High-Speed Rail and CO2


One of the less-publicized components of the stimulus package was an $8 billion commitment to develop a high-speed rail (HSR) network in America. This is no more than a down payment, given the very large sums needed to build HSR (University of Minnesota transportation scholar David Levinson estimates that the proposed California segment alone will cost $80 billion, or more than $2,000 per Californian; given my state’s financial problems, this is going to require a very large bake sale).

Since this policy appears to be a personal favorite of the president and other officials, the needed funding may well materialize down the road. But it is worth asking whether the touted benefits will too.

Backers cite many gains to be reaped from HSR, including relieving crowding at airports and on highways; cutting the need for expensive new air and road infrastructure; preventing road fatalities; reducing travel times and costs; promoting economic development (particularly in areas not well covered by air service); improving travel reliability; boosting productivity; spurring technological advances; stimulating the economy/creating jobs; and, because HSR will run on electricity and may require less energy to move each passenger, reducing greenhouse gas emissions.

This is a long list and the blog is a short medium. So for now let’s just consider the final point about HSR’s environmental benefits. Under some conditions, there is no doubt that an HSR system would reduce greenhouse emissions. Unfortunately, a study undertaken by the consulting firm Booz Allen Hamilton for the U.K. Department for Transport raises some troublesome questions about whether these conditions can be met in reality.

Booz Allen considered two potential U.K. HSR lines (London-Manchester and London-Edinburgh/Glasgow). They found that the CO2 emissions required to move HSR passenger seats were about the same as those required to move automobile seats — hardly a slam dunk for rail. In fact, intercity bus came out considerably cleaner than HSR on a per-seat-mile basis.

HSR would emit less on a per-seat mile basis than air travel. But the major caveat is that all of these figures consider emissions from operations only, without taking into account the very large amount of pollution that will be created in the construction of the HSR system.

When the emissions spewed by all those earth movers, tunnel boring machines, bulldozers, trucks, cranes, etc. are taken into account, the carbon advantage for HSR vis a vis air travel largely evaporates.

What would the bottom line be if the proposed U.K. lines were built? It all depends on how many people shift from air to rail; the more HSR passengers the better. But the authors found that even if the mode split on the proposed London/Manchester line shifted from 50-50 air/rail (approximately the current distribution) to 100 percent rail ridership, emissions over a 60-year period would be lower if the HSR line was never built.

The picture for HSR is somewhat brighter for a proposed London/Scotland line, because that corridor currently has a low 15 percent rail share (meaning there’s more scope for people to switch to rail). But to make the rail line worthwhile, HSR would have to capture almost two-thirds of the air/rail split. This is not impossible, but it may prove difficult given that air travel has a considerable speed advantage over HSR and will continue to do so in the future. (For reference, even the California HSR authority, which strongly supports the program, forecasts that rail will attract only one third of the air/rail split in the California corridor).

There are, of course, many uncertainties involved in this type of analysis. Technological advances may make electricity production cleaner in the future (though airplanes’ fuel economy may improve as well). Obviously Britain is not the U.S. (although given its high population density and short distances, Britain may actually be a better place for HSR than most areas of our country).

Still, the results of this study deserve careful consideration. No one argues that futuristic rail isn’t a great way to get from Main Street U.S.A. to Epcot Center (usually). But given the very severe budget constraints we are currently facing, a program as costly as HSR should be evaluated very thoroughly despite its considerable allure.

Hat tip: David Bayliss


I applaud the attempt to examine a "green" policy and see whether or not it will actually reduce green house gas emmissions, and I largely agree with the points in this post. But I do have a (large) quibble: when evaluating the impact on emissions of HSR, you obviously need to look at the changes in how people utilize mass transit. I do not know anyone who currently uses rail lines for interestate travel outside the eastern corridor (between Boston and DC) with any regularity. If a major component of emissions impact is:
" how many people shift from air to rail; the more HSR passengers the better"
then wouldn't the US market represent a major opportunity because so few people currently use HSR? I imagine we have lots of marginal air plane passengers who would be willing to try HSR because they dislike flying or find it too expensive.
Full disclosure: I am not an environmentalist.


