A volcanic winter is a period of global cooling that follows a large eruption, caused not by ash but by sulfur flung into the stratosphere. It has darkened skies, failed harvests and toppled empires — the year 536 ADwas so cold and dim it gets called the worst year in history. Here's the science, and a ranked tour of the eruptions that chilled the planet, each tied to the VEI data in our eruption database.
By VolcanoDB Research Team. Data: Smithsonian Global Volcanism Program (VEI grades across 11,079 eruptions); cooling estimates from published ice-core and proxy studies, cited inline.
Worst year to be alive
536 AD
VEI 7 eruptions in our DB
7
VEI 6 eruptions in our DB
52
Pinatubo 1991 global cooling
~0.5°C
What is a volcanic winter?
A volcanic winter is a temporary drop in global temperatures caused by a large volcanic eruption. The eruption injects sulfur dioxide into the stratosphere, where it forms a haze of sulfuric-acid aerosols that reflects sunlight back to space. The result is one to three years of cooler, dimmer conditions worldwide — enough to cause crop failures, famine and, in the worst cases, historical collapse. The biggest confirmed example, the 536 AD event, dimmed the sun for 18 months and ushered in the coldest decade in more than 2,000 years.
Why Sulfur, Not Ash, Freezes the Planet
The intuitive villain is ash — a black plume blotting out the sun. But volcanic ash is heavy and falls out of the sky within days to weeks, close to the volcano. It causes local darkness, not a global winter. The real driver sits one layer up.
A powerful explosive eruption punches gas past the weather layer and into the stratosphere, 15–50 km up. The key gas is sulfur dioxide. Once there, SO₂ combines with water vapour to form microscopic droplets of sulfuric acid — a reflective aerosol veil that stratospheric winds smear around the entire globe within months. That veil bounces a fraction of incoming sunlight back to space, so less energy reaches the surface. Temperatures fall. Because the stratosphere has no rain to wash the droplets out, the veil lingers for one to three years before gravity finally clears it.
This is why two eruptions of the same size can have wildly different climate effects. What matters is how much sulfur reaches the stratosphere — not how much lava or ash lands on the ground. A relatively modest-looking eruption rich in sulfur can out-cool a bigger, drier one.
1. Eruption
SO₂ blasted 15–50 km into the stratosphere
2. Aerosol veil
SO₂ → sulfuric-acid droplets circle the globe
3. Cooling
Sunlight reflected; surface cools 1–3 years
536 AD: The Worst Year in History
The Byzantine historian Procopius wrote that "the sun gave forth its light without brightness, like the moon, during the whole year." He wasn't exaggerating. Beginning in 536 AD, a dry fog spread across Europe, the Middle East and Asia, dimming the sun for around 18 months. Tree rings record the coldest decade in more than 2,000 years; snow fell in China in summer; crops failed from Ireland to Mesopotamia. Medieval historian Michael McCormick has called 536 the beginning of "the worst period to be alive."
The cause was volcanic — in fact a one-two punch. Ice-core sulfur spikes show a major eruption in 536 (increasingly attributed to Iceland), followed by a second large eruption in 540. Together they kept the Northern Hemisphere locked in cold for a full decade, and the resulting famine and social stress set the stage for the Justinian plague that swept the empire from 541. It is the clearest case in the historical record of volcanism reshaping human events.
The Volcanic Winters That Changed History, Ranked
No one has pulled these into a single ranked list with the VEI data attached — so here it is. Each eruption below appears in our database; the cooling figures come from published ice-core and proxy studies. They are ordered roughly by climatic severity.
The largest eruption of the last 2 million years. Long tied to a human population bottleneck — though 2026 research argues people adapted rather than nearly went extinct.
The 536 event
536–537 AD
6–7
~1.5–2.5°C (N. Hemisphere)
Often called the worst year to be alive. A dry fog dimmed the sun for 18 months; crops failed from Ireland to China. Followed by the Justinian plague.
The single greatest stratospheric sulfur release of the Common Era. Helped tip the Northern Hemisphere toward the Little Ice Age; medieval famine and mass burials followed in Europe.
Huaynaputina
1600 AD
6
~0.8°C (N. Hemisphere)
The largest eruption in South American history. Triggered the Russian famine of 1601–1603, which killed an estimated two million people.
Laki
1783–84 AD
4 (flood basalt)
~1°C (N. Hemisphere)
An eight-month fissure eruption that released roughly 120 million tonnes of SO₂. Killed about a fifth of Iceland's population and spread a toxic 'Laki haze' and hard winter across Europe.
Produced 1816, the 'Year Without a Summer' — snow in June in New England, failed harvests across Europe, and the wet, dark holiday where Mary Shelley wrote Frankenstein.
