The answer depends on whom you ask

How much global warming results from air travel?

Posted by Clark Williams-Derry (Guest Contributor) at 10:49 AM on 18 Dec 2007

Read more about: travel | air travel | climate | energy | greenhouse-gas emissions

Over the past few days, I've been trying to pull together some data on how airplane travel affects global warming, as part of a broader project on transportation and climate change.

My stunningly obvious conclusion: it's complicated. Worse, different calculation methods yield wildly different results.

Take, for instance, this brilliant chart (below) from the Stockholm Environment Institute, comparing many of the major online emissions calculators. Emissions are represented by the light blue lines. As you can see, the online calculators find that a Boston-D.C. round trip has the impact of somewhere between 0.19 tons and 0.48 tons of CO2 emissions, depending on which calculator you use. (By the way -- as I discuss below, it looks like the Atmosfair calculator is probably the most accurate and comprehensive.)

airplane emissions calculators - 390

The high-end estimate is over two and a half times higher than the low-end estimate. What's up with that?

To be brief, the complications in calculating air travel emissions include:

Distance of flight. Generally speaking, short flights use more fuel per mile traveled. It takes a lot of fuel to lift a plane, and shorter flights have more lift per mile traveled. Taxiing also burns through fuel, and the amount of taxiing doesn't vary that much by the length of the flight. So you can think of short flights as "city" driving for a car -- with lots of idling and acceleration -- and long flights as "highway" driving. That said, fuel efficiency drops for very long flights, since the plane has to carry extra fuel for a long flight -- and it takes fuel to carry fuel.
Altitude. Although long flights tend to use less fuel per mile, they also tend to fly higher, and stay at high altitudes for longer. And high-altitude flying has some troubling side effects. On net, airplane contrails, and the cirrus clouds they form at high altitudes, are believed to trap heat in the atmosphere, especially at night.
Timing and location. Flying at night, and at northern latitudes, may increase the formation of heat-trapping clouds. During the day, though, clouds may reflect sunlight back into outer space.

The problem is, there's tons of uncertainty about all of these effects, particularly for high-altitude emissions. (See this IPCC report, especially this page, for more on this.) But on average, the IPCC recommends multiplying the CO2 emissions from fossil fuels by 2.7 to get the total global-warming impacts from a given flight. That is, the comprehensive global warming impacts (counting contrails, clouds, and trace gases) of flying are about 2.7 times as high as the CO2 alone. Note, however, that according to this pdf, a more recent study pegs that figure at 1.9, rather than 2.7.

Confused? I certainly am.

As should be obvious, these complications can make it really difficult to estimate the climate impacts of any given flight. And to add to the confusion, if you really want to be accurate you'll need to account for the type of plane you're flying, how old it is, whether you're flying first-class or economy (first class seats take more room on the plane), and how full the plane is. Sheesh! It makes calculating the climate impacts of your car a piece of cake.

There are two ways around these complications. First, you can make some simplifying assumptions -- some rules of thumb that work pretty well, all things considered. And second, you can rely on the best, most reliable online calculator you can find.

For the second tack, the Stockholm Environment Institute recommends Atmosfair, since it has the broadest and most comprehensive assessment of the climate impacts of flying -- covering not just CO2 emissions, but also other gases, contrails, and the like. As a consequence, Atmosfair tends to give a higher estimate of climate impacts than the other flight calculators -- depressing, but probably pretty accurate.

For the first tack -- well, any rule of thumb is going to be overly simplistic. But for my purposes, it may be best to stick with the CO2 releases from airplane fuel only, divided by short, medium, and long-distance flights. The World Resources Institute (PDF) issued some widely used estimates, based on a 2005 report by a British government agency (PDF), that have been used at this reputable GHG protocol site. However, The British agency updated its figures in a 2007 report, here (PDF), allegedly based on this EU report (PDF) -- though I can't figure out how. So I'm going to use the most recent British figures (see table 6 in the pdf) as the best available estimate of CO2 releases per passenger-mile. Converting from metric, they are:

0.56 pounds per mile for short flights,
0.46 pounds per mile for medium flights, and
0.37 pounds per mile for long flights.

Of course, that doesn't match up at all with Atmosfair's results. But it'll have to do for now. I'm done with complexity.

For story: The answer depends on whom you ask
8 Comments | Post a Comment

Sources of error

Atmosfair does have a 23 page PDF laying out their methodology.

It says, in part:

"A compromise between accuracy and data volume was struck when designing
the Emissions Calculator. The most important factors are simulated, if at all
possible, without giving an exaggerated impression of accuracy."

It lists the following as major uncertainties:

Aircraft type
Seating
Seat occupancy rate
Engine type
Condition of aircraft and engines
Specific fuel consumption as a function of flight distance
Detours
Holding patterns
Weather
Current state of atmosphere
Scientific knowledge level to IPCC 1999

a sibilant intake of breath

by sindark at 11:30 AM on 18 Dec 2007

Planetary Boundary Layer (PBL)

I agree that it's complicated - I've worked on lower level atmospheric emissions from aircraft for years. These emissions are usually only counted within the mixing layer that extends up to the PBL, which can be 3,000 to 5,000 feet during the day. Above the tropopause, emissions do not generally mix down (except in severe storms).

