Nuclear Winter: The Science and the Silence

There is a particular kind of dread that settles over you when you read about nuclear winter for the first time and realise that the most devastating consequence of a nuclear war would not be the blast, not the radiation, not the immediate death toll — but a slow, dark, frozen silence that descends over the entire planet for years afterward. I first wrote about this back in 2011, when the topic felt theoretical and Cold-War-distant. Returning to it in 2026, with nuclear arsenals still very much in play and geopolitical tensions running at their highest since the 1980s, it does not feel theoretical at all.

This is everything I know about nuclear winter — every scientific fact I have gathered, every model I have studied, every implication I have followed to its uncomfortable end. I am not writing this to frighten you. I am writing it because this is one of those subjects where understanding the stakes is, itself, a form of protection.

What Is Nuclear Winter, Actually?

Nuclear winter is a hypothesised severe and prolonged cooling of the Earth's climate that scientists believe would follow a large-scale nuclear war. The word "hypothesised" is doing some work there — not because the basic mechanism is in doubt, but because no one has ever run the full experiment and lived to publish the results. What we have instead are increasingly sophisticated climate models, firestorm case studies, and satellite observations of large wildfires that together paint a picture that is, to put it plainly, catastrophic.

The core mechanism is this: nuclear weapons detonated over cities would ignite enormous firestorms. Those firestorms would inject massive quantities of black carbon soot — not dust, not ash, but specifically the fine, dark, light-absorbing particles produced by burning petroleum products, plastics, and synthetic materials — directly into the stratosphere. Once there, that soot would spread around the globe within weeks, blocking incoming solar radiation and causing surface temperatures to plummet. Crops would fail. Ecosystems would collapse. A famine of planetary scale would follow.

It is worth pausing on the word "stratosphere." Most weather happens in the troposphere, the lowest layer of the atmosphere, where rain, wind, and turbulence naturally clean out particulates within days to weeks. The stratosphere is above all of that. There is no rain up there to wash soot out. Once black carbon reaches stratospheric altitudes — roughly 12 kilometres or higher — it stays there, circulating globally, for months to years.

The 1983 TTAPS Study: Where the Term Was Born

The concept of nuclear winter was not always called nuclear winter. In the 1970s and early 1980s, researchers were already exploring the atmospheric effects of large-scale nuclear conflict, but the conversation was dominated by immediate blast effects, radiation, and electromagnetic pulse. The climate angle was secondary — until December 1983, when a paper published in the journal Science changed everything.

The paper, titled "Nuclear Winter: Global Consequences of Multiple Nuclear Explosions," was authored by five scientists: Richard P. Turco, Owen B. Toon, Thomas P. Ackerman, James B. Pollack, and Carl Sagan. Their initials gave the study its name: TTAPS. It was Richard Turco who coined the term "nuclear winter" itself — a two-word phrase that compressed the forcing mechanism (soot) and the response (cooling) into something that the public and policymakers could actually hold in their minds.

The TTAPS team ran a one-dimensional climate model through various nuclear exchange scenarios between the United States and the Soviet Union, ranging from a few hundred megatons to full-scale thermonuclear war. Their finding: the smoke and soot from fires ignited by nuclear detonations — particularly from burning urban materials like petroleum fuels and plastics, which absorb sunlight far more efficiently than wood smoke — would remain suspended in the stratosphere for months. The result would be what they called a "nuclear twilight," a dramatic reduction in sunlight reaching the Earth's surface, leading to subfreezing temperatures across large portions of the Northern Hemisphere even in summer.

"Nuclear winter" — two words that captured a civilisation-ending consequence nobody had thought to model until then. — Richard P. Turco, TTAPS team, 1983

The TTAPS study was not published in isolation. Simultaneously, the Soviet Academy of Sciences produced its own independent modelling, reaching broadly similar conclusions. The convergence of American and Soviet scientific findings gave the hypothesis unusual credibility across Cold War lines. Carl Sagan, the most publicly visible of the TTAPS authors, used his platform to bring nuclear winter into mainstream awareness — making it a genuine factor in the arms control debate of the 1980s.

The Mechanics of a Firestorm

To understand nuclear winter, you need to understand what a firestorm actually is — because it is the firestorm, not the nuclear explosion itself, that does the atmospheric damage. A nuclear detonation over an urban area would ignite simultaneous fires across a vast area — potentially hundreds of square kilometres. Under certain conditions, those fires merge into a single self-sustaining conflagration with its own weather system: a firestorm.

Firestorms are categorically different from ordinary large fires. The heat they generate creates powerful updrafts — convective columns of superheated air rising at hurricane force. These updrafts act like chimneys, drawing in fresh oxygen from the surrounding area (which is why firestorm victims sometimes suffocated in air raid shelters even without fire touching them) and, crucially, lofting soot and smoke far higher than ordinary fires can manage. A large enough firestorm can inject combustion products directly into the lower stratosphere.

