Nuclear math doesn’t add up
Writing in the conservative US magazine National Review, Robert Bryce of the Manhattan Institute criticises the Democratic Party platform for omitting any mention of nuclear power, and accuses the Democrats of failing to “do the math”. Unfortunately, although he throws some numbers about, he doesn’t do any math to support his key conclusion
But even if we doubled the rate of growth for wind and solar — and came up with a perfect method of electricity storage (which of course, doesn’t exist) — those renewables aren’t going to replace nuclear energy any time soon
So, I’ll do the math for him.
Here’s the data Bryce supplies
* In 2015, America’s nuclear plants produced 839 terawatt-hours of electricity (this industry source says 798), compared to 253 for hydro, 193 for wind, and 39 for solar
* The current rate of growth is 7GW per year for wind, with output of 2.6 TWh/GW, and 5 GW per year of solar with output of 1.5 TWh/GW
That’s enough to check Bryce’s claim. The gap between nuclear and renewables is currently 552 TW/year. The current rate of addition of capacity for solar and wind combined amounts to additional generation 25.7 TW/h per year. Doubling that gives 51.4. So, assuming a doubling of the current installation rate, it would take 10.5 years for renewable generation (including hydro) to surpass nuclear. Excluding hydro it would take 15 years. So, if the goal of policy were to replace nuclear with renewables, the answer is that it could be achieved in the fairly near future.
Of course, replacing nuclear power is not the goal of climate policy. The objective is to replace coal and, as far as possible, gas as sources of electricity.
So,Bryce is right to say that ““widespread retirement of America’s nuclear power plants would make it extremely difficult if not impossible”. If we didn’t know that a rapid shutdown of nuclear power would be a big problem for a decarbonization policy, the examples of Germany and Japan make it pretty clear.
But Bryce is shifting the ground. he fact that The Democratic platform says nothing about nuclear power implies that they don’t have a policy for a rapid shutdown. In fact, the Obama Administration has consistently supported an “all of the above” policy including loan guarantees for nuclear power.
The problem is that new nuclear power is hopelessly uneconomic, and even existing plants are finding it hard to cover their costs. Bryce doesn’t propose a solution, presumably because any solution would require government intervention on a scale so massive as to horrify the readers of National Review.
Still, it might be worth looking at the possibilities for an expansion of US nuclear power in the foreseeable future. Apart from a handful of plants under construction there’s no chance of any new ones in the 10-15 year time-frame we just derived. So, let’s look out a bit further to the early 2030s. On past experience would require starting construction by 2020, which means we can confine our attention to proposals that have already started the licensing process.
In the first flush of enthusiasm for the “nuclear renaissance” of the early 2000s, proposals were put forward to build around 30 new units, typically of 1 GW each. Four of those have gone ahead (over time and over budget as usual), and should be completed by 2020. At least ten have been abandoned, while the remainder are in limbo, being kept alive as options but without any actual prospect of occurring. That’s a maximum of around 15 GW, or around 100 TWh per year.
The new additions have already been offset by closures. Even with the most friendly policy environment possible, we’re bound to see more over the next 15 years, given that many plants are already well beyond their initially planned operating life.
So, even with an all-out effort to implement every currently proposed nuclear power plant, and to keep closures to a minimum, we might get a net addition of 50TWh a year. That’s the equivalent of two years of additional solar and wind at the current rate, which has been increasing steadily over time.
* For completeness, I should mention Small Modular Reactors, which are often seen as the nuclear hope of the future. The only project that’s currently active is that of Nuscale which aims to produce a demonstration plant by 2025. Even if SMRs prove economic (there is no reason to believe this), large scale deployment will be well into the 2030s.