Friday 28 August 2015

Carbon emissions from TV versus books, revisited

A big surprise for me when I wrote my book was that watching TV (usually) generates less carbon emissions than reading a book. However at that time I was considering a paper book whereas these days we use e-readers a lot of the time. How much difference does that make? For a fair comparison, I am using carbon emissions per hour watching or reading.

The trouble with carbon foot-printing is that there are so many variables it is impossible to give straight answers. If you twist my arm I would say that the TV is now only a little better than the e-reader in a typical case. In fact if you look at the chart below you would think that the e-reader was scarcely better than the paperback book. However, the worst case for the paper book is when it is read only once, which is likely, whereas the e-reader worst case is a lifetime of only 25 books, which is much less likely. The more you use the e-reader, the lower the emissions. If you read newspapers on your reader too, then the carbon savings mount up rapidly.

At the end of this post I offer answers to personal questions such as, should I watch the TV tonight and should I buy an e-reader?

Estimates of carbon emissions per hour from watching TV or reading a book.

This chart shows my best estimates for carbon emissions per hour with the assumptions described below. I have included the embodied emissions as well as the power use of the TV or reading device. Also I have included the programme production and transmission for TVs and the authoring and publishing emissions for the books as well as printing for the paper version.

Watching TV

The carbon emissions from watching TV come in three categories: the production and transmission of the programme, the electricity use watching the TV and the embodied emissions in the TV itself. The latter is highly dependent on how much you watch the TV. By varying the lifetime from 2-15 years I have estimates of between about 5 and 40 gCO2/hour: at the lower end this is negligible and at the upper end it is half the emissions. However, there are other variables to consider as well. Taking into account other factors you could double or halve the overall totals.

Carbon emissions per hour from watching TV

The assumptions I have made are:

Number of people watching: 1.5
More people -> less emissions
The BBC reckons the average is 1.46 viewers [1]. In real situations you don’t get fractions of people but perhaps you watch alone half the time and half the time with a friend. If you are always together you can reduce the emissions by 25%. If you are always alone, add half as much again.

TV embodied emissions 200kg CO2
Bigger TVs-> more emissions
Typical hours of use 4.5 hours/day
Lifetime: 2, 5 or 15 years
More use and longer lifetime -> less emissions
I have taken the embodied emissions of the TV as 200kg, typical of a 32” LCD display [1]. For the per-hour emissions you divide by the total use time. At 4.5 hours/day, the national average, and replacing it every 5 years, the total use time is 8200 hours so the embodied emissions are 24 gCO2/hour. However, if you have a bigger TV this could be double, or if you watch it only 2 hours/day then the use time is down to 3,600 hours and, again, the per hour emissions more than double.

If there is only one of you watching, then you might use a smaller TV or you could use a laptop. A laptop has only about half the embodied emissions [2] and uses less power too.

TV on mode power consumption 80W
Bigger TVs -> more power-> more emissions
80W is a little high for a 32” TV on the market today but you might need to add in a set top box and/or a sound system too. If you have a larger screen you will use correspondingly more power. TVs now have EU energy ratings from A* downwards. However, as always with ratings, they compare TVs of similar size and functionality. A large A* TV can use more power than a smaller A or B rated TV.

Programme production and transmission: 11W
For production, the BBC reckons on 10 W/viewer/hour averaged across all programmes and including repeats [1]. Some programmes are much more carbon intensive than others to produce. For example outside broadcasts and those involving lots of travel have high emissions. However I expect they often have more viewers too.

Each channel is transmitted in the same way regardless of how many people are watching. For BBC1 the emissions are less than 0.5 gCO2/viewer/hour but for BBC3 and BBC4, because they have fewer viewers, it is about ten times that and for BBC Parliament, the emissions come to 196 gCO2/viewer/hour, about 400 times as much as BBC1 [1]. Of course not watching BBC Parliament does not actually reduce the emissions as long as the BBC is required to broadcast it

Reading a book on an e-reader

I have estimated the carbon emissions for reading a book on an e-reader in four parts. Those due to authoring and the publishing office are divided by the number of books sold to get the emissions per book and then the number of hours to read the book. The other two parts depend on the reading device. Even more so than with the TV, it is the embodied emissions of the reader that are the biggest problem so the more you use it the better.
Carbon emissions from reading on an e-reader

Here are my assumptions:

Time to read book: 10 hours
More time-> less emissions/hour
The time to read the book rather depends on the length and nature of the book. I think 10 hours is reasonable for a 500-page novel or a shorter non-fiction book.

