another factor is peak hour public transport vehicles usually return empty or with just a few counter-flow passengers after they’ve reached the terminus.

Buses rank worse because they tend to be relegated to low volume feeder services and marginal routes. When buses are used for trunk services as they are with Bus Rapid Transit, they perform much closer to rail.

If we only consider services with high load factors – say peak hour CBD services – then trains and trams are much, much more sustainable than cars, measured in terms of passenger kilometres.

The performance of public transport also improves when lifecycle emissions are taken into account. Trains and buses don’t sit in parking spaces most of the time – they’re used for the greater part of each day.

Further, public transport enables us to live and work in denser locations. All other things being equal, density means we travel fewer kilometres and live in smaller dwellings that consume less energy e.g. to heat and cool.

The relative performance of trains and trams will also be better in cities where the electricity supply is cleaner e.g. a high proportion of hydro power.

On the other hand, if likewise we only consider the sorts of car trips that tend to have high occupancy levels – like non-work trips – cars will perform better too.

Public transport is necessarily more than crowded peak hour services – it has to be viewed as a system. As a society we expect governments to provide frequent services all day and late into the night when patronage falls off.

We demand a decent standard of service in locations like outer suburbs where public transport struggles for ridership because it’s very hard to compete with the flexibility and privacy offered by cars.

Nevertheless, I think we have to acknowledge that public transport isn’t as green as is customarily assumed. The fact that low patronage services address important social objectives doesn’t change that equation. It shouldn’t lead us to ignore the facts.

I think this offers an important “so what?” insight. Environmental sustainability is very important but it isn’t the sole purpose of public transport.

As a society, we invest billions of dollars of public funds in transit in order to provide mobility for those who have no other choice. We also do it to sustain dense agglomerations of activity. We’d continue to do so even in the absence of climate change.

It’s of course highly benficial that mass transit is more sustainable than driving. However we shouldn’t for a moment shy away from providing public transport services where they’re warranted because some types of services have (relatively) high emissions.

The real fallacy is the idea that we have to judge the sustainability of every activity equally, irrespective of the cost of mitigation or the value it delivers. It’s erroneous to think only in terms of costs and ignore benefits.

What we should do is acknowledge some activities are so socially beneficial that their relatively modest – or even poor – environmental performance is acceptable. I’d propose many “marginal” public transport services in outer suburbs should be in that category.

At the same time we should focus on improving the environmental performance of activities where the pay-off is high relative to the financial and political cost. It’s not wise to fight a war with equal intensity on all fronts.

(Visited 23 times, 1 visits today)


Leave a comment

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

21 thoughts on “Is public transport as green as you think?

  1. Bill Bunting

    We are at the cusp of some major changes in transport, but these changes are largely at ground zero. Firstly there is the electric vehicle, slightly above ground zero with the Prius and moving forward with the Volt and others. But to me the most significant is the up coming, slated for production 2013, VW XL1 100klms/ltr diesel electric hybride 2 seater. This vehicle will drop commuter fuel costs to 12% of current consumption for single driver long commutes.

    The second major change is the autonomous drive which is already being partially installed in production vehicles in the form of space managing cruise controls and autonomous parking systems. With both of these elements in the same vehicle most of the autonomous hardware is built into the vehicles (speed, braking and steering servo control along with the speed and vehicle near proximity spatial perception). The rest of the Autonomous control can be a retrofit from that hardware status.

    For me the merging of the VW XL1 and autonomous control represents the ultimate freedom from the tyranny of driving, and it achieves this with a fuel efficiency that is a quantum leap improvement in fuel consumption/cost and CO2 emissions. Furthermore this vehicle weighs in at 800kg (half the resources required for production) much of which is composite (Carbon as the main resource consumed) for the body and aluminium (readily recycled) for the engine complex. Should the Envia battery of some other comparible prove successful then the addition of 50kg will give this vehicle an electric capacity over 20Kwhrs which will give it a very substantial all electric range in future versions. So you get the very best of range, fuel flexibility, minimum energy consumption, solar energy compatibility, CO2 emission reduction and commuter functionality. Furthermore the XL1 is a narrow vehicle which will provide advantages in both road space and parking.

    Now rethink those figures in comparison to busses and trains and we are looking at a very different possible future world. Once the miniturisation of commuter transport gets a foot hold, and as fuel prices continue to rise, the SUV ballon in vehicle sizes will be arrested. Family vehicle will remain a feature of the future but they will begin to reduce in size, I predict.

    The other feature of Autonomous Control is that there is the very real prospect for driverless taxis and mini busses. One of the failings of our public transport is the complete lack of mini busses in favour of the “one size fits all” 12 tonne buss. It is way past time that got a rethink.

    I for one look forward to 8 or 10 seater driverless all electric microbusses which can be called at will by text message to my phones location and pay for that trip with an Octopus type debit card. This is a very real prospect which could be in use in 6 to 8 years. This is the best future for local public transport interlinking with long haul or underground rail systems. I see this as being fully compatible and commercially viable beside the VW L1 class personal commuter vehicle.

