The Summer of the Dry Fog: How Iceland’s Laki Eruption Poisoned a Nation and Darkened Europe

On the morning of Sunday, June 8, 1783, the people of the Síða district in southern Iceland were getting ready for Whitsunday services when a black cloud rose over the highlands to the north. Within hours it had spread across the sky, and fine black ash was falling on the farms. The ground shook. That night, from behind the mountains, came a red glow that did not fade. A crack in the earth more than 25 kilometres long had begun to open, and for the next eight months it would pour out lava, ash and gas on a scale that no one alive had ever seen.

The eruption is known in Iceland as the Skaftáreldar, the Skaftá Fires, after the river whose gorge the lava filled. Outside Iceland it is usually called Laki, after the small mountain the fissure cut through. It killed no one directly. There was no great explosion and no wall of fire racing toward a town. What it did instead was quieter and far worse. It poisoned the grass that Iceland’s sheep and cattle lived on, and the famine that followed killed about one person in six on the island. Its gas drifted over Europe as a stinking, sun-dimming haze that people called the dry fog. And it gave Benjamin Franklin, then living outside Paris, an idea that was about two centuries ahead of its time.

When this site told the story of the Year Without a Summer, the famine year of 1816 that followed the eruption of Tambora, Franklin’s 1783 guess about an Icelandic volcano got a single sentence. It deserves more than that. Laki was a different kind of eruption from Tambora, it hurt people in different ways, and of all the great eruptions in the historical record it is the one volcanologists worry most about seeing again.

A fissure opens on Whitsunday

Iceland sits on the Mid-Atlantic Ridge, the seam where the North American and Eurasian plates are slowly pulling apart. In much of the island that stretching happens along long cracks rather than at tall, cone-shaped mountains. Magma rises into a crack, and when it reaches the surface it comes out along the whole length at once, as a line of fire fountains. Geologists call this a fissure eruption. Laki belongs to the volcanic system of Grímsvötn, a volcano hidden under the Vatnajökull ice cap that is the most frequently active in Iceland, and Grímsvötn itself erupted on and off from 1783 to 1785.

The Laki fissure did not open all at once. Studies of its craters and of the eyewitness record show that it came in about ten separate episodes, each starting with a short explosive phase as rising magma met groundwater and then settling into a long period of fountaining. New segments kept opening toward the northeast. In the end the line held more than 130 craters. Lava fountains rose an estimated 800 to 1,400 metres into the air, and the heat drove the gas higher still, to roughly 10 to 15 kilometres, into the upper atmosphere.

The volumes are hard to picture. Over eight months the fissure produced about 14 cubic kilometres of basalt lava, one of the largest lava outpourings anywhere on Earth in the last thousand years. The flows ran down the gorge of the Skaftá, a canyon said to have been up to 200 metres deep, filled it, and spilled out onto the lowlands, swallowing farms. A second great flow went down the valley of the Hverfisfljót to the east. Together they covered roughly 600 square kilometres. Most of the lava came out in the first five months, and the eruption finally stopped on February 7, 1784.

On the Volcanic Explosivity Index, the scale usually used to compare eruptions, Laki rates only a 4, well below Tambora’s 7. That index measures how much rock an eruption blasts into the sky, and Laki was not especially explosive. Its weapon was gas. The magma was rich in sulfur and fluorine, and the fissure let it all out: an estimated 120 million tonnes of sulfur dioxide and about 8 million tonnes of fluorine compounds. That is roughly six times the sulfur that Mount Pinatubo put into the atmosphere in 1991.

Today the crater row is part of Vatnajökull National Park. The cones are coated in thick grey-green woolly moss, which grows so well on young lava in Iceland’s damp climate that a visitor can walk among the craters of one of history’s deadliest eruptions without seeing much bare rock at all. Getting there takes a four-wheel-drive vehicle and several unbridged river crossings.

The fire priest

Most of what we know about the first months of the eruption comes from one man. Jón Steingrímsson (1728–1791) was the pastor at Kirkjubæjarklaustur, a church farm on the Skaftá river in the lowlands below the highlands, and the dean of the district. He kept careful notes from the first day, and in 1788 he finished a detailed account of the eruption and its aftermath. It is still one of the best eyewitness descriptions of a volcanic disaster ever written, and modern volcanologists have used it to date the separate episodes of the eruption almost day by day.

His descriptions of what the gas did to the land are hard to read even now. In one passage he wrote:

This past week, and the two prior to it, more poison fell from the sky than words can describe: ash, volcanic hairs, rain full of sulfur and saltpeter, all of it mixed with sand. The snouts, nostrils, and feet of livestock grazing or walking on the grass turned bright yellow and raw.