Yes, it is costly in environmental terms to build HSR lines. But what about the cost of building roads (or even just maintaining them)? What about building airports?


A distinction needs to be made between direct and indirect emissions.

If HSR is 100% electric, then it doesn't have to emit anything. Cars and planes do. We can't make any investment in sustainable technology without CO2 emissions in the short term.

100% electric technologies do not emit anything directly. They run just as well on nuclear, wind, solar, etc. as they do on coal. Once we replace the power plants, we will have a 0 emissions infrastructure in place without any additional investment or emissions.


What if you include the carbon it took to build all the freeways between SF and LA?


The start up costs for almost all forms of transit are high. This same argument could have been used for the Erie Canal and the Interstate Highway System.


It seems this study doesn't take into effect the reduction in carbon emission attributed to lifestyle changes surrounding HSR. When a society moves toward a rail-based transportation infrastructure from a primarily road-based one, people tend to settle in denser, walkable neighborhoods with smaller home sizes. This reduces the amount of car use, overall, while smaller living spaces reduces the energy used (and, hence, the carbon footprint) to heat and cool those houses. A road-based transportation infrastructure, on the other hand, promotes sprawling, low density, segregated land uses and large houses that use more energy per capita.

Finally, it should be noted that the California HSR was never about reducing carbon emissions, but primarily about reducing long term costs. It was estimated that building all the necessary road and runway space to accommodate the same amount of passenger growth would have far exceeded the costs of building the HSR system.




I think your article over looks the fact that diversity of transportation, especially adding HSR, reduces wasteful energy use which is where most of the C02 pollution is taking place. UK studies really don't apply here because in general the UK is not nearly as wasteful with their transportation as we are. They don't have a I405. It doesn't take them 2 hrs to get to work if there commute is over 30 miles.

Your analysis, thoughtful to your title did forget two central aspects.
1 .Growth rate of general population and travelers. Though building HSR will have some negative environmental impact , it will still reduce the overall environmental reduce impact as our population grows. It is much cheaper, efficient and less environmental impact to control C02 at central locations which is what electricity offers. Even building the HSR environmental impact can be negated with current technology and practices. The real question is what will the C02 impact be if we don't build one?

2. Basic theory of diversity. Your analysis missed this key aspect. By spreading the transportation love where movers take more efficient forms of transportation for their needs will have less impact with C02 emissions. The extremely positive aspect of HSR and C02 is the lack of wastefulness it has. Did the analysis factor in the C02 wasted every time a plane sits on the tarmac due to over capacity? Every time a car is stuck in traffic due to over capacity? Adding HSR will open capacity and reduce wasteful energy consumption.

The work /C02 ratio out of construction equipment isn't nearly as wasteful as it would be on other forms of transportation.



The whole "green" movement is so impossibly convoluted and complicated that it cannot possibly be implemented with any assurances that it will achieve the desired outcome. It's an example of an affluent society trying to salve its conscience and in the process making matters worse, or at least no better.

@2 Robert--A few questions: Do you expect that just because we build HSR we will not maintain the roads? By building an HSR project, wouldn't we be simply adding on to existing "greenhouse gas" emissions? What about building multiple train stations that can accommodate the HSR, parking, offices, maintenance facilities, etc.?

Outside of the largest Northeastern metro areas, ridership spikes when the cost of operating a car spikes. Otherwise, they mostly operate at a loss or break even. (At least that's the way it was when I lived in Atlanta.) So then you have to add the societal cost of that dislocation to the cost of HSR.

Like I said, the green movement is strictly a project of importance to affluent societies. Less affluent societies aren't interested, as India just made crystal clear to Sec. of State Clinton.


d trembley

Hmm. The big point is that the cost to move any goods (including people) is going to go up. Oil, diesel: they are only going to get much more expensive, maybe sooner than we project or wish.

Not building HSR might be penny wise but pound foolish as the only place we have to live is the future.


While David MacKay does not incorporate building costs (to my knowledge), he calculates HSR as 27x more efficient than cars if the train is full. It would take a very empty train to have the same (lack of) efficiency as a car. I'm not sure where his estimates grossly diverge from that of Booz Allen's.

