EVEN HARVARD CATCHES OBN TO WHAT WE'RE DOING

FROM THE GUARDIAN
The carbon cost of Googling
Climate researchers say two Google searches emit 7g of CO2 – the same as boiling an electric kettle. Do their numbers add up?
Comments (…)
Google

Net giant Google is central to our lives – but is it energy efficient? Photograph: AFP

Can two Google searches really produce as much carbon dioxide as boiling enough water in an electric kettle for a cup of tea? That's what Alex Wissner-Gross, an environmental fellow at Harvard University, is claiming. "Google operates huge data centres around the world that consume a great deal of power," says Wissner-Gross in forthcoming research about the environmental impact of computing, which calculates that every Google search produces 7g of CO2. "Google are very efficient, but their primary concern is to make searches fast and that means they have a lot of extra capacity that burns energy."

It should probably be noted at this point that Wissner-Gross is also the co-founder of Enernetics, and its associated website www.CO2stats.com, which, according to the Boston Business Journal, allows "websites to get analysis of how energy-efficient they are and sells carbon offsets to help them reach a neutral status". So let's first congratulate Wissner-Gross on getting himself and his company talked about all over the internet, including here. But does his claim stack up?

Without any published data to hand it's hard to tell. All Google is saying is that it is takes the issue seriously, but that "the energy used per Google search is minimal". It adds: ""In the time it takes to do a Google search, your own personal computer will use more energy than we will use to answer your query." (If this is true, it surely makes a mockery of Wissner-Gross's claims as there's no way an average computer uses as much power as an electric kettle when it's boiling water.)

So let's do some crude sums based on what we know and what is being claimed. Google receives millions of search queries every day from all over the world. Estimates vary about quite how many queries it receives, but they seem to range from 200m up to 500m. Let's, for the sake of argument, take the top figure as a worst-case scenario.

If Wissner-Gross is correct then 3,500 tonnes of CO2 (500m x 0.000007 tonnes) are emitted every day through all of us performing Google searches. Or put another way, 1.28m tonnes a year. That's about the same as Laos emits each year, the 151st biggest emitting country in the world.

I'm torn between thinking that this sounds like an awful lot – "Shock: Google emits as much as a country!" – or whether it doesn't sound too bad, given, for right or wrong, how integral Google now is to many of our lives. What is certain is that the environmental impact of information technology as a whole is considerable and ever rising.

A widely quoted figure is that the global ICT sector produces as much CO2 each year as the global aviation industry – about 2-3% of total global emissions. It is helpful, therefore, that Wissner-Gross's claim is at least providing a needed spur to debating the ICT sector's impact, and how best to reduce it.

Ultimately, though, I suspect this particularly quotable nugget will have little impact on the searching habits of internet users. Nor should it, really. We can each monitor how much electricity our own computers use – and aim to keep it at a minimum – but it can only ever be Google's responsibility about how much power its servers and related hardware use. Perhaps there's even an argument for saying that internet searches have helped to reduce net emissions by greatly reducing the need to make physical journeys in search of information, say, a trip to the local library or bookshop?

(NB: At least one cup of tea was consumed during the making of this blog.)