The “volcanic hairs” were Pele’s hair, thin threads of volcanic glass spun out of the fountains by the wind. The yellow, raw skin on the animals was the work of acid. Water in the streams turned tepid and bluish, and the plants withered and went grey.

By July the main lava flow had filled the gorge and was moving down toward Kirkjubæjarklaustur itself. On Sunday, July 20, 1783, Jón gathered his parishioners in the church while the lava advanced, and preached what became famous as the Eldmessa, the Fire Mass. When the service ended, the story goes, the people walked out to find that the flow had stopped. The lava front halted short of the church, at a spot still called Eldmessutangi, Fire Mass Point. Many of his parishioners believed that his prayers had saved them. Jón, who had been watching the flows closely for weeks, may well have known that the ground there favored them. Either way, the sermon became one of the best-known episodes in Icelandic history.

In 1974 the district consecrated a memorial chapel at Kirkjubæjarklaustur in his honor, built a little east of the old church site. He is buried in the old churchyard there. The lava did not take the farm, but the worst was still to come for the people who had heard him preach.

The mist hardships

In Icelandic the years after the eruption are called the Móðuharðindin, the mist hardships, after the haze. To understand why they were so deadly, it helps to know how Iceland lived in the 1780s. It was a poor province of the Danish crown, with a population of about 50,000 people scattered on farms along the coast and in the valleys. Most families lived from sheep and cattle, and hay was the main crop. Milk and other dairy products supplied about half of the calories people ate.

The eruption hit that system at its weakest point. The fluorine coated the grass, and the animals that ate it developed fluorosis: their teeth grew lumps and wore down unevenly until they could not chew, and their bones grew swellings and broke easily. Acid rain and the haze stunted the hay crop over much of the country. Through the following winter the animals starved or died of poisoning. By most estimates about 80 percent of Iceland’s sheep died, along with about half of its cattle and half of its horses. Milk yields fell by half.

Then the people began to die. Iceland’s population was 49,609 at the start of 1783 and 40,381 at the start of 1786. Most of the deaths came in 1784 and 1785, when about 11,000 people died in a country where an ordinary year saw about 1,500 deaths. That is roughly 8,000 deaths above normal, about one-sixth of the people alive before the eruption, and births dropped too. Some relief came from Denmark in the summer of 1784, but it was modest, and some regions fared better than others depending on whether people could turn to the sea for food.

For a long time historians blamed the deaths on hunger and the diseases that follow hunger. From the 1970s on, some researchers suggested that the haze itself, and fluorine in food and water, might have poisoned people directly. A 2024 study by Claudia Wieners of Utrecht University and Guðmundur Hálfdanarson of the University of Iceland went back to parish burial registers and found that the old explanation holds up. Where and when people died matched where food ran out, not where the haze was thickest, and the symptoms the pastors recorded looked more like starvation and scurvy than fluorine poisoning. Laki killed Icelanders mainly by killing their animals.

A dry fog over Europe

While Iceland suffered, the gas went abroad. A rare high-pressure system sat over Iceland that summer and steered the plume southeast. The haze reached Bergen in Norway within days, then Prague by June 17, Berlin by June 18, Paris by June 20, Le Havre by June 22 and Britain by June 23. It was nothing like ordinary fog. It did not lift in the sun or blow away in the wind. It smelled of sulfur. Ships stayed in port because sailors could not see, and the sun rose and set the color of blood.

In the village of Selborne in Hampshire, the naturalist and curate Gilbert White, who had kept a daily nature journal for years, wrote one of the most quoted descriptions of that summer in a letter later collected in his famous Natural History of Selborne:

The summer of the year 1783 was an amazing and portentous one, and full of horrible phaenomena… the peculiar haze, or smokey fog, that prevailed for many weeks in this island, and in every part of Europe, and even beyond its limits, was a most extraordinary appearance, unlike anything known within the memory of man… The sun, at noon, looked as blank as a clouded moon, and shed a rust-coloured ferruginous light on the ground, and floors of rooms; but was particularly lurid and blood-coloured at rising and setting.

White noted the haze in his journal from June 23 to July 20. The heat was so intense, he wrote, that butchers’ meat could hardly be eaten the day after the animal was killed, and the flies drove the horses half frantic. Country people looked at the red sun “with a superstitious awe.” On the morning the haze arrived in Suffolk, Sir John Cullum wrote to Joseph Banks, the president of the Royal Society, that a tenant had found the ground white with what looked like frost in the middle of June, and that the tips of the barley and the leaves of oats and young conifers had turned brown and withered. Modern researchers think much of that damage was done by the acid in the haze.

The summer of 1783 turned into one of the hottest on record in England, with violent thunderstorms and hail reportedly large enough to kill cattle. And people began to feel ill. Letters and newspapers across Europe reported headaches, breathing trouble and eye irritation. In France, the death rate in Chartres rose noticeably in August and September.