Tom M

For shame. This is the kind of calculation that involves very intricate and complex assumptions about any number of data points. Consequently, the results of these calculations vary substantially from one another, and it is very difficult to get an idea of what the real differences might or not be if the assumptions are not provided with the results. Doing otherwise is at best setting up a straw man and at worst a piece of deceptive propaganda. In this environment, a single study is relatively meaningless; what is needed is a broad array of studies that can be compared with one another. This is particularly dangerous in light of what appears to be a concerted effort by industry to discredit the train (and "global warming" issues in general). A preliminary question: who paid for these studies? Examples of such complications include how much of the infrastructure is taken into account (e.g. is airplane/car/train manufacture considered?) and what that the infrastructure looks like (e.g. what about a general move away from coal?). A particularly essential assumption that remains undiscussed is the time frame involved, because the expense is much, much higher on the front end for building a high speed network than it is to run it. There are a number of fundamental reasons to think rail traffic would be fundamentally more efficient. To begin, most people seem pretty confident electric plugins would be more efficient than cars (this is mostly to do with internal combustion issues). Yet trains should for any number of reasons be more efficient than cars. First, with trains electricity does not have to converted from AC to DC and back as often. Second, trains have a much lower rolling resistance than cars. Third, weighted by carrying capacity trains have a *much* lower and sleeker profile and should have less air resistance despite traveling at twice the speed (i.e. the profile difference should be greater than a factor of four). Fourth, by carrying capacity, train lines should cost about what an equivalent road would cost. I fully admit that the most important variable is how much people will actually use a given rail line, but this would work itself out if the externalities caused by pollution were truly figured into the price. If the difference is great enough the incentive will be for business to promote rail travel and make it more effectively used.



@3 Noah--Where to start?

"If HSR is 100% electric, then it doesn't have to emit anything. Cars and planes do."

Whaaaa? HSR emissions occur upstream, beginning with the excavation/extraction of the fuel needed to power the electric plant.

"We can't make any investment in sustainable technology without CO2 emissions in the short term."

CO2 emissions will be with us short and long term. No getting around it.

"Once we replace the power plants, we will have a 0 emissions infrastructure in place without any additional investment or emissions."

Look, we're all supposed to be underwater in 100 years. That electrical infrastructure will not be replaced before we are all living in rafts.

What's the prediction? Sea levels will rise a foot over the next century? I will make a bold prediction. 100 years from now, not one person will have gotten wet because of that rise in sea levels. Further, any dwelling/person that is currently within 10 yards of a coast line will....move inland about 20 yards. Problem solved for another 100 years.

I absolutely don't understand the desire to completely overturn our entire economy/lifestyle for what has, according to many studies, little chance of success.

The day that our current methods of energy production become too costly, they will be replaced. Not before.



I'll second some of the comments here.
I don't believe that the comparrison is very fair. You are assuming that no new infrastructure would be required to keep up with future growth. California's air and car corridors are basically at capcity so any passenger growth in the long term requires the same sort of infrastructure work and emissions as you are criticizing.

I'd like to see what happens to air travel if you add up the huge new airports (with their associated car trip to the park & ride lot) or worse, adding lane miles to highways from LA to SF.

My guess:
If you add the Seat Mile Emissions of any mode of transportation (total of per trip emissions and amortized construction emissions from the creation of that mode of transit) Rail would end up with a significant advantage in emissions once the expanded capacity is taken into account.

Lastly, the fact that Electricity sources are fungible once the power is generated means that the Rail Network is the only system that could get better emissions over time without replacing the vehicles (airliners can loast 40 years!).



A high speed rail network would be faster than or the same speed as a flight from Edinburgh to London.

Currently, with a regular train, the journey is 4 hours 45 minutes. A flight takes around one hour. However, both Edinburgh and London train stations are very central. This means you can arrive for a train just five minutes before it leaves, you need no connecting transport, there is no time consuming security and you do not have to wait for your bags at the other end.

Even with a low speed train, the time is about the same. A high speed train would be much faster than air travel.

