E-mails not all that ‘green'

D. BALASUBRAMANIAN

Share  ·   print   ·   T+  

Costly infrastructure: The use of personal computers, servers, storage centres cost energy and hence more CO {-2} emission. — photo: AFP

During Deepavali time, these days, we are inundated with electronic greeting cards, and we too send several such e-cards ourselves. We believe that by switching from paper to electronic mode of communications, we are “green”, and that in doing so we have saved paper and thus done a bit to save the environment and generate less CO {-2}.
Well, perhaps just a bit but it appears not as much as we are led to believe. “E-mails are not so green” reports a news item in a recent issue of the journal Science .
The often-quoted estimate by Mr. Matthew Yeager of Computacentre (Europe's largest IT infrastructure company) claims that sending an e-mail attachment of 4.7 megabytes (MB) creates as much greenhouse gas as boiling a tea-kettle 17.5 times.
His study claims that an e-mail of 1 MB would be the equivalent to the emission of 19 grams of CO {-2} {-.} And if that mail is copied (cc'd, as we type) to 10 people, its impact is 73 grams of CO {-2} {-.}
Well, I was astonished to read this, since I too believed that I was saving the planet a bit by using my PC to communicate with people, instead of “snail mail”.
Keira Butler explains the matter in an issue of the magazine The Atlantic (August 12, 2010). She says “Say you send a picture to 20 people by email.
Each one has to download it. That means the use of equipment such as personal computers, servers, storage centres (not to mention printers for hard copy, if used)”. All these cost energy and hence more CO {-2} emission.
It is a matter of scale. Matthew Yeager points out that the current amount of data storage across the globe is 1.2 zettabytes (ZB) of stored data. This requires equipment with a mass equivalent of 20 per cent of the island of Manhattan, New York City! Put another way, this level of stored data is the equivalent of all of the US' academic libraries multiplied by half a million! And the data storage is expected, by the year 2020, to grow to 35 ZB (incidentally, zetta is a sextillion, or 10 raised to the power 21 (or 1 followed by 21 zeros).
The scale increases thousand-fold each time from million or mega, to billion (giga), trillion (tera), quadrillion (peta), quintillion (exa), sextillion (zetta), septillion (yotta) and so forth).
E-mail is thus not all that green. And e-mails with attachments are worse. Yeager estimates that in a 100-people company where each employee sends on average 33 e-mails a day and receives 58, the greenhouse gas emission linked to emails would be around 13.6 tons of CO {-2} per year.
And a study by the French government's Environment and Energy Management Agency (Ademe) suggests that if each of these 100 employees sent 10 per cent less emails for a year, they would save CO {-2} emissions equivalent to one round-trip flight between Paris and New York.
Talking of CO {-2} emissions by airline traffic, I was reminded of what Dr. Jeremy Nathans of Johns Hopkins wrote to me (by e-mail, not snail-mail) when we invited him to come to Hyderabad for delivering the Champalimaud Lecture in 2009.
He declined coming in person, stating that he is doing his bit to the environment by not flying all the way from Baltimore and back. We had him lecture electronically (video talk real time; I should now estimate how much CO {-2} he would have saved by not flying but video-lecturing).
To get an estimate of how much power is consumed by electronic communication, go to the website < http://whatsthisgottodowithstoragefiles.wordpress.com/2010/08/wired-uk-july-2009-internet-electricity.pdf >.
They point out that 30 per cent of the input power in each computer is used in powering the chips, 30 per cent of the energy entering a microprocessor is turned into heat, and that 123 billion kilowatt hour (kwh) per year is how much electricity it takes just to keep the Internet's servers running.
And traditional IT environments, says Yeager, tend not to be overly efficient in scale. Traditional infrastructure — server plus storage plus network plus operating system plus application — all lead to wastage in efficiency. Combine this with what Keira Butter points out in The Atlantic , you get an idea of how much energy is lost in electronic communications. Yes, e-communication does save trees, is more efficient and produces less CO{-2}than paper-based communication. But the scale of it is what needs to be kept in mind.
Take Facebook usage. It is estimated that its users alone are uploading over 1000 photos per second, or 3 billion photos per month. Recall the tea kettle boiling equivalent of sending a 4.7 MB attachment, and you get the idea.
What should we do?
So what should we do? There are several ways of saving energy and cutting down greenhouse gas from our end. First, free up the memory space in the computer. Clean up the e-mail box (in and out mails) periodically. Not doing these means greater demand for storage and energy used by that storage.
Second, limit the number of recipients for each e-mail (cut down the number of cc's to).
Third, cut down the size of the attachments (boil less tea- water).
Fourth: enter the URL address directly rather than use a search engine. Cut down the times you “Google”, “Yahoo” etc.
Fifth: don't leave your computer and accessories on overnight (as many offices do), not even on ‘ sleep mode' (even if that eats up only 1-10 watts).
Sixth: laptops use 15-60 watts while desktops use 250W. Cut down the power by doing more ‘offline' work than online. Finally, remember Facebooking and Twittering burn carbon and make CO {-2}. Talk more and twitter less!
D. BALASUBRAMANIAN
dbala@lvpei.org