Pinatubo
1991 AD
6
~0.5°C (global, 2 yrs)
The modern benchmark. Injected ~20 million tonnes of SO₂; satellites tracked global temperatures dropping ~0.5°C for two years. The clearest instrumented volcanic winter on record.
VEI grades from Smithsonian GVP records (Samalas is catalogued under the Rinjani complex; Laki under the Grímsvötn system). Cooling estimates are best-available figures from proxy and ice-core research and carry real uncertainty, especially for older events.
The VEI Threshold: How Rare Are These Eruptions?
The Volcanic Explosivity Index runs from 0 to 8, each step representing roughly a ten-fold jump in erupted volume. Climate history lives at the top of the scale. Across the 11,079 eruptions catalogued in our database:
Only 7 reach VEI 7— the Tambora / Samalas class, capable of catastrophic multi-year cooling. That's roughly one every 1,500–2,000 years.
52 reach VEI 6 — the Pinatubo / Krakatau class, enough for a measurable ~0.5°C global dip lasting a couple of years.
181 reach VEI 5 — the Mount St. Helens class, which can cool a region but rarely the whole planet.
No confirmed VEI 8 super-eruption has occurred in recorded human history — the last, Toba, was around 74,000 years ago. That rarity is worth holding onto whenever a headline warns that a supervolcano is "overdue." The events that cause the deepest volcanic winters are, thankfully, the rarest eruptions on Earth.
Could Yellowstone Cause a Volcanic Winter?
It's the question that drives most searches on this topic, so here's the straight answer: physically yes, realistically no time soon. A full VEI 8 eruption of the Yellowstone caldera would loft enough sulfur to cause years of global cooling, crop failure and hemispheric ashfall — a volcanic winter beyond anything in written history. But the USGS estimates the annual probability of such an eruption at about 1 in 730,000, and current monitoring shows no indication of an impending eruption of any size. Yellowstone's earthquakes and ground movement are normal behaviour for a large, well-behaved caldera.
Would We See the Next One Coming?
Better than any generation before us. Satellites now measure stratospheric SO₂ directly — the instrumented record of Pinatubo in 1991 is the reason modern climate models handle volcanic forcing so well. Global networks track ground deformation and earthquake swarms around the world's highest-threat systems, and the nine Volcanic Ash Advisory Centres monitor eruption plumes for aviation in real time. We can't stop a volcanic winter, but for the first time we would very likely watch the sulfur climb into the stratosphere as it happened — and know what was coming.
Frequently Asked Questions
What causes a volcanic winter?
It's not the ash — it's the sulfur. A large eruption blasts sulfur dioxide (SO₂) high into the stratosphere, where it reacts with water to form a haze of tiny sulfuric-acid droplets. That aerosol veil spreads around the globe and reflects incoming sunlight back to space, cooling the Earth's surface for one to three years until the droplets settle out. Fine ash falls out within weeks and matters little; the stratospheric sulfur is what drives a volcanic winter.
How long does a volcanic winter last?
Typically one to three years. The sulfate aerosols that cause the cooling gradually fall out of the stratosphere over about 1–2 years, so temperatures recover fairly quickly — Pinatubo's ~0.5°C dip in 1991 had largely faded by 1994. The exceptions are the giants: the 536 AD event was reinforced by a second eruption in 540, extending the cold for a full decade, and the very largest eruptions can nudge the climate system toward a longer cold phase like the Little Ice Age.
How big does an eruption need to be to cool the climate?
Measurable global cooling generally requires VEI 6 or larger — the scale of Pinatubo 1991 or Krakatau 1883. Catastrophic, history-altering cooling takes VEI 7 (Tambora, Samalas) or the extremely rare VEI 8 (Toba, Yellowstone's past eruptions). These are genuinely rare: of the 11,079 eruptions in our database, only 7 reach VEI 7 and just 52 reach VEI 6. It's less about size on the ground than how much sulfur reaches the stratosphere.
Could a volcano cause an ice age?
Not a true ice age on its own, but a large eruption can help push an already-cooling climate over a threshold. The 1257 Samalas eruption and a cluster of 13th–15th century eruptions are widely linked to the onset of the Little Ice Age, a several-hundred-year cold period. A single eruption's direct cooling lasts only a few years, but if it lands during a susceptible period it can trigger feedbacks — expanding sea ice and snow cover — that lock in cold for much longer.
Could Yellowstone cause a volcanic winter?
In theory, yes — a full VEI 8 Yellowstone eruption would inject enough sulfur to cause years of global cooling and crop failure. But the USGS puts the annual probability of such an eruption at roughly 1 in 730,000, and there is no sign one is brewing. A Yellowstone super-eruption is one of the least likely natural catastrophes you could worry about, even though it would be among the most severe if it ever happened.