Thursday, January 22, 2009

What is the carbon footprint of an Internet search?

What is the carbon cost of running a search on Google or any other search engine?

In 2007, Rolf Kersten of Sun Microsystems published that an Internet search generates seven grams of CO2. When the Sunday Times re-published this figure, Google published their own calculations of 0.2 grams of CO2 generated for each search.

The Numbers Guy at the Wall Street Journal believes the truth is out there, and there seems to be some consensus that the 0.2 gram figure is probably the more accurate one. I know some of the hardware designers at Google (Google designs their own servers these days) and they were hired partly for their expertise in designing high performance but low power "green" servers, so the 2007 figure is certainly outdated.

Searching via the Internet is certainly more efficient than driving to the library and using Inter Library Loan to have books and journals shipped around the country so we can do our research. But because Internet searches are so efficient and easy, we do so much more of them these days.

In the same vein, Internet publishing is as easy as the push of a virtual button on a computer screen. I tried my hand at paper publishing in my college days and shortly beyond, but there's certainly no way I'd generate as much content as I do now with a blog. How much carbon do I generate each time I push a post that is read by a few hundred people? The individual impact is tiny -- probably on the order of a gram of CO2 -- but I do it over and over and over and over again. I have somewhere around 3,000 posts on Cyclelicious now, and most of those posts link out to other websites, and each of those posts lives on to generate traffic and search engine results and are also stored at search engine cache farms.

This is the crux of Jevon's Paradox -- making resource use more efficient means we consume that much more quickly, because it's much more available. Jevon's Paradox also suggests efforts to improve fuel economy just doom the planet that much more quickly. The widespread availability of a 50 MPG Prius, for example, enables that many more people to live out in the boonies and buy more gasoline. The individual impact may be less, but the aggregate impact becomes multiplied.

A high technology society is a high energy society. What do you think? Can we afford to continue a high technology lifestyle?

6 comments:

  1. Thought-provoking post... I think there's a tendency to downplay the impact of these individual "small actions". Hmmm... I wonder what the carbon impact of this comment is :)

    ReplyDelete
  2. What's the CO2 footprint of just using (or leaving on) your own PC for an hour? Probably that's more significant than the footprint of your Google searches.

    ReplyDelete
  3. For the comparison to traditional print media this is absolutely true. Internet is leaner.


    But you forgot the most vital component. How do we get to use the internet? With power and electrity. Somewhere, huge server farms have to consume lots of it. As the internet usage across the world doubles or triples and large amounts of data capture and storage are needed every year, that consumption will only increase unless we switch to alternative sources of energy.

    ReplyDelete
  4. We are suffering from an over-abundance of needless technology. Give up all that crazy gadget love and live a leaner, high-tech lifestyle.

    ReplyDelete
  5. When I saw that 7 gram figure floating around a few days ago I really couldn't believe it. Still server farms take huge amounts of energy. At least it's easier to generate green electricity than fuel for cars.

    ReplyDelete
  6. We've got more energy than we know what to do with, we're just stuck on inefficient, harmful sources.

    As for Google, I would hazard a guess that the flip side of that coin is that the more use Google gets, the more efficient it is.

    ReplyDelete