Chris Anderson today argues that “dead-tree” magazines have a smaller net carbon footprint than web media.

I’ve looked at this question a number of times this year, and I think Anderson makes an extremely important point. The very term “dead tree edition”, long used jokingly by new media types, implies a assumption that digital media is inherently more environmentally-friendly than print. That assumption is deeply flawed and needs to be challenged.

Anderson’s post is a great Gedankenexperiment, but he could have also drawn on a growing body of evidence on this subject, including supply chain audits by a number of newspaper publishers and one excellent comparative study between print and online carbon emissions published last month by the Royal Institute of Technology in Sweden. (That study’s footnotes point to even bigger pile of evidence.)

It’s easy to quibble with Anderson excluding the impact of his magazine’s distribution network as “carbon neutral” — it isn’t, even if the US Postal Service “runs the same routes whether they’re carrying our magazines or not”. There is some marginal cost to delivering thousands of magazines — paper is heavy and increases the size of lorries needed by the postal service.

But the key environmental issue in the print supply chain is not distribution, but the energy used in paper production, which Anderson also suggests (very wrongly) to be “carbon neutral”.

A study conducted by the Carbon Trust and newspaper group Trinity Mirror in late 2006 found that each copy of the Mirror produces 174 grams of CO2.

The study found that that about 70 per cent carbon that goes into the supply chain of printed publications arises in the paper-production process, and that the key source of emissions at that stage was the source of electricity used by the paper plants. It concluded that sourcing more newsprint from Scandanavian countries with cleaner electricity generation would cut back a British newspaper’s carbon footprint.

Similar results were found in audits conducted by the Guardian and the Gazette of Montreal this year. In mid-2006, the Heinz Center published a report that studied impact of producing the paper used by Time magazine. It found 61 per cent of Time magazine’s emission’s came from paper production.

For online media, meanwhile, the most important factor is the (actually quite enormous) energy consumption of 24/7 web servers and the air-conditioning they require. In addition, there is the the energy used by users’ PCs and some tiny fraction of the entire process of manufacturing and later disposing of all of these electronics.

Unlike printed products, which result in the same one-of bundle of emissions no matter how many people read them and for how long, online media’s environmental impact depends on how long consumers use their PCs.

The recent life-cycle analysis from Sweden showed just how big the impact of all this is. Reading the web version of a newspaper produced less carbon emissions that its printed equivalent — but only if you read it online for less than 20 minutes per day. Moving to an e-reader device with a passive e-ink display increases the length of time before digital media has the same impact as print by another 10 minutes.

Once again, the key variable was the source of energy used by the digital devices. Move to a country with lots of sustainable energy sources, and you can use the web longer before reaching the same level of emissions as buying a newspaper.

The Swedish study says it did not include “Internet infrastructure” in its analysis. This presumably means all of the routers and servers between the publisher’s datacentre and the end users’ machines. If I understood this correctly, this will cut back the 20 minute mark even further.

All the evidence I’m aware of concerns the production of newsprint products at daily newspapers. As a monthly magazine, so the volume of paper Wired uses is much smaller. On the other hand, the Trinity Mirror study noted in passing that producing glossy magazine paper requires more energy than newsprint.

The real question seems to be “how long is online greener?” How long spend do you spend consuming the Wired brand every month? If it’s more than around 30 or 40 minutes, buying the magazine is probably the greener choice. Otherwise, stick to the web site.