Michael Peters

> given that air travel has a considerable speed advantage over
> HSR and will continue to do so in the future

Does this "speed advantage" take into consideration all the time wasted at the airport? When I lived in DC I could get to NY considerably faster on the train than I could by plane because getting to the airport, checking in, going through security, etc added so much more time to the trip that it wasn't worth it.

Because of the space needs, airports are rarely close to the city. In contrast, train stations are much smaller and are usually very close to where people live.

Avi Rappoport

Airplanes are dependent on high-quality liquid fuel. If for any reason the supply was disrupted or the cost went way up, the cost comparison would tilt towards rail.

The amount spent on repairing and upgrading freeways is enormous, and also currently dependent on fossil fuels.

There are 40,000 deaths from auto accidents every year. Extrapolating from that, it's boggling how many injuries to people and damage to cars there must be. While one big train crash could be spectacularly awful, I doubt the stats would work out worse than that.

If building High Speed Rail also means re-building a working rail system for local use and goods transport, I am so for it.

steve from virginia

Here, the perfect is the enemy of the good. We don't need high speed rail, we need reliable regular speed rail, we need a lot more of it, and we need it yesterday.

We have an economic crisis, a fuel crisis and a climate crisis. We don't have the means or the time to support a luxury train system.

Passenger and 'hot-shot' freight trains routinely traveled over 100mph ... in the 1920's. Clickety- clack, indeed. This was with steam locomotives. There is no reason why acceptable service cannot be provided with current electric locomotives and rail technology. Giving Americans the rail service of pre- WWII would be a giant leap forward (backward).

The transport problem has two parts; one is purpose and the other scale. HSR is a gee- whiz toy that has no purpose other than to advertise itself (and enrich constructors). As auto-mania recedes into the far distance, large numbers of auto trips will require replacing. Moving large numbers is what matters, the fillip of added speed does not add equivalent value.

Moving the large numbers requires capacity more than performance, Any solution requires large scale. $8 billion would improve current standard passenger rail service. $80 billion would go a long way to electrifying the entire national rail network.

Here is a good article:

Standard, electrified rail is the way to go, as results can be felt immediately. Leave HSR for another time.


Johnny E

The speed advantage of airlines evaporates when you factor in getting to and from the airports, standing in line at security, waiting at boarding gates, checking luggage, ....

High-speed trains are roomier, you can walk around whenever you want, the seats are more comfortable, the bathrooms are bigger, the scenery is better, they're less affected by bad weather, if there's a delay in one city the whole network probably won't go down.

Airports are at full capacity with little room for expansion. Jet exhaust stinks and planes sit on the taxiways forever waiting to take off or circle for hours waiting to land. If an electric train has to wait it's not burning fuel.

Check out the NOVA documentary about Global Dimming and the role of contrails in climate.

John Jefkins

5,000km of new high speed line have opened across Europe recently and vast numbers of people have switched from air travel to rail instead.

For Eurostar between London and Paris for example rail emits 10% of the equivalent air journey and 200mph rail has now taken 80% of the market.

That 80% figure has been gained on every route where the journey time has been reduced to 2 or under 3hrs.
There is an S curve where one can predict market share from journey time. Even 5 hr journeys at 200mph are winning 40% of the old air market.

A new line will thus not be needed all the way to Scotland for the reduction in journey time to take most of the air market. The existing line allows 125mph - so a new line for just half the distance (with a speed of 200mph) will suffice. That means we just need to build a line just past Manchester to win most of the Scotttish air market.

200mph rail thus has a proven track record of winning air markets and reducing CO2 emissions to 10% of what they were before for the equivalent journey.

It does matter that (a) the trains are electric and (b) how you generate the electricity though! The more wind power, wave power and nuclear in the mix and the less coal you use, the more CO2 is saved,

John Jefkins - London - Great Britain



Okay, good job doing the first step of deeper thinking, now on to the second:

In California, electricity generation is about 50% low-carbon natural gas and 50% carbon-free sources (hydro, nukes, and other renewables). So operating CO2 is going to be dramatically lower than in England, which relies mostly on coal.

If an HSR powered by coal is a wash for CO2 when compared to air, it seems highly likely that a low-carbon electricity version would provide a pretty substantial environmental benefit.