We have real-world case studies. The Allied firebombing of Hamburg in 1943 and Dresden in 1945 produced genuine firestorms. The atomic bombing of Hiroshima in August 1945 ignited a firestorm that destroyed approximately 13 square kilometres of the city. More recently, the 2021 British Columbia wildfires produced pyrocumulonimbus clouds — the atmospheric structures associated with fire-generated storm systems — that injected smoke measurably into the stratosphere. Scientists studying these events have been able to verify and refine their models of how smoke behaves at high altitude.

The crucial insight from the TTAPS study — one that earlier models of nuclear war's atmospheric effects had missed — was the importance of what is burning. Cities are not forests. Modern urban environments are packed with petroleum products, synthetic polymers, and plastics that, when burned, produce a disproportionately large quantity of black carbon per unit of fuel. Black carbon particles are tiny — typically smaller than a micrometre — and dark. They absorb visible light with extraordinary efficiency. The TTAPS team recognised that the smoke from a nuclear-devastated city was qualitatively different, and far more climatically potent, than the smoke from burning wood or grassland.

How Much Soot Are We Talking About?

The numbers involved in nuclear winter scenarios are hard to visualise without a frame of reference, so let me try to give you one. Mount Pinatubo's 1991 eruption — one of the largest volcanic events of the twentieth century — injected roughly 20 million tonnes of sulphur dioxide into the stratosphere, causing a global average temperature drop of about 0.5°C for two years. That volcanic winter was considered a significant climate event. Nuclear winter, even in moderate scenarios, dwarfs it.

Those figures come from models by Alan Robock and colleagues at Rutgers University, who have been the most active researchers in this field over the past two decades. The 5 Tg scenario from a regional India-Pakistan exchange is particularly sobering — not because the scale of conflict is extreme, but because it is so plausible. Both nations have growing arsenals, a history of conflict, and contested territory. A war that would claim perhaps 20–100 million lives directly would also trigger, via the soot pathway, what researchers project could be the deaths of 1–2 billion people from famine globally.

The Stratosphere Does Not Forget

One of the most counterintuitive aspects of nuclear winter is that the stratospheric soot does not behave the way most people expect atmospheric pollution to behave. In the troposphere, where we live and breathe, particles are rained out within days to weeks. The stratosphere has no such cleaning mechanism. Once soot reaches stratospheric altitudes, it is effectively locked in for months to years.

There is an additional wrinkle that makes the situation even worse: the soot heats the surrounding air as it absorbs solar radiation. This creates a self-lofting effect — the heated soot rises higher and higher into the stratosphere, moving it further from the troposphere where it might eventually be removed. Computer models have consistently shown that the initial soot injection actually rises significantly higher over the weeks following a nuclear exchange. Once it is up there, global circulation patterns spread it latitudinally within a matter of weeks, making it a global phenomenon, not a regional one.

What Nuclear Winter Does to the Planet: A Chain Reaction of Catastrophes

Temperature Drop

In a full-scale nuclear exchange scenario, climate models project midsummer temperature drops of 10–20°C averaged across Northern Hemisphere agricultural regions, with local extremes reaching 35°C below normal in continental interiors like Russia. Even in the India-Pakistan 5 Tg scenario, global average surface temperatures would drop by roughly 1–2°C — enough to eliminate anthropogenic warming gains of the last several decades and meaningfully disrupt growing seasons across the world's breadbasket regions.

Crop Failure and Nuclear Famine

This is where the death toll projections become staggering, and where most mainstream discussions of nuclear war fall short. The direct casualties of a nuclear exchange — horrific as they are — would be dwarfed by the downstream famine. Alan Robock and colleagues modelled global food production under various soot injection scenarios and found that injection of more than 5 Tg of soot into the stratosphere would lead to mass food shortages persisting for several years, with livestock and aquatic food production unable to compensate for reduced crop output in almost all countries.

For the 37 Tg scenario — roughly corresponding to a large regional exchange — projected direct deaths from famine amount to 1–2 billion people, assuming stored food supplies are consumed within months and international food trade collapses. This projection assumes that Southern Hemisphere nations like Argentina, Australia, and New Zealand, which are large food exporters, maintain enough local production to feed their own populations — but that their surpluses would not reach a world with collapsed trade infrastructure.

In 2025, researchers at Pennsylvania State University used a crop-specific agroecosystem model to simulate how nuclear winter would affect global corn yields across nearly 40,000 locations worldwide. Their findings reinforced the severity: even moderate soot injections produce yield reductions that cascade into multi-year famine conditions across both hemispheres.

