Roughly one fifth of the global carbon dioxide emissions seem to be from transport. According to Wikipedia, which uses data from EDGAR, the figure is 14%. WWF claims that transport accounts for 26% of the emissions, and the U.S. Environmental Protection Agency (EPA) has come up with a 27% figure for the U.S.. European Environmental Agency arrives to a 20% share of total CO2 emitted in Europe. Rather different figures, but in any case transport, both cargo and passenger travel, is one of the major factors in global warming. The biggest culprit seems to be production of energy, for both electricity and heating.

The share of airplane emissions of all transport is still low, below 10% according to most sources, but rapidly growing. For the energy efficiency per kilometer, air travel doesn’t do too well. The previously referenced EPA report states that in 2001, energy required to move a ton-mile of air cargo was 7.5 times greater than heavy-duty trucks, over 17 times that of ships, and 83 times greater than rail. For person traffic, they estimate about 4.5 liters of fuel spent per 100 km and per seat. Aviation Environment Federation (AEF), a UK non-profit, has a very readable comparison between car, rail and air travel, which lacks credibility by not listing references, but uses a very similar fuel consumption figure (4.8 liters of fuel per 100 km per seat). That calculates to about 150 grams of carbon dioxide per passenger kilometer — the same than a fairly fuel-efficient modern car with only the driver inside.

Various sources (see e.g. Aviation and the environment in Wikipedia) estimate that the total global warming effect of air travel is about two times compared to the carbon dioxide alone, due to other gases emitted and the high altitude. They are usually referring to The Intergovernmental Panel on Climate Change (IPCC), which has published a report on Aviation and the Global Atmosphere. I found that report rather hard to read and interpret.

Altogether, flying seems to be approximately as harmful or somewhat more harmful per kilometer than driving a car alone. A car full of people or traveling by train is clearly more efficient. It was more difficult to find data on sea travel, but the efficiency figures of the EPA report would place ships somewhere between planes and trains. However, due to the larger amount of cargo transported by sea, the total impact of shipping seems to be even larger than airlines. For personal travel, worst of all are probably high speed motor boats, which can consume really excessive amounts of fuel per passenger. Fortunately such boats are rarely used for traveling very long distances.

Overall, flying to a far-away destination is clearly not eco-travel. Even if the impact was only the same than a fuel-efficient car (the minimum figure), a trans-atlantic round trip flight would be equivalent to 15000-20000 km of driving, a distance which a typical car owner covers in one year.

Personally, I’m trying to proportionate the length of my trips to the time spent at the destination. I’m not going to fly to Hong Kong just to spend a weekend there, even if I’d find a ridiculously low price for the ticket. However, I could still take the same flight to start a two month excursion in China. Of course the environmental impact of the flight is exactly the same, but this way I end up taking less of such flights per year. Having enough time I’ll also enjoy the trip much more, making it better worth the high cost.