Decarbonizing the U.S. electricity sector by phasing out existing nuclear is technically, economically, and environmentally better
This post is a complement to a prior one, which dealt with: (1) Is nuclear clean? and (2) Can new nuclear power plants reasonably be expected to contribute significantly to decarbonization in a timely way, given the climate crisis?
Answers: (1) No. (2) No.
What about existing nuclear power plants? Let me start this post as well with the position of the Nuclear Energy Institute, the industry’s think tank. It claims that “the policy environment” does not, at present, value the “role [of nuclear] in providing clean energy”; it advocates for “[p]olicies that value nuclear energy’s carbon-free electricity can ensure that these plants will continue to be part of the energy portfolio of the future.” (NEI, p. 6) In other words, the industry wants money – billions of dollars a year – to keep existing, mostly depreciated plants (average age: 40 years) operating in addition to the revenues from electricity sales. Their advocacy has succeeded as states have passed 100% policies in places such as New York, New Jersey, and Illinois, strapping ratepayers already paying hundreds of millions a year to keep a few aging nuclear plants open.
As we look solely at emission levels, let’s recognize two things at the outset. In a world where fossil fuels are used in essential sectors like transportation and construction, all power plants, including solar, wind, and nuclear, will have some emissions when they are built and decommissioned. But, as a thought experiment (Einstein was fond of them, so the process has an excellent pedigree), if we singularly consider only nuclear, only wind, or only solar, what would our emissions be with any one of them as our energy system. The answer: close to zero (except maybe for the cement). A 2017 paper in the journal Nature Energy, that looks at a lifecycle emissions of these power plants in a future (2050) low carbon energy system, provides significant substantiation of this thought experiment.
Second, extensive work, done by many independent researchers, on “lifecycle emissions” points to a simple conclusion: that CO2 emissions associated with solar, wind, and nuclear are much lower than the lowest emission fossil fuel plants. 
States have been using or are considering a mix of nuclear, wind and solar. But there is a very important technical reason, with major economic implications, to phase out nuclear as the grid is decarbonized and more and more solar and wind feed power into the grid. Nuclear power plants do not operate well with solar and wind as the amounts of these variable renewable sources increase to significant levels.
There is general agreement that most of the growth in new electricity capacity will come from solar and wind, which are by far the most abundant and economical new electricity sources. Unlike nuclear, which is 24/7 when it operates normally, the output of solar and wind fluctuates a great deal. To get an average of one kilowatt of output for the whole year requires roughly four to five kilowatts of solar capacity and two to three kilowatts of wind. (The exact values depend on location, but that is not important to the present argument.) As a result, at a point when much or most of the annual electricity requirements come from solar and wind, the peak installed solar and wind capacity – the maximum they can produce at the best resource moment – will be several times the actual peak load. In other words, while wind and solar would fall short for many hours, they would also greatly exceed the electricity requirement for many hours in the year. The latter phenomenon is called “overgeneration”.
This simple technical fact means that solar and wind need complementary resources that can fill the gap when they are short and can backdown or absorb the excess when there is excess wind and solar supply. And these complementary resources must do it fast. When the wind drops suddenly or a cloud goes over a utility-scale solar installation, the output can drop precipitously in seconds or a couple of minutes. As the cloud passes, the power output will surge. Thus, complementary sources must supply more power fast when the cloud covers the solar and then back down fast when it passes. Batteries do this extremely well. Hydropower plants respond very rapidly too. Nuclear power plants are a very bad fit for this requirement. It takes two hours to change the power level by 50 percent even in advanced reactors of the light water design, the design used in all operating U.S. power reactors. That is far too slow. Power supply and demand must be balanced in seconds; if not, brownouts and blackouts are the result.
The inability to change power levels rapidly means that when nuclear is the main complement, it is solar and wind that must be curtailed. This is economically and ecologically poor. It curtails the cheapest sources of electricity that cost almost nothing to operate, once built (technically known as “low marginal cost of production”) in deference to nuclear, which costs more, makes radioactive waste and plutonium, and increases demand for uranium, meaning more radioactive waste.