Ozone Depletion and UV Surge

Temperature drop and crop failure are not the only consequences. A 2021 study published in the Journal of Geophysical Research by Charles Bardeen, Douglas Kinnison, and colleagues found that the fires of a large nuclear exchange would produce nitrogen oxides in enormous quantities. At stratospheric temperatures elevated by soot absorption, these nitrogen oxides would react to deplete the ozone layer at rates far exceeding normal. The resulting surge in ultraviolet radiation reaching the surface would add a further assault on already-stressed agricultural systems and ecosystems — injuring plants, killing phytoplankton, and increasing cancer risks for surviving human populations for years after the initial exchange.

Ocean Acidification

The oceans would not escape. Researchers modelling the oceanic response to nuclear conflict found that nuclear war has the potential to increase surface ocean pH and decrease aragonite saturation — the measure of how easily marine organisms can form calcium carbonate shells. This would exacerbate the ocean acidification already underway from anthropogenic carbon dioxide emissions, potentially collapsing shellfish populations and disrupting the marine food web at its base.

The "Nuclear Niño" Effect

Climate modelling has also identified a curious ocean-atmosphere coupling response that researchers have termed the "Nuclear Niño." The differential cooling between continental interiors and ocean surfaces would alter large-scale atmospheric circulation patterns in ways that mimic — but dramatically amplify — the effects of El Niño. Monsoon systems in South and Southeast Asia, already complex and fragile, would be severely disrupted. The coastal storms predicted in earlier nuclear winter models — where cold continental air masses collide with relatively warmer oceanic air — are part of this broader circulation disruption.

The Kuwait Oil Fires: A Cautionary Tale in Modelling

No discussion of nuclear winter is complete without acknowledging the controversy that nearly killed the field in the 1990s. When Saddam Hussein ordered the burning of 736 Kuwaiti oil wells at the end of the 1991 Gulf War, some nuclear winter researchers — notably Carl Sagan — predicted that the fires could produce regional cooling effects comparable to nuclear winter scenarios. They did not. Local temperatures dropped, but the global atmospheric impact was minimal. Critics seized on this as evidence that nuclear winter models were overblown.

The criticism, while pointed, misunderstood the key distinction: the Kuwait fires produced significant smoke, but the fires were not large enough, hot enough, or concentrated enough to create firestorm conditions. Without firestorm dynamics, smoke rises only into the troposphere, where rain scavenges it within weeks. The key variable in nuclear winter is not simply the volume of fire — it is whether firestorm conditions develop, which drives soot into the stratosphere. The Kuwait experience was, if anything, an illustration of exactly why firestorms matter so much to the model.

The Science in 2024 and 2026: Where We Stand

After a dormant period in the 1990s following the Cold War's end, nuclear winter research experienced a significant revival in the 2000s, driven by growing concerns about the nuclear arsenals of India and Pakistan. A generation of more powerful climate models — including the NASA GISS ModelE and the Whole Atmosphere Community Climate Model (WACCM) — allowed researchers to run three-dimensional atmospheric simulations of far greater sophistication than the TTAPS era.

The broad conclusion of this modern research: the original TTAPS findings were not wrong. If anything, the newer models, incorporating more realistic ocean-atmosphere coupling, full stratospheric chemistry, and better soot lofting dynamics, project consequences that are broadly consistent with or in some respects more severe than the 1983 estimates.

There remains genuine scientific debate about key variables. The Rutgers University team (Robock, Toon, and colleagues) and the Los Alamos National Laboratory team disagree significantly about the probability of firestorm formation in a regional nuclear exchange and about what fraction of generated soot would actually reach the stratosphere versus being scavenged in the lower atmosphere. Lawrence Livermore National Laboratory, in its own modelling, occupies a middle ground, noting strong nonlinearities in the soot dynamics — meaning that small differences in fire intensity or combustion conditions produce large differences in the final soot load at stratospheric altitudes.

Since 2023, the US National Academies of Sciences, Engineering, and Medicine has been conducting an Independent Study on the Potential Environmental Effects of Nuclear War, with a comprehensive report in preparation. The committee was still working as of early 2025, reflecting the complexity and genuine scientific uncertainty that remains in parts of the problem — but none of that uncertainty addresses whether the basic mechanism is real. It is.

Natural Analogues: Volcanic Winters and the K-T Event

Nuclear winter does not exist without precedent in nature — though the precedents are either insufficient in scale or deep in geological time. Volcanic eruptions inject sulphur dioxide and ash into the stratosphere, producing "volcanic winters" of measurable but limited duration. The 1815 eruption of Tambora in Indonesia caused the "Year Without a Summer" in 1816 — crop failures across the Northern Hemisphere, food riots in Europe, and weather so cold and gloomy in Switzerland that Mary Shelley, stuck indoors, wrote Frankenstein. Tambora was a catastrophic eruption, yet it produced a global temperature drop of perhaps 0.5–1°C for two to three years.