Counting the dead in England

Did the haze kill people in Britain? In 2004 Claire Witham and Clive Oppenheimer of the University of Cambridge went through the burial records of 404 parishes in 39 English counties to find out. They found two clear spikes in deaths. The first came in August and September 1783, with about 11,500 more deaths than normal, around 40 percent above the average for those months. The second came in January and February 1784, with about 8,200 more. Together that is about 20,000 extra deaths, and in both spikes the east of England was hit hardest.

The winter spike, they concluded, could be explained by cold alone: January 1784 was brutal. The summer spike was harder to explain. A very hot July can raise deaths a month or two later through gut infections and fevers, but by their estimate the heat accounted for only about 30 percent of the late-summer excess. That left room for the haze, and they suggested that fine acid droplets in the volcanic cloud may well have contributed, much as coal smoke trapped under a temperature inversion killed thousands during London’s Great Smog of 1952.

Not everyone is convinced. The 2024 Iceland study pointed out that the deaths in England and France did not rise until August, even though the haze had been thick since late June, and that crisis parishes and normal ones often sat side by side. The honest answer is that the dry fog probably made a bad summer deadlier, but no one can say by exactly how much.

To people at the time, 1783 felt like a year when the whole natural order had come loose. Earthquakes had devastated Calabria in southern Italy in February. Mount Asama in Japan erupted that summer. Off the Reykjanes peninsula in southwest Iceland, a submarine eruption had built a small new island earlier in 1783, which the sea soon wore away. Gilbert White heard of it as “a volcano sprung out of the sea on the coast of Norway.” And on the night of August 18, a brilliant fireball crossed the sky over Britain from north to south, splitting into pieces as it went.

The painter Paul Sandby pictured the scene from the terrace of Windsor Castle, where the Italian natural philosopher Tiberius Cavallo was among the watchers. Cavallo wrote up the meteor for the Royal Society’s Philosophical Transactions, describing a light so strong that the whole landscape below the terrace was lit up. To many in Britain, the fireball was one more strange sign in an already frightening summer, and it would turn up again, in a very different role, in Franklin’s thinking.

Franklin’s conjecture

Benjamin Franklin spent the summer of 1783 at Passy, then a village just outside Paris, where he was the American minister to France. He was 77, and he was busy. The peace treaty that ended the American Revolutionary War was signed in Paris that September, and in the same months he went to see the first balloon flights. It was also a lively time for scientific showmanship in the city: the chess-playing automaton known as the Mechanical Turk was on its European tour that year, and played games in Paris.

Franklin had been interested in weather all his life, and the fog caught his attention. In May 1784 he wrote a short essay called “Meteorological Imaginations and Conjectures.” He offered it to the physician Thomas Percival in Manchester, who read it to the Literary and Philosophical Society of Manchester on December 22, 1784. It was printed in the society’s Memoirs the next year. In it Franklin described the summer in plain words:

During several of the summer months of the year 1783, when the effect of the sun’s rays to heat the earth in these northern regions should have been greatest, there existed a constant fog over all Europe, and great part of North America. This fog was of a permanent nature; it was dry, and the rays of the sun seemed to have little effect towards dissipating it… They were indeed rendered so faint in passing through it, that when collected in the focus of a burning glass, they would scarce kindle brown paper.

That detail about the burning glass suggests he had tried it himself. Franklin then built a chain of reasoning: because the sun’s rays were weakened, the ground stayed cooler, so the first snows did not melt, so the snow reflected still more sunlight and chilled the winds. “Hence perhaps,” he wrote, “the winter of 1783-4 was more severe than any that had happened for many years.”

He offered two possible causes for the fog. One was that the Earth had passed through the smoke of “some of those great burning balls or globes which we happen to meet with in our rapid course round the sun,” which “are sometimes seen to kindle and be destroyed in passing our atmosphere.” That is, meteors, and the great fireball of August 18 was fresh in everyone’s mind. The other was “the vast quantity of smoke, long continuing to issue during the summer from Hecla in Iceland, and that other volcano which arose out of the sea near that island.”

He had the right country and the wrong mountain. Hekla, the most famous volcano in Iceland, had not erupted since 1766, and European newspapers sometimes reported any Icelandic eruption under its name. The real source was Laki, about 70 kilometres to the east. But the core idea was right: a volcano’s emissions could spread over half the world and cool it. Both of Franklin’s guesses, in fact, describe things that we now know can change climate, and a researcher who reprinted the essay in 2010 called it the first time that either volcanoes or meteors had been put forward as a cause of climate change.

