According to Stefan Krause of the UK's University of Lancaster, in 2006 around 7,730 delegates travelled a total of 15 million km to reach Vienna from 91 countries. That travel produced roughly 5000 tonnes of carbon emissions. Krause found that just 18% of the delegates were responsible for around 3000 tonnes of carbon, as they came by plane from another continent. European delegates generated between 800 and 1800 tonnes of carbon, depending on whether they arrived by train or air. That said, there's quite a lot of uncertainty in the data, as carbon emissions depend on the type of plane, train or car as well as the passenger loading.

Krause's colleague Wouter Buytaert, who argued that the meeting's carbon footprint was bigger than necessary, told environmentalresearchweb that the average carbon footprint of a scientist is probably higher than that of "normal" people, as a result of travel to conferences. "We have a higher footprint and tell others to lower their footprint," he said. "We have a moral responsibility to set an example." Both Krause and Buytaert took a 25-hour train trip to reach Vienna.

Martin Juckes of the British Atmospheric Data Centre agreed. "The public will hear first the words 'flown to another conference'," he said, "and miss out on any further message about the environment."

So what are the alternatives? Conferences could make more use of online networking technologies such as Skype or AccessGrid to bring scientists together, while researchers could also network using web resources such as MySpace, choosing who to share their ideas with. As participants suggested at the debate, the EGU could provide low-carbon travel grants to help people arrive by train rather than plane when the low-carbon option is more expensive. This year, two delegates cycled to Vienna from Amsterdam - one member of the audience proposed the pair should get free registration. The EGU could also use its power to get a bulk discount on train transport, in the same way that air companies provide reductions for major conferences.

Krause, Buytaert and colleague Tobias Krueger looked only at the carbon footprint generated by scientists travelling to the meeting - it could be interesting to look at the footprint of the conference as a whole, including the air conditioning and lighting use of the conference centre. The EGU reduced the size of the general assembly abstract book by more than half this year to save resources (and cut carbon emissions).

Arguing that the EGU's carbon footprint was not bigger than necessary, former EGU president John Ludden said that the conference's 80 splinter groups and business meetings saved carbon emissions by removing the need for lots of smaller meetings. "We should be bringing more people here to make ourselves more effective," he added. "We should be talking with the energy industry and economists." Ludden also suggested spending the money that might otherwise be spent offsetting the meeting's carbon footprint to bring, say, 100 Chinese or Indian scientists to the conference in order to boost environmental knowledge in countries that are developing fast.

It looks like it's a debate that will continue. Next Krause and Buytaert would like to see the EGU conduct a survey of conference participants to find out exactly how they travelled to the meeting. And, while the conference will take place in Vienna again in 2008, the EGU is currently investigating options for the location of future annual meetings - ease of transport to the site could well be a factor.