So do high-altitude jet plane emissions even count? My understanding is that many parts of the stratosphere are actually cooling, not warming up.

Then as a minor point, you say that taxi operations (ground aviation movements) are fairly dependable and short. Tell that to folks like me who have sat for hours on the tarmac, only to find out that the plane is out of fuel and has to take on a few more thousand pounds!

Onward through the fog

by Sam Wells at 11:33 AM on 18 Dec 2007

taxi

you say that taxi operations (ground aviation movements) are fairly dependable and short.

I think his point was that taxi operations don't vary in any predictable fashion, and aren't related to the length of the flight. They are, however, potentially a big wild card on any given flight.

by GreenEngineer at 12:30 PM on 18 Dec 2007

And this doesn't begin to cover...

...flights from cargo planes, private jets, military planes, and the like.

And if we really wanna get technical 'bout it, how much do the rocket and satellite launches add?

by Tasermons Partner at 4:04 PM on 18 Dec 2007

cargo included

To Green Engineer, I realize that taxi time averages are hard to get but for large international airports, but the data is there. I use FAA raw data, anyway.

To Taser, general aviation (small planes), regional taxi panes, and commercial cargo planes are usually included in the airport activity statistics in separate columns. Helicopter activity is typically not. And because of Homeland Security, good luck on getting any military data, anywhere. It is a bunch. Forget rockets.

To all, I am still having a problem with a distance-based function. I think that is bogus, since cruising it usually at 20,000 to 30,000 feet. It's all about take-offs and climb-out, descent approach and landing. If we were talking international ship emissions, I would say just the opposite. //sam

Onward through the fog

by Sam Wells at 5:47 PM on 18 Dec 2007

CO2 Emissions are Dependent on Weight!

News flash: planes use more fuel when they carry more weight. Why is no one talking about this? Nor is anyone talking about the fact that as Americans get fatter and fatter, we use more gasoline when we drive even if nothing else changes. Fat people also consume more food, which uses more energy to produce and transport.

I weigh only 110 lbs and travel very light -- lightweight wheeled 22" suitcase with not much in it. I often site next to HUGE guys whose jeans and shoes alone weigh as much as all the things in my suitcase. I also keep my car virtually empty of stuff except for emergency equipment. My contribution to CO2 emissions is a lot lower than those of people who let their BMI get up over 20, not to mention 25, 30, etc.

Losing weight is something most Americans could do to help save the planet.

by rivergal at 2:46 PM on 19 Dec 2007
[ Parent ]

A Few Comments

Kudos to Clark Williams-Derry for the helpful post examining what's behind carbon offsetters' air travel calculators. It's great to see this information gathered in one place, even if the only definite conclusion that can be drawn is it's complicated. :)

I noticed the chart from the Stockholm Environment Institute says, "Calculations were done on 7/3/06 if not indicated otherwise." A year and a half is a long time, especially in a quickly-moving industry like carbon offsets. Carbonfund.org has been using an updated calculator for some time. It includes, among other improvements, a multiplier for radiative forcing and the ability to choose departure and arrival locations.

You can read about the methodology for the air travel calculator and all the Carbonfund.org calculators here: http://www.carbonfund.org/site/pages/carbon_calculators/c ....

Thanks for this great post and keep up the good work. Transparency is the name of the game for carbon offsetters, and we applaud any effort to increase it. To that end, we'll be featuring a posting on the Carbonfund.org blog in the near future detailing the decisions behind our assumptions for air travel.

William from Carbonfund.org

by William Bert at 11:54 AM on 20 Dec 2007

Advice I once received from a farmer...

Many years ago, I met a teacher while on a flight to New Orleans. He complained to me that the rural school he work at provided his class with computers, but refused to purchase mice for those machines. As there were only a few computers, he purchased them with his own money so his kids could get more use out of the machines.

The teacher also farmed to supplement his income, and he drew an analogy between the computer mice and the price of nails on the farm. He said that because the cost of grain represented 80% of his total costs, he watched the price of grain very closely. But if he needed a box of nails to repair fencing, he just bought the nails regardless of what they cost.

The IPCC report on Aviation cited above suggests that CO2 emissions from aircraft represented 2% of total anthropogenic CO2 emissions in 1992. I don't know whether that number has increased in the last 16 years, and I have heard that high-altitude emissions are worse than ground-level emissions of the same magnitude. That said, however, I am largely of the same mind as the farmer: if coal burning represents more than 40% of global emissions, liquid petroleum another 40%, and tropical deforestation somewhere in excess of 10%, I don't see air travel as a particularly significant concern. By the farmers reasoning, air travel is looking a great deal like the box of nails.

Assuming absolutely no improvements in flight efficiency, air travel could increase by 50% and it would still represent less than 3% of total emissions.

If the high-altitude emission issue is far more significant than I am accounting for in my reasoning, or if there is some other reason why air travel is represents far more than 2-3% of the problem, I am open to change my position.

by picassotrigger at 4:09 PM on 15 Apr 2008

ADVERTISING POLICY

You are not logged in. Thus, you cannot post a comment. If you have an account, log in. If you don't have an account, well, by all means go make one! Meet you back here in five.