  2. Mena

    Studies must have been done looking at the benefits of avoided traffic from those using public transport and bicycles. This would have a positive benefit on the emissions of the cars. I catch the train because I cant be bothered sitting in Perth’s seriously horrendous traffic. However, if I, and others like me, decided to drive, the whole system would grind to a halt. The emissions from cars would surely increase from the additional time it takes to get where they are going. This is an extra benefit of public transport eh?

  3. sanjay muktar

    The full story of how the government of NSW got rid of Sydney trams would be an indication of the corrupt ways our leaders use to justify their transport decisions and the effect they have on air pollution now.

  4. Dylan Nicholson

    Krammer56 “like most people I expect, I use all three depending on the trip”

    If only…

  5. duke the lost engine

    sure we care about marginal emissions per passenger-km when we’re considering choices to increase patronage on existing services – eg our personal mode choice, or policies like densifying around transit, changing price incentives.

    but when we’re considering whether to introduce new transit services, we must grapple with average emissions per passenger-km (more specifically, average emissions/pkm on the new service being considered).

    generally, i think we should be more seriously grappling with the inevitibilty that moving people consumes energy, and if we’re serious about emissions reductions, sometimes the best policies will be those that encourage less travel

  6. Smith John

    Alan #15

    If the marginal user tips the need for an extra service, you’re still looking at average vehicle occupancy of over 50 per cent, which is probably much higher than the all day average of most Australian PT at present. So there is probably considerable potential for putting more people on existing services at zero marginal cost, at least in off peak times.

    Certainly vehicle/km of service is a strong predictor of patronage; more vkm (especially bus) are needed for serious increase of PT use; but that by itself won’t necessarily increase average vehicle occupancy.

    But there are things that a well run PT system can do to increase patronage independently of vkm – real time information systems, better network design to support’anywhere to anywhere’ travel within the same total vkm budget; infrastructure to improve the speed and reliability of trams and buses.

    These things will tend to increase average vehicle occupancy. They are more likely to happen in an environment of increasing PT use.

  7. Alan Davies

    Smith John #14:

    Provided, that is, you’re not the marginal passenger who tips the need for a whole new bus, or worse, train! Best to think in terms of ‘average marginal (energy etc) cost’, I think. This is getting into the “if the plane is going anyway is it OK to use it?” debate (which is not a bad thing).

  8. Smith John

    ‘Energy use by mode’figures vary widely between sources and should be approached with caution unless they explain their inclusions and assumptions very clearly.

    Since the interst of the comparison is in assessing options for change, the marginal user is more important than the average user.

    The comparison with respect to the marginal user is likely to be much more favourable to public transport since:

    – as car use increases, energy use per person/km is likely to increase because of congestion;

    – as public transport use increases, energy use per person/km is likely to decrease because of better average vehicle occupancy; and because of the efficiencies of things like bus lanes reducing congestion for buses, which are likely to go with increased PT use.

  9. Krammer56

    One way to look at it is that you can make a PT network greener by increasing patronage – if you increase car use it just goes the other way! Maybe this suggests where we should be putting our effots and money!
    Of course, bikes are better again, but like most people I expect, I use all three depending on the trip.

  10. Dylan Nicholson

    I would accept though you’d really have to go all out with a PT network vastly more comprehensive, frequent and reliable than we’re close to now to get to the point you could actually make significant reductions in the amount of road infrastructure and number of cars we might need. At best we can hope to reduce the need for more roads and more cars as population increases.

  11. Dylan Nicholson

    IkaInk, on your last point, provision of good public transport systems clearly has a huge impact on how many cars households choose to own. It would be a rare family in even the outer parts of larger Japanese or many European cities that would bother owning two cars. It’s a rare one in anything other than the inner parts of Australian and American ones that own less than two.
    And it’s not just the embedded energy in the cars – it’s all the roads/bridges/tunnels etc. too, which need constant maintenance (involving CO2-intensive materials) with such high usage.

  12. IkaInk

    @SBH – The embedded energy of a car certainly doesn’t come close to the energy required to run the average car for a lifetime. This article isn’t particularly authoritative, but as far as my limited research goes its accurate, and it summarises studies that have been performed by the Argonne National Laboratory, which is part of the US Department of Energy, in order to allow “researchers and analysts to evaluate various vehicle and fuel combinations on a full fuel-cycle/vehicle-cycle basis.”

    You can take a more in depth look at the Argone information directly, but the site is pretty clunky.

    All this aside, you’re dead right that embedded energy in vehicles and infrastructure should be considered when deciding which is cleaner and greener. However, its a really complex equation, especially when so many people will choose to own a car regardless of whether they catch the train to work, or even use public transport most of the time.

  13. Alan Davies

    Dylan Nicholson #1, SBH #7:

    No they don’t, hence the para above:

    The performance of public transport also improves when lifecycle emissions are taken into account.

    Tom the first and best #8:

    Average car occupancy is higher than JTW occupancy (circa 1.1 in Australia IIRC) but lower than non-work travel. Actually, I just checked the photo I linked to and it’s 69 people in 60 cars. I’d say that’s average JTW car occupancy (JTW is circa a fifth of all trips). Not in a position right now to check the figure for Australian cities, but the US discussions I’ve linked to assume an average occupancy, all purposes, of 1.6 (I’ve seen a similar figure cited for the UK).

