The Cost of Tweets (CO2 Emissions)

What’s the cost of tweets sent daily on Twitter in terms of CO2 emission?

According to data from Twitter, published in February 2010, every day around 50 million tweets are sent over the popular micro blogging platform.

Considering an average time of 10 seconds to write one single tweet and the average consumption of 250 watt/hour for one single computer, the total daily consumption amount to 35 kW h (kilowatt hour).

Multiplying this value for the average estimate of CO2 emission per kW, we obtain the daily value of CO2 emission which rises up to 492,000 Kg. Obviously this value is only an rough estimate (rounded down) but, just to give you an idea, it’s like driving a car producing 150 grams of CO2/Km for 3,3 million kilometers.

MORE LINKS

  • Reply
    damian
    August 27, 2010 at 1:33 pm

    And thats just for the twitter users . If you add to the total emissions the twitter servers it will probably the double(?) Anyone got a calculation?

  • Reply
    Paperjam
    August 27, 2010 at 1:50 pm

    This number is way off. Most tweets are made from mobile devices, which use considerably less power than a computer.

    • Antonio Lupetti
      August 27, 2010 at 2:02 pm

      It’s true. But I didn’t consider the time to read each tweet which is considerabily more important that the kind of devices from which tweets are sent.

    • Flo
      August 27, 2010 at 4:50 pm

      Retweets and tweets sent by clicking “tweet this” buttons or automatically (via facebook or by blog software etc.) take no time at all. Lots of users use their mobile devices to tweet AND to read tweets. In the remaining cases the computers are usually turned on anyway and the interactions with twitter take place in the working pauses. So I would estimate the actual additional CO2 caused by Twitter is way lower.

  • Reply
    Milos Sutanovac
    August 27, 2010 at 2:31 pm

    Loving the details & colors, Antonio.

    @Paperjam:

    Off or not, isn’t this shocking enough?

    • Antonio Lupetti
      August 29, 2010 at 3:05 pm

      Thanks Milos! :)

  • Reply
    Don
    August 27, 2010 at 3:44 pm

    That calculation seems to be assuming that those resources wouldn’t be used for some other computing process such as browsing the web, email, etc. So the calculation based on time to write tweets is only meaningful if the person would have otherwise shutoff their computer, ended up being on it longer than normal, or only started it it up in the first place to write something on Twitter.

    I guess I don’t see the point since you could do this for anything that you use your computer for.

    • Jonathan
      September 3, 2010 at 2:03 pm

      Posted to say this but you seem to have gotten a step ahead on me :)

  • Reply
    Flo
    August 27, 2010 at 4:52 pm

    And don’t forget how much CO2 is saved by the fact that people are forced to keep it short!

  • Reply
    Nripesh
    August 27, 2010 at 10:33 pm

    calculations are good … but i think most of the people writes or read the tweets while their PCs are already running and the got some spare time for that…
    could you justify to Turn-Off and On the PCs every time when you got a small break or just want to get refreshed while working and you know that after a while you have again get back to work…

  • Reply
    redwall_hp
    August 30, 2010 at 10:42 am

    I have a few problems with the PC power consumption average. Firstly, a *lot* of tweets are sent from mobile devices. How much power does an iPhone or a Nokia prepaid phone use? Also, computer power consumption varies wildly. Laptops are generally far more energy efficient than desktops, and some desktops (like the new Mac Mini) are very energy efficient.

    There’s also a figure that doesn’t seem to be included: the servers. How much power are Twitter’s servers using? How about each user’s DNS servers? How about all of the servers that every single HTTP request passes through?

    Calculating CO2 emissions like this is really a losing proposition. There’s always something else to account for, and you really have to wonder whether its fair to count the users’ usage at all…

  • Reply
    John
    August 31, 2010 at 6:35 am

    Don’t forget the cost of actually writing the tweet in respect of loss of productivity. Therefore requiring longer to carry out work tasks (assuming you are tweeting for non work related matters).

    Trying to find out the correct figure is an extremely difficult task with numerous possible variables. Like the post though to open up the discussion.

  • Reply
    cooljaz124
    August 31, 2010 at 4:30 pm

    I dont know, why most guys are going for the accuracy of the calculation ! Hes got a point .

    Ps : yeah, love the design too. Nicely done.

  • Reply
    Gaurav Mishra
    September 4, 2010 at 4:03 am

    Good illustration. Rest no comments

  • Reply
    Joel
    September 6, 2010 at 8:22 pm

    Great illustration, poor data analysis.

  • Reply
    OzymandiasX
    September 17, 2010 at 2:23 pm

    There is a mistake in the math itself.

    The initial number we basing these calculations on is the number of tweets/day. We calculate the amount of CO2 produced by this number of tweets to be 20,833.3 kg. This is the weight of CO2 produced by the 50 million tweets.

    To multiply this by 24 (which is what it appears is done) is erroneous. This may seem confusing as we earlier converted to hours, but that was so that we could use the existing statistics on energy which commonly use the kilowatthour measurement. The weight in CO2 is total weight produced by those tweets, not the weight per hour. If you don’t believe me, do the calculations yourself and pay attention to the units.

    The correct total (based on the assumptions and averages listed) is 20,833.3 kg of CO2 produced by the 50,000,000 tweets per day.

  • Reply
    vasiauvi
    September 26, 2010 at 6:35 am

    I will definitely not tweet this post :)