Book size: 5Mb
Storing the book (on a WD Red 4TB HDD, modern spinning disk): 0.5 mW/GB
Delivering the book to the e-reader: 0.05 kWh/GB
Negligible anyway
Based on these assumptions, storing the book in the cloud and delivering it to your e-reader over the internet is negligible. For a discussion of the download carbon see ‘Mobile network energy in fridge units’.

Reading the book: 2W
Almost negligible.
A tablet sized reader with a backlight might use 2-3W, the carbon emissions are then about 1 gCO2/hour, so practically negligible. A Kindle Paperwhite takes considerably less than this: the battery stores 5.25 Wh and is supposed to last 21 hours so 0.25W. That is definitely negligible.

Reader embodied emissions: 40kg (Kindle)
Lifetime of reader: 25 - 100 books
Longer lifetime -> lower emissions/hour
This is by far the most important component of the overall emissions. Even if you read 100 books on the device, it still comes to 40g/hour, 3/4 of the total. However, you can use your reader for newspapers as well as books. Counting a newspaper as equivalent to half a book, a year's worth would be 180 books.

The embodied emissions of electronics like tablets and computers are roughly related to physical size [2]. A larger device will have more embodied carbon. However, you might do a lot more with it than read books, in which case the proportion of emissions related to book readings would be less.

NB. The authors of [2] suggest that their method typically underestimates emissions by a factor of 2 or 3 because there are aspects of production not included. To be conservative I have multiplied their figure given by 3 giving a total of  40 kg. This is still much lower than the figure of 186 kg I have used in previous work [10] but footprinting methods have improved since that was published, back in 2009. Also, Apple says the emissions from their iPhone are 94 kg and it is simply not plausible that a Kindle would be twice as bad as that.

Number of copies sold: 5000
More copies sold -> less emissions/book
Top titles sell over a million copies in their lifetime – many self-published books sell less than a hundred. According to [4], 5000 is regarded as healthy sales for a small publisher. This number matters because the authoring and publishing costs are fixed for the title and I have divided by the number sold to get the figure per book. Self published books sell even less so they are unlikely to be on your reading list.

For the less successful books, I think we can say the carbon emissions for preparing them are the responsibility of the people who choose to write them, not the people who don’t read them.

Computer time: 18 months at 4 hours/day at 60W
Printing drafts: 5 drafts, 2 sided, 2 to a page, 80g/cm paper
Embodied emissions of computer and display :146 kg
Total 221 kg
This is based on my experience writing my book. It may be a bit different for fiction. Since I did this at home I have only counted the electricity consumption and embodied emission of the equipment I used, not the heating and lighting. I have allowed 477kg for my Mac Mini [5] and 300kg for my display [2] and I have allocated 20% of the use of these to writing my book. The total for all authoring comes to 221 kg CO2, mostly due to embodied emissions in the equipment. That sounds a lot but divided by 5000 copies sold, at 10 hours reading time that is only 4 gCO2/hour.

Publishing office:
300 books published/year by 120 staff [6]
Office emissions 100 kgCO2/m2 [7]
Each person occupies 10 m2 [8]
Total 400kg
With these assumptions, the total emissions come to 400 kg for the title, so rather more than the authoring but still only 8 gCO2/hour for a reasonably successful book.

Reading a book on paper

The authoring and publishing emissions for a paper book are the same. The big difference is in the paper and printing. Another big difference is that you can lend the book to a friend instead of them buying their own copy. However, most books are only read once, which is the worst case.

Carbon emissions from reading a book on paper

Number of times you read the book: 1, 2, 3
For paper books, read more times-> less emissions/hour
Most books in my library I only read once. If they are good I might lend them to a friend and/or read them again. However, very few books will be read more than 3 times. In fact some books don't get read even once - if they are no good I might not bother to get to the end.

