Recently it occurred to me that having my dual-core 2.8Ghz computer running while I read a document had the potential to be far more damaging to the environment than printing the document out, so I set out to calculate which was worse, using the best resources I could find. If anyone finds more accurate numbers, please let me know.
To make the comparison I will try to estimate the embodied energy cost of a single printed page of A4, and compare that with the energy used by my computer while running for the length of time it takes to read the page. I'm only addressing the marginal costs of course, e.g. I'm not considering the energy it took to create the computer or printer, and amortising that over the number of pages printed/read.
So what is the embodied energy cost of a single printed page of A4?
A University databook put the embodied energy of "Paper" (which I have assumed to be bleached white printing paper) at 36.4MJ/kg. So for 80gsm (grams/square metre) copy paper this comes to 2.9 MJ / square metre. A4 paper has an area of 1/16th of a square metre, so a single sheet of A4 has an embodied energy of around 180kJ.
How much energy does it take to run your computer?
Figures vary wildly between processors and monitors, so I've taken a rough estimate at 200W. Hopefully this should be an upper-end estimate representation of a standard desktop system. However, electricity generation is far from 100% efficient. Coal Fired power stations (which for simplicity we assume are the only source of UK Electricity) had an energy efficiency of 36% in 2008, so that 200W used requires 560W of fuel use. At 560W, it takes 320 seconds to clock up 180kJ; or 5.5 minutes.
So allowing all those caveats and approximations, if you're going to spend more than 5 minutes reading each page, the environment would be better off if you printed it out and switched off your desktop.