  14. Tom the first and best

    Firstly, the posters depicting the space of buses versus cars is aimed at discussions about contested traffic, that is often at times of high demand for travel (including peak hour), when bus patronage is highest (at capacity is not uncommon in the situations talked about) but the average car occupancy is actually slightly lower than other times. It is therefore a reasonable depiction.

    Secondly trams have a more even patronage because they have greater usage during the day, on weekends and in the evenings. This is less so with trains but could be helped with things like building a station at Southland and extending the double track and electric wires to Leawarra for Monash University`s Peninsular Campus. Buses have such poor efficiency in Melbourne, mostly not because of low density but because of poor timetable coordination, low frequency and slow indirect routes. More bus lanes in congested areas would also help.

    Thirdly, patronage has mostly grown more strongly outside peak hour than in, since 2008 so the figures would be a little better.

  15. SBH

    Alan, do the calculations take into account the carbon cost of producing the vehicles? I have read that the carbon emissions of making a car outweigh the carbon produced by the car during it’s life.

    ‘The real fallacy is…’ and the next para – couldn’t agree more.

  16. Alan Davies

    Michelle Imison #4:

    You’re a bit selective in what you quote, Michelle. I offered three reasons for public investment in transit: environmental sustainability, mobility for those without a choice, and supporting dense agglomerations of activity.

    If your preference to live car-free confers a social benefit (e.g. greater sustainability), then in principle that’s a valid justification for public subsidy. However I’d be less convinced if there’s only a private benefit for you.

  17. Dylan Nicholson

    Actually Michelle my thought was more along the lines of “we invest billions of dollars of public funds in building roads, leaving little for other alternatives, in order to ensure most of us have no choice but to use them”…

    Of course as a cyclist I use those roads too, but if they were primarily meant for cyclists they wouldn’t need to be cost even a fraction of what they do now.

  18. Michelle Imison

    “As a society, we invest billions of dollars of public funds in transit in order to provide mobility for those who have no other choice.”

    Er – no. That might be *one* reason to invest billions of dollars in public transport, but some of us actually use it (and other modes, such as bicycles) because we prefer to be car free. Do I detect a tiny hint of pitying condescension there, Alan…? 🙂

  19. IkaInk

    The other factor with a number of ‘under-performing’ routes is that these routes help create the network effect which then allows other routes to perform well. You can’t look at individual routes in isolation when measuring the overall performance of the network. Each route has a bearing on how other routes perform both in positive and negative manners. Two parallel routes, or two routes that have a few key shared trip generators will likely result in a lower ridership per vehicle than if only one route was running. On the other hand a route that branches off to a trip generator from a trunk line will result in higher ridership on the trunk line, even if its own patronage isn’t very high.

    Additionally, if we were to start cutting services (as the Freakonomics article suggests), then we must also factor in the additional congestion that would be created. Travelling in congested traffic will not be as efficient as travelling in free-flowing traffic.

    @boscombe – The main running expense for a bus company is the driver, not fuel; so running smaller buses only really makes sense when it also means they can pay the drivers less. For the same reason that people continue to buy cars that serve a purpose they’ll only need on rare occasion, and not their most common purpose (e.g. a large station wagon to commute to work, because occasionally the owner likes to travel with the family to a holiday destination, etc), bus companies do the same with their fleet. It makes little sense for them to buy a number of small buses to run on a set route when later in the day the same route may carry a full large bus worth of passengers. To do so they’d need extra land to house the extra buses, they’d need to run extra empty loads to and from the depots and they’d likely need extra drivers as instead of a driver simply swapping at a terminus, one driver would need to be coming into the depot while another was leaving in the other bus.

    All that aside, I’m sure that at least some of Melbourne’s (and many other cities) could start switching some of the larger buses to smaller ones. There are some routes that would virtually never reach a full load, and even if these routes do become busier overtime, more than likely there will be new routes opening up in newer suburbs (or their should be) that will then have smaller loads. However I guess because there are many different bus operators it might make moving appropriate sized buses across the network difficult.

  20. boscombe

    Shocking things, those buses. Everytime I come across one I have to put the car window up and re-circulate air inside the car because of the pollution they belch out. I go to the beach early (sorry Dylan, have gone back to driving since bikes are just too troublesome) and see the first of the buses lumbering along with only one or two passengers.

    Why is there only one size of bus? Surely those 5.00am buses could be mini-buses or even just people movers?

  21. Dylan Nicholson

    Did any of those articles discuss the environment cost of providing the infrastructure and manufacturing the vehicles? Surely once that’s factored in, trains running on reasonably clean electricity sources should be doing a lot better than ~40% less emissions than having all the passengers in cars instead (inevitably sitting idle in traffic for much of the time and not running at particularly high fuel efficiency levels!).

    Still, it’s one of several reasons it’s worth trying to encourage people out of PT and on to bikes where possible 🙂

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details