Printing a book: 1.8 kgCO2/book
This figure comes from [9]. I believe this includes delivery too as well as setting up, paper and printing. Presumably it depends on the size of the book at least to some extent, but the figure is 6 times what I would expect for paper alone - though half the books that are printed aren't sold anyway. They end up pulped and recycled. Using a print on demand service reduces this waste.

Your personal choices

This is all very complicated – you are probably thinking, what does it mean for me? Here is some advice.

Should I watch TV tonight, instead of reading my book?
Well if you’ve bought the book, then you might as well read it. (If it is a paper book, then you can get more out of it by lending it to a friend to read instead of buying their own.) On the other hand, if you have a TV and don’t use it, the embodied emissions are massive.

The fact is, this is the wrong question to ask. The important choices are what sort of TV to get, how often to replace it, and whether or not to get an e-reader.

What sort of TV should I get?
Generally speaking, smaller TVs are lower in carbon emissions, both from the point of view of embodies emissions and power consumption. However, this is only a rule of thumb and you really need to check the estimated annual power consumption for each model. You will find it on the EU energy rating label – if you can find the label. All TVs must have one but, especially if you are looking online, they aren’t always obvious.

If you don’t watch TV much, then maybe you don’t need one at all, or you could use a laptop computer instead that you have anyway.

How often should I replace my TV?
The longer you keep it, then the less important the embodied emissions become. From the chart you can see these are relatively minor after 5 years at 4.5 hours/day (about 8000 hours) for a middling size TV. If you use the TV less than this, perhaps you could live with the old one a bit longer. However, if you watch 10 hours a day then it would make little difference if you replaced it every 2 or 3 years.

Should I buy an e-reader?
You need to read at least 20 books on an e-Reader to justify the embodied carbon emissions compared to the equivalent books. If you really don’t read much, but you get a daily newspaper then that would also justify the e-Reader - a quality paper like the Guardian is equivalent to half a book or thereabouts.

Having got the e-reader, you should of course use it instead of paper whenever you can. The power consumption is tiny so the additional carbon emissions for reading each book is minor. However, every paper book you buy means another 1.8kg or so of carbon emissions just for the paper and printing.

You aren’t an average person so I can’t give you straight answers, but your choices still matter.
When I am doing consultancy work for an organisation such as Department of Energy and Climate Change, I only need to deal with national averages. For example, if I were asked to estimate the impact of improving the efficiency standards of TVs then I would calculate the carbon emissions saved using the average number of viewers as 1.46 people, the average use time of a TV, and the average size of TV. If I could find more detailed data I would break down the population by TV size, using average of hours of use and lifetime of each size of TV.

However, you aren’t an average person and and that is why I haven’t been able to give you straight answers here. Arguably, since you don’t have control over how books are written or how TV programmes are made I shouldn’t have included those processes. However, I think it is important to know where the main emissions are. After all, if the programme-making emissions were much more than the TV watching there would be little reason to worry about what sort of TV you used. As it turns out, the programme making emissions are relatively minor (assuming the BBC is representative) but your choice of TV and how often you replace it matters quite a lot. Similarly, for books the emissions due to authoring and publishing are minor, at least for the books you are likely to buy. However, your e-reader will not pay back in carbon emissions unless you read 20 or so books on it.

[1] A comparison of the carbon footprint of digital terrestrial television with video-on-demand, Jigna Chandaria, Jeff Hunter, Adrian Williams
 (BBC white paper) March 2011
[2] Comparing Embodied Greenhouse Gas Emissions of Modern Computing and Electronics Products, Paul Teehan and Milind Kandlikar | Environ. Sci. Technol. 2013, 47, 3997−4003
[3] Life cycle assessment of tablet PCs, Paul Teehan (Green Electronics Council) 2013
[4] Bestsellers and book sales expectations (
[5] Mac Mini environmental report (Apple) 2014
[6] CMAP #2 : How books are made from Charlie’s diary by Charlie Stross
[7] Building Performance Evaluation Programme: Early Findings Non Domestic Projects (Innovate UK) 2015
[8] Occupier Density Study (British Council for Offices) 2013
[9] Environmental Book Printing Overview (printondemand-worldwide,com)
[10] The Environmental Impact of Amazon's Kindle, (Cleantech group) 2009

No comments:

Post a Comment

Comments on this blog are moderated. Your comment will not appear until it has been reviewed.