Nuclear proponents claim that new reactor designs would greatly improve upon this. But even then, significant changes in power level to prevent solar and wind from being curtailed take far too long. Further, curtailing nuclear means that it is no longer operating 24/7 as designed; its costs are based on that assumption. For instance, if a Small Modular Reactor (SMR) followed solar and wind to accommodate supply variations its costs would increase significantly. That’s on top of already high costs. Changing power in a nuclear reactor frequently also creates wear and tear issues and, if too fast and frequent, potentially safety issues as well.
The mismatch between the technical capability of nuclear and the variable characteristics of wind and solar was one of the main reasons that PG&E, California’s largest utility, decided to shut down its Diablo Canyon plant in the mid-2020s. It was looking ahead to California’s growing solar portfolio and realized that the plant would no longer be economical toward the end of the 2020s. Here is what the joint report about its retirement said about overgeneration:
Overgeneration conditions can force the system operator to take action to curtail generation (e.g., dispatch generators down, or even disconnect supply from the grid) in order to maintain electric system reliability. Retirement of Diablo Canyon on the timeframe agreed to in the joint proposal will allow for increased flexibility for the California electric system so as to help maximize the value of solar and other variable resources that will be a crucial part of meeting PG&E’s renewable targets and California’s renewable and GHG emissions goals.
While there is an effort to keep Diablo Canyon open for a complex of energy production of semi-tested technologies that would hardly have time to be evaluated, much less tested fully and built commercially PG&E, the plant’s owner, dismissed the report saying that the matter had been decided and the closure “approved by the California Public Utilities Commission and the state Legislature.”
Incompatibility with growing wind and solar is not the only problem. Coal and nuclear plants use vast amounts of water, mostly freshwater. Once-through cooling systems that can and do draw in billions of gallons of water a day, heat it up and discharge the water back to rivers, lakes, and seas. As heat extremes worsen, thermal plants sometimes fail to meet the discharge temperature limits meant to protect aquatic life and ecosystems. So, they have to curtail their power at the very times when power requirements are highest due to the heat.
Fresh surface and ground water withdrawals by thermal power plants (mainly coal and nuclear) in the United States amounted to 95 billion gallons a day, about two-and-a-half times the withdrawals of all public water supply systems in the country and a third of freshwater withdrawals for all sources, including agriculture (from USGS data). Most of the water is returned to the rivers and lakes, but heated up. As a result, some, about a billion gallons a day evaporates, lost to use.
The most economical thermal power plants from a water standpoint use cooling towers. That is also the most polluting cooling method, because algicides and fungicides are used to keep the towers operating efficiently. Solar photovoltaics and wind turbines use almost no water. One of the great, unappreciated benefits of shutting down thermal power plants (fossil fuel and nuclear) is that a vast amount of water will be liberated for other uses by the transition to solar and wind. The value of this will increase (skyrocket at some point in some places?) as water becomes a more and more contested resource. Recall Mark Twain: “Whiskey is for drinking; water is for fighting over.”
By any reasonable economic and technical standard, nuclear power should be phased out in an orderly way if we want to have a reliable, resilient, and affordable energy system. New nuclear is a plutonium-producing economic lemon. Existing nuclear is on its way to becoming a technical dinosaur in an age where the very term “baseload” plant on which it hangs its hat is headed to a museum of electrical lexicography. Utility-scale solar and wind are roughly four times cheaper than new nuclear – plenty of economic room to fix the variability for when the wind doesn’t blow, and the sun doesn’t shine. Rather than subsidizing aging, plutonium-producing, water-gobbling reactors, it would be far better to phase out nuclear and let Mother Nature stream her energy through the grid complemented by technology that works with her rhythms to keep the lights on.
Coming: (1) Keeping the lights on with solar and wind. (2) A creative niche for green hydrogen in a clean energy system
Opening photo credit: Byron Nuclear Generating Station (source: Bill Tracey)
 I’m leaving the discussion of “carbon capture and sequestration” for another day.
 MJB&A, Joint Proposal for the Orderly Replacement of Diablo Canyon Power Plant with Energy Efficiency and Renewables, June 21, 2016, p. 3; italics added.