IT鈥橲 midnight, again. I鈥檓 re-reading the recent correspondence to this
magazine about how much carbon dioxide nuclear power stations produce.
Ian Hore-Lacy of the World Nuclear Association cites studies which conclude
that building and fuelling nuclear power stations consumes between 6 grams and
40 grams of carbon per kilowatt-hour of electricity generated
(28 July, p 72).
Coal-fired power stations consume almost 1000 grams.
I want to check this. But how? I could pore over paper copies of the studies,
but that would consume significant amounts of carbon, in coffee-brewing alone.
So I try a different tack.
Advertisement
Some fraction of the waste nuclear power stations create will need to be
stored or buried, with warnings that disturbing it is A Bad Thing. The warning
will have to last a long time. Plutonium-239 has a half-life of about 24,000
years. After 10 half-lives鈥攕ay 250,000 years鈥攐ne-thousandth of the
original amount will remain.
Somehow, I鈥檓 not confident of designing a sign that can be read 250,000 years
hence. Better leave a message with people鈥攁 self-sustaining community of
security guards. Taking a sunny view of the technological and cultural future,
you may imagine a happy band with jobs for life. Picture them relaxing with
their families in their gardens, occasionally poking the barbecue coals.
Less optimistically, envision a rather puzzled hereditary caste of
security-priests, huddling round their camp fires as the apprentices chant
operations manuals in a language that was already dead when they were written in
2053.
Conveniently, both the barbie and the bonfire model lead to the same carbon
consumption. Estimate about half the current world average, or 1/30th of the
US鈥檚 consumption: 0.2 tonnes per head per year. Multiply that by a community of
100 by 250,000 years and you get 5 million tonnes, or 5 脳 1012 grams. The
lifetime output of a nuclear power station is around 1011 kilowatt-hours. So if
the waste is stored on-site at each station, we need to add 50 grams per
kilowatt-hour to the total estimated by Hore-Lacy.
The security-guard problem doesn鈥檛, by itself, bring nuclear power into the
same climate-changing league as coal-fired behemoths or off-road vehicles
pootling down to the postbox. But it does show the hazards of making a
prediction without taking into account the longest timescale and the biggest
numbers you can think of.