The reason volcanic winters are less severe than nuclear winter projections lies in the nature of the aerosols. Volcanic eruptions primarily inject sulphur dioxide, which forms reflective sulphate aerosols. These are relatively bright particles — they scatter sunlight back into space, which cools the surface, but they do not absorb it in the way black carbon does. Soot absorbs light. It is one of the most efficient light-absorbers in the known atmosphere. A given mass of black carbon has enormously greater climate-forcing potential than a comparable mass of volcanic sulphate.

Go back 66 million years and you find the closest natural analogue: the Chicxulub asteroid impact that ended the Cretaceous period. The impact vaporised enormous quantities of carbonaceous rock, injected debris and soot globally, halted photosynthesis, and drove the extinction of approximately 75% of species on Earth. Owen B. Toon and Alan Robock, in their 2025 book Earth in Flames, draw this parallel explicitly — not to say nuclear war would cause the same scale of extinction, but to situate nuclear winter within the history of catastrophic atmospheric events that the Earth has actually experienced.

The Policy Dimension: Nuclear Winter and the Arms Race

One of the genuinely remarkable things about nuclear winter science is that it apparently worked. Alan Robock and colleagues published a 2023 paper in Atmospheric Chemistry and Physics arguing that awareness of nuclear winter was a significant factor in ending the Cold War arms race. They cite the example of Soviet General Secretary Gorbachev, who in a 1985 interview referenced nuclear winter as a key reason for his commitment to nuclear abolition — and whose subsequent arms control agreements with the Reagan administration began the dramatic reduction of the global arsenal from a peak of roughly 70,000 weapons in 1986 to around 12,000 today.

The 2017 Nobel Peace Prize went to the International Campaign to Abolish Nuclear Weapons (ICAN), whose evidence base explicitly incorporated nuclear winter science. The Treaty on the Prohibition of Nuclear Weapons, which entered into force in 2021, draws on the humanitarian and environmental consequences of nuclear weapons — consequences that nuclear winter research made legible to policymakers.

Despite this, the roughly 12,000 nuclear weapons that remain — approximately 4,000 of them deployed and ready for use — are still more than sufficient to trigger the most severe nuclear winter scenarios. The Doomsday Clock of the Bulletin of the Atomic Scientists stood at 90 seconds to midnight in 2023, the closest it has ever been, precisely because the geopolitical environment has deteriorated while the arsenals remain.

Where Would You Even Go?

This is the question people always ask when they encounter nuclear winter for the first time, and it deserves an honest answer. Climate models are consistent on one point: Southern Hemisphere locations would fare considerably better than Northern Hemisphere ones in most scenarios. Argentina, Australia, and New Zealand experience less severe cooling, being surrounded by ocean masses that buffer temperature change and being far from the likely conflict zones. These are also large food-exporting nations that produce more calories than their populations require.

But here is the brutal qualifier: the same models that identify these relative safe zones also project that international food trade would collapse following a nuclear exchange. The food surplus of Argentina or New Zealand would not reach a population in a different continent dealing with failed infrastructure, destroyed trade routes, and societal breakdown. The benefit of being in a safer location depends almost entirely on whether the political and logistical infrastructure required to distribute food survives — which, in a full-scale nuclear exchange scenario, it would not.

Survival planning — bunkers, long-term food stores, underground agriculture, aquaponics, UV-protective clothing — is technically possible to think through, but it scales very badly beyond small groups. The realistic conclusion of nuclear winter science is not "here is how to survive" but "here is why we must prevent this from happening."

The Connection to Other Nuclear Events

Nuclear winter is specifically about the atmospheric consequences of nuclear war — it is distinct from the effects of nuclear accidents like Chernobyl or Fukushima, which I have written about elsewhere on this blog. Those events involved catastrophic releases of radioactive material into the local and regional environment, but they did not involve the firestorm-scale soot injection that drives nuclear winter. The atmospheric signature of nuclear accidents is chemically and physically different from the atmospheric signature of nuclear warfare. I mention this because the two are sometimes conflated, and the distinction matters for understanding both types of hazard clearly.

If anything, thinking about nuclear winter alongside nuclear accident risk underscores a broader point: the decision to maintain and potentially use nuclear weapons is not merely a military or strategic decision. It is a decision about the planetary atmosphere, the global food system, and the survival prospects of people who have no say in the matter whatsoever — people in countries that possess no nuclear weapons, living on continents far from any conflict.

A Final Word

Carl Sagan once said that if we do not destroy ourselves, we will one day venture to the stars. Nuclear winter is the scenario in which we do destroy ourselves — not in the fireball, but in the long, cold, dark years that follow. The science is clear. The mechanism is real. The question is whether we find it compelling enough to act on.

I hope we do.