Count Nuclear Waste, Not Just CO2

How much carbon does the average American or European emit per year? How much does the world emit? And if you know the answers to those questions, maybe you can also tell Craig Morris how many tons of nuclear waste the world has? He tried, and failed, to find out.

Radioactive Waste

Radioactive Waste in Nevada – how much is there? (Photo courtesy of National Nuclear Security Administration / Nevada Field Office)

Last week, we read alarming reports about CO2 concentrations in the atmosphere are approaching 400 parts per million – and counting. In the run-up to the second anniversary of the disaster in Fukushima, we also read articles by proponents of nuclear power – such as this one from Sunday in New Scientist. In addition to reiterating the pro-nuclear argument that “relatively few people will suffer serious health effects from the nuclear fiasco,” the article states: “the world still fears nuclear technology more than it fears climate change.” Is that true?

Counting carbon emissions is no easy task. The shift from coal to shale gas, for example, has lowered the emissions we count, but questions remain about emissions we may not count, such as line losses and methane escaping from the mining process. To make things worse, small businesses (such as the small mines in China) have trouble estimating their carbon emissions.

A German power bill showing nuclear waste production

A German power bill showing nuclear waste

You would think it would therefore be much easier to count nuclear waste. After all, there are not really any small businesses involved in the nuclear sector, and nuclear waste production is heavily monitored.

For instance, a Wikipedia site lists countries by per capita carbon dioxide emissions. A Google search reveals a wide range of websites and articles with related information.

Yet, there is no Wikipedia site on per capita nuclear waste production. The entry on radioactive waste does include a paragraph on “legacy waste,” which offers such scientifically exact estimates as “millions of gallons,” “thousands of tons” and “huge quantities.” Later, we see that “about 12,000 metric tons” of high-level waste are produced every year. Amazingly, the website includes a chart of “uranium and thorium release from coal combustion” (cumulatively 2.9 million tons up to 2040); apparently, coal plants release more radioactive material than nuclear plants do!

Good data are hard to find

Doesn’t the International Atomic Energy Agency provide updates every year? There is an estimation (PDF) with data from 2008 (published in 2010), and it shows that high-level waste is not the same as spent fuel, which is not the same as “LILW” (liquid intermediate level waste). The totals are then given in scientific notation, such as 1.8 E5 for MTHM – any questions?

We state emissions of heat-trapping gases in terms of carbon-equivalent emissions so people get the message. Is there no way to state the amount of nuclear waste for laypeople?

The study is not updated annually, but the IAEA does explain on its website that a 1,000 MW nuclear plant produces around “30 tons of high-level solid packed waste per year.” France has around 62,000 MW of installed nuclear capacity, so the country should therefore be producing 62×30 tons per year, or 1,860 tons – equivalent to 28 grams per Frenchman. So why does a presentation from 2006 (PDF) claim that the French “only” produce 10 grams of high-level waste – still just half of the Nuclear Energy Institute’s estimate of “20 metric tons of used nuclear fuel” from of “typical nuclear power plant in a year”?

If we were actually more concerned about nuclear risks than climate change, why would we know so much about carbon emissions and so little about nuclear waste production?

Volkswagen counts nuclear waste production from its power plants

Volkswagen counts nuclear waste production from its power consumption

As you might expect, things are different in Germany. For instance, Volkswagen owns its own power plants and is pleased that it has lower nuclear waste production than the German power mix (see chart to the right). Do other power producers outside Germany report their nuclear waste production? Is anyone outside Germany counting it along with carbon emissions?

Craig Morris (@PPchef) is the lead author of German Energy Transition. He directs Petite Planète and writes every workday for Renewables International


Craig Morris (@PPchef) is co-author of Energy Democracy, the first history of Germany’s Energiewende.


  1. Martin Vermeer says

    The difference between CO2 reporting and radioactive waste reporting is that CO2 is well-mixed in the atmosphere and emissions are “fungible”: every kg of CO2 emitted has the same climatic effect, no matter where it comes from.

    With radioactivity, not so. And BTW kilograms or tons or gallons etc. are silly units, because they don’t reflect the biological activity, or even the radioactivity, of the waste. E.g., contaminated clothing or building materials etc. may be quite voluminous but radiating only very little.

    A better unit for radiactivity is the Becquerel, which expresses how much radioactive decay is going on. But that again doesn’t say how much will reach the internal organs if you’re standing by the side of a source, as penetration power depends on the type of radiation.

    Then there’s biological activity: Sr collects in the bones, I in the thyroid gland, leading to continual exposure. Noble gases like Xe again dilute into the wide atmosphere and are chemically inactive (Radon is a special case). Tritium is also fairly harmless, once it is diluted into the seriously huge hydrogen pool of the world ocean, from which it is chemically indistinguishable.

    What we need is a measure for the energy deposited in tissue: the Gray. But the biological effect of radiation differs for different radiation types, and is expressed in Sieverts. Only for gamma radiation is 1 Gy = 1 Sv. And as said, there is no simple link between either Gy or Sv, and Bq.

