Science

Nuclear Energy

The one large-scale low-carbon power source that terrifies people — and whether the fear has been proportionate to the evidence.

The short version

Nuclear power generates approximately 10% of the world's electricity with near-zero carbon emissions, making it among the largest sources of low-carbon electricity alongside hydropower.
By deaths per unit of energy produced, nuclear is statistically among the safest energy sources ever deployed — safer than coal, oil, natural gas, and comparable to or safer than solar and wind.
The three major accidents — Three Mile Island (1979), Chernobyl (1986), and Fukushima (2011) — produced vastly different outcomes and are often conflated in public risk perception.
The case against nuclear is primarily economic: new plants in Western countries cost 3–5 times original estimates and take a decade to build, while solar and wind costs have fallen 90% since 2010, changing the competitive landscape entirely.

What it is

Nuclear power generates electricity through nuclear fission — the splitting of heavy atomic nuclei (primarily uranium-235 or plutonium-239) to release energy. When a neutron strikes a fissile nucleus, the nucleus splits, releasing energy as heat and emitting additional neutrons that trigger further fissions in a controlled chain reaction. This heat is used to produce steam, which drives turbines — a thermodynamic process identical to coal or gas plants, but fueled by the energy of atomic binding forces rather than chemical combustion. The result is a power source that produces roughly 1 million times more energy per unit of fuel mass than fossil fuels, generates negligible direct carbon emissions during operation, and runs continuously at high capacity factors (typically 90%+ availability) regardless of weather. Commercial nuclear power has been operating since the 1950s and currently supplies approximately 10% of global electricity from approximately 440 operating reactors across 30 countries.

The nuclear fuel cycle involves mining and enriching uranium, fabricating fuel rods, operating reactors, and managing spent fuel — each stage with distinct environmental and safety profiles. Uranium mining produces radioactive tailings that require long-term containment. Reactor operation generates spent nuclear fuel: highly radioactive material that must be stored in water-filled spent fuel pools initially, then eventually in dry cask storage or permanent geological repositories. The absence of a permanent repository for high-level nuclear waste is a genuine unresolved problem in every country with a nuclear program. The United States' Yucca Mountain repository was licensed and funded, then abandoned for political reasons in 2010. Spent fuel currently sits in temporary storage at reactor sites across the country — a situation that is not immediately dangerous but is not a solution.

The history of nuclear accidents has shaped public perception more than the statistical risk profile justifies. Three Mile Island (1979) was a serious mechanical failure and partial core meltdown that released small amounts of radioactive gases; the NRC's own health studies found no measurable increase in cancer rates among the surrounding population. Chernobyl (1986) was categorically different: a reactor design with fundamental safety flaws operated by a crew in violation of procedures during a test that caused an explosion and fire without a containment structure — a scenario impossible with Western reactor designs. The WHO's 2005 assessment attributed approximately 4,000 eventual cancer deaths to Chernobyl radiation exposure; the direct death toll was 31. Fukushima (2011), triggered by a tsunami that overwhelmed backup cooling systems, caused one radiation-related death and no measurable increase in cancer rates in the surrounding population, according to the United Nations Scientific Committee on the Effects of Atomic Radiation. The 2,200 deaths from the Fukushima disaster were caused by the evacuation itself — stress, exposure, and disruption to medical care — not radiation.

When the statistical record of deaths per unit of energy produced is compiled across energy sources, nuclear power's safety profile is consistently among the best. A 2021 study in the Lancet reviewing decades of data found that coal causes approximately 24.6 deaths per terawatt-hour of electricity produced; oil, 18.4; natural gas, 2.8; solar, 0.02; wind, 0.04; and nuclear, 0.07 — comparable to wind and solar, and orders of magnitude safer than fossil fuels. These numbers include Chernobyl. The public perception inversion — that nuclear is uniquely dangerous while coal plants, which emit mercury, particulates, and radioactive ash continuously, are routine — reflects the psychology of dread risk rather than the actuarial record.

Why it matters

Nuclear's relevance to climate change is simple in principle: it is a large-scale, dispatchable, low-carbon electricity source at a moment when the world needs to decarbonize its electricity supply rapidly while maintaining grid reliability. Wind and solar are intermittent — they generate electricity when the wind blows and the sun shines, which doesn't always coincide with demand. Nuclear runs continuously, providing the baseload power that grids need to balance variable renewables. Countries that have closed nuclear plants — Germany most prominently — have seen their electricity become more carbon-intensive in the short term, not less, as nuclear was replaced partly by natural gas and coal imports. France generates roughly 70% of its electricity from nuclear power and has among the lowest per-capita electricity-sector carbon emissions in Europe. The physics of the grid make nuclear harder to simply replace than its opponents anticipated.

The case against new nuclear construction is primarily economic, and it has grown stronger as renewable costs have collapsed. New nuclear plants in the West have been economic disasters. The Vogtle Plant in Georgia — the first new U.S. nuclear plant in decades — was licensed in 2012 at an estimated cost of $14 billion. It came online in 2024 at an actual cost exceeding $35 billion, nearly a decade late. Similar cost overruns have plagued projects in Finland, France, and the United Kingdom. The root cause is the loss of the skilled workforce and supply chains that enabled the construction of the previous generation of plants in the 1960s–70s; rebuilding those capabilities requires a learning curve that costs money. Meanwhile, utility-scale solar electricity costs have fallen approximately 90% since 2010, and wind costs have fallen similarly. The competitive arithmetic has shifted dramatically against new conventional nuclear construction.

Advanced nuclear designs — small modular reactors (SMRs), molten salt reactors, and next-generation designs that use spent fuel or thorium — are being developed and promoted as solutions to both the cost and waste problems of conventional nuclear. SMRs are designed for factory manufacture and modular deployment rather than custom construction, theoretically capturing the cost benefits of standardization. As of 2025, the first commercial SMRs are beginning operation in China; Western SMR projects remain in earlier development stages. Fusion power — which would produce vastly more energy than fission with far less waste and no proliferation risk — continues to make genuine technical progress, with the National Ignition Facility achieving fusion ignition in December 2022, but commercial fusion remains decades away under the most optimistic projections.

The nuclear debate has become a proxy for a broader cultural conflict about risk, technology, and the nature of environmental values. Opposition to nuclear has been a core position of the environmental movement since the 1970s, rooted in legitimate concerns about safety, waste, proliferation risk, and the entanglement of civilian and military nuclear programs. The emergence of climate change as the defining environmental challenge of the 21st century has fractured this consensus: some environmentalists — including prominent figures like Stewart Brand, James Lovelock, and more recently a minority of climate scientists — argue that closing existing nuclear plants in the name of environmental values is a climate mistake. Others maintain that the economics, waste, and proliferation arguments have not changed and that the renewable alternative makes nuclear unnecessary. This is a genuine empirical dispute with legitimate arguments on multiple sides — which makes it an unusual subject for a debate that often generates more heat than the plants themselves.

Nuclear Energy

The short version

What it is

Why it matters

Sources & Further Reading