As we choose how we will decarbonize and depollute the grid, a rearguard action is still fighting hard for nuclear energy. Michael Shellenberger is one of the prime advocates of a lot of nuclear in the mix. I recently interacted with him on Twitter when he was defending the anti-wind Caithness group list of wind energy-related fatalities, a list which when I reviewed it had 50% tiny wind generation, suicides, and airplanes running into nearby guy wires in the mix. It was not the basis for statistical analysis, culled as it is from press headlines by anti-wind types with little critical judgment.
To be clear, nuclear is very low fatality per TWH. Wind and solar generation are well down the list as well, and wind is a lot better than people like those behind Caithness and Shellenberger are willing to admit simply so that they can make a case for nuclear. But when we talk about fatalities, we have to put those numbers in context in any event.
DARA maintains a series of research on climate vulnerability. In 2012, the organization indicated that there were already hundreds of thousands of deaths due to climate change per year. It projects 700,000 deaths a year solely due to climate change by 2030 and 6 million a year from the carbon intensive energy model we currently have.
As a result, any assessment of mortality has to include avoided deaths due to speed and scope of deployment, which is something nuclear can’t match easily.
So let’s look at three scenarios. The first is business as usual, the second is a nuclear-only replacement of the current fossil fuels, and the third is a renewables only replacement of the fossil fuels.
This first chart models out a year-by-year view of reductions in CO2 emissions with the base scenario flat lined, renewables happening quickly, and the deferred reductions from nuclear.
Making the point more brutally clear, this chart depicts cumulative emissions by 2050 for the grid.
How are these numbers arrived at?
Electrical generation produces about 1.7 billion metric tons in the USA per year with a mix of about 70% coal and 30% gas generation. Projecting that forward to 2050 would see 52,700 billion tons of CO2 emitted.
The coarse-grained benefits of nuclear seem good on paper. Nuclear produces about a tenth of the CO2 per MWH full lifecycle as coal and around a quarter of gas. That would probably see about 1.4 billion metric tons of saving per year when all the reactors were up and running. Sounds good, but…
It takes a median 15 years to build a single new nuclear plant per the global fleet’s stats, and replacing all of the coal and gas plants is a major program. Call it 30 years all in before it could all be replaced. So in 15 years we’d likely start seeing a reduction and in 30 years we’d see the maximum reduction. Nuclear advocates like to claim it’s a lot faster than that, but the reality of the assessment is that speed was something that was achieved early on in some programs, but more recently it takes a lot longer. Even if it were 10 years instead of 15 years, the benefits of nuclear vs renewables would still be poor.
And nuclear is expensive. Unsubsidized, it’s $100 to $150 per MWH, or about 10–15 cents per KWH. That’s the wholesale price, not the retail price, so add a bunch for the retail price for consumers and businesses.
Utility scale solar is about the same CO2 as nuclear while wind is about half of either. If we split the difference and do a 50:50 mix of wind and solar instead, we’d see another 150 million tons of savings of CO2 per year.
And wind and solar are an awful lot faster to build than nuclear, with first power within two years, and full replacement possible in fifteen years. If we compare the savings over the 30 years, we would get triple the benefit with a saving of around 33 billion tons for wind and solar vs 11 billion tons for nuclear.
And wind and solar are a lot cheaper than nuclear. Right now unsubsidized onshore wind and solar are under $40 per MWH or 4 cents per KWH, and many places are already seeing $20 per MWH. So that’s 2.5 to 7.5 times cheaper than the nuclear.
Oh, and you have to start spending the money on the nuclear now in order to get the deferred reduced benefits much later. While spending on new electrical generation isn’t a zero-sum game, there is a moderately non-elastic amount of money to spend on it. Any money spent on slow-to-build nuclear reduces the amount available to spend on fast-to-build renewables.
Economically and environmentally, nuclear doesn’t make much sense. I’m happy for every plant that gets turned on, refurbished, and not prematurely shelved because right now that means that gas and coal aren’t being used, but we can go further faster with wind and solar. There is no rational explanation for any substantive investment in nuclear. Shellenberger’s investigation into fatalities per TWH using anti-wind energy data is at best deeply misguided.