    Confused yet?

    I think the problem is not an unwillingness to report as much an inability to report in a way that would be transparent to the readership, given the above complications. But perhaps something could be developed.

    • Craig Morris says

      Martin, interesting comment, but you do know that “carbon” emissions is already a simplification of a much more complex reality. Other things are converted into carbon-equivalent emissions, some things are not counted (sometimes depending on where they are emitted), and not all emissions are counted or counted correctly (such as methane from agriculture – cow farts and rice fields – or emissions from small coal mines in developing countries).

      It should be possible to count the tonnage of spent fuel rods, don’t you think?

  2. Martin Vermeer says

    Actually, no. Most of the tonnage of the spent fuels is made up of U-238 (in practice an inert “filler” material, and only extremely weakly radiating), and precisely how much of that, depends on the degree of enrichment of the original fuel, and the amount of “breeding” that has taken place, i.e., conversion of U-238 to plutonium, before the rods were taken out.

    Actually a more sensible metric would be the amount of material fissioned. But additionally one should measure the amount of actinides formed by neutron activation, as these materials, with lifetimes of thousands of years, form a separate, special problem.

    So I think it’s doable, but it’s not simple.

    • Craig Morris says

      Right, then we agree that the answer is yes, not no: we can count nuclear fuel waste in the same way that we count carbon emissions – by oversimplifying a complex issue.

  3. Martin Vermeer says

    Sure, I agree we _can_ count nuclear fuel waste in this way; but you have still failed to convince me why this would be interesting.

    CO2 released into the atmosphere _has_ as detrimental effect directly connected to its amount. That makes establishing the amount of this gas a relevant and interesting undertaking. Warts and all.

    What I think you overlook is that nuclear waste, of whatever type, is only a _potential_ environmental and health hazard, though a serious one. That makes the precise amount of the stuff sitting in storage ponds and dry casks or whatever, of limited interest. An actually interesting factor would be the safety culture, which can make an orders-of-magnitude or even yes-or-no difference for the actual hazard. All major nuclear accidents have been failures of safety culture. I hardly need to elaborate.

    I even see a small danger that if we start focusing on quantification like this, we lose the perspective on the less quantifiable but actually more important safety culture aspect.

    • Craig Morris says


      You can see nuclear waste as “only a _potential_ environmental and health hazard, though a serious one” or as a time bomb.

      Assuming that the age of fossil energy will be over in a couple of hundred years, carbon emissions from energy consumption will eventually taper out. You and I would probably agree that it’s going to be a wild ride. If there are positive feedback effects, it may change the planet irreversibly for many forms of life. Or things may calm down in a couple of centuries, possibly without so many humans.

      Whoever and whatever is left will then have to stand guard over this nuclear waste for an unimaginable time span. That is the best-case scenario.

  4. Martin Vermeer says

    BTW why make it complicated? The metric you’re after already exists, in the form of the total amount of GWh generated by nuclear power plants. Think: the thermodynamic efficiency of nuclear plants lies in the narrow range 30-40%. The amount of energy released by a single fissioned atom, or by a gram of fissioned material, is also within a narrow range irrespective of what the fissioning nuclide is.

    The total amount of radiation that the fission products combined will release over time may vary a bit more, but still not by orders of magnitude…

    I would be hard pressed to construct a more precise metric than this.

  5. Martin Vermeer says

    > Whoever and whatever is left will then have to stand guard over this nuclear waste > for an unimaginable time span. That is the best-case scenario.

    Don’t quite agree. One may hope that we get through the anthropocene in a state that will allow us to dispatch this responsability. I would expect it however to be one of the lesser challenges. If we come through with a somewhat intact global civilization, we will also have the technology to handle this waste much better than just by guarding it forever: remember it (the actinides) are _fuel_. You can burn it up by fissioning.

    The actual fission products don’t have this “forever” problem: they halve by themselves every 30 years. And even if some part of it somewhere would escape into the environment, that would be bad, but not the end of the world. A Chernobyl replay at worst.

    No, the worst threat remains weapons proliferation. Especially in a world destabilized by climate disruption. That, for me, is the main reason to want to avoid the complication of a global nuclear power economy. And I’ll believe nuclear power is “safe” when the UN Security Council loses interest in Iran’s “peaceful” programme.

  6. Martin Vermeer says

    Craig, an illustration of my point: the spent-fuel pond atop reactor building 4 in Fukushima. There have been collected over the years (decades?) a great many fuel-rod assemblies, that now have to be removed, at great effort and expense, from the top of un unstable wreck.

    Proper regard to safety would have transferred each of these assemblies to dry-cask storage after a couple of years of cooling down. This single issue represents many orders of magnitude of unnecessary risk taking. Focusing here on the _amount_ of spent fuel, quantified by whatever metric precise or otherwise, would be a distraction at best. At worst, I am reminded of Goodhart’s law:

Leave a Reply

Your email address will not be published. Required fields are marked *