Franklin closed with a suggestion that reads like a research program. It seemed “worth the inquiry,” he wrote, “whether other hard winters, recorded in history, were preceded by similar permanent and widely extended summer fogs,” because if so, people might use such fogs to predict a hard winter and prepare for it. Almost nobody took him up on it. As the Year Without a Summer story describes, the link between volcanoes and climate did not become mainstream science until after Krakatoa in 1883, a century later.

The long winter

Franklin was right that the winter was severe. In Europe, January and February 1784 were among the coldest months on record, and when the deep snow melted that spring, floods tore through towns along the rivers of Germany and central Europe. In North America the winter of 1783–84 was one of the longest and coldest ever recorded. New England had its longest stretch of below-zero temperatures, Chesapeake Bay stayed frozen for a record time, and the Mississippi River froze at New Orleans, with ice floes reported in the Gulf of Mexico.

The cold even touched American politics. Congress was then meeting in Annapolis, on the Chesapeake, and it needed delegates from nine states to ratify the Treaty of Paris. Snow and ice kept many of them from arriving, and the treaty was not ratified until January 14, 1784, cutting close to the deadline for exchanging copies with Britain.

The effects reached farther than anyone in 1784 could have guessed. In 2006 a team led by Luke Oman and Alan Robock of Rutgers University ran climate models of the Laki cloud and found that by cooling the northern continents it could have weakened the African and Indian monsoons. That fits the records of an unusually low Nile flood in 1783 and a severe famine in Egypt the following year, and of drought in parts of India. Japan suffered the great Tenmei famine in the same years, though it had its own volcano, Asama, and years of poor weather before either eruption. Some historians have also pointed to the run of bad harvests in France in the 1780s, which helped drive bread prices up before the French Revolution of 1789. Laki was at most one ingredient in that story, but it was a real one.

Reading the haze

Modern estimates of what Laki released come largely from the work of the volcanologists Thorvaldur Thordarson and Stephen Self, who in 2003 combined the chemistry of the lava with the historical record. They estimated that the eruption emitted about 122 million tonnes of sulfur dioxide, most of it carried up to heights of 9 to 13 kilometres by the eruption columns. There it turned into tiny droplets of sulfuric acid, a veil of aerosol that hung over the Northern Hemisphere for more than five months.

The droplets eventually fell out of the sky, and some of them landed on the Greenland ice sheet, where they are still preserved. Laki leaves a clear sulfate spike in Greenland ice cores, the frozen archives that scientists began drilling at Camp Century in the 1960s. Spikes like it, matched to known eruptions such as Laki and Tambora, are how researchers now date older eruptions that nobody wrote down.

Laki also showed that eruptions can harm people in more than one way. Tambora’s 1815 blast threw its sulfur high into the stratosphere, where it spread around the whole planet and cooled it for a year or more. Laki’s eruption columns reached only the upper troposphere and the lowest stratosphere, and the weather that summer brought much of the gas back down to ground level as acid haze that people breathed. Roughly a fifth of the sulfur never rose high at all: it seeped out of the cooling lava fields and hung over Iceland itself. Volcanic gas does not need an explosion to kill, as the people around Cameroon’s Lake Nyos learned in 1986.

When Iceland’s fissures open again

Iceland’s fissures have not stopped opening. On August 29, 2014, after two weeks of earthquakes beneath the Bárðarbunga volcano, lava broke out at Holuhraun, a plain north of the Vatnajökull ice cap. The eruption ran until February 27, 2015, and produced about 1.5 cubic kilometres of lava, the largest eruption in Iceland in more than 200 years, which is to say since Laki.

Holuhraun was about a tenth the size of Laki, but it gave scientists a chance to watch a Laki-style eruption with modern instruments. A team led by Anja Schmidt, then at the University of Leeds, found that the fissure emitted as much as 120,000 tonnes of sulfur dioxide a day in early September 2014. On average its daily output was at least three times that of all human sources in Europe combined, and early on it was about eight times higher. The eruption released roughly 11 million tonnes of sulfur dioxide in all. The gas crossed the ocean at low altitude. Monitoring stations picked it up as far as 2,750 kilometres away, and on one September day a station at Ennis in western Ireland recorded about 524 micrograms per cubic metre of sulfur dioxide for an hour, a level Europe had not seen since 1990.

In 2011 Schmidt and her colleagues had already asked what a full Laki-scale eruption would do to today’s Europe. Using a global aerosol model and modern health statistics, they estimated that fine sulfate particles would roughly double across central, western and northern Europe during the first three months, and that the extra pollution could cause about 142,000 additional deaths from heart and lung disease, with a range of 52,000 to 228,000. That is likely more than a typical flu season kills. Since 2012, Britain’s National Risk Register has included a large, gas-rich Icelandic eruption of this kind among the hazards the country plans for.

Nobody can say when the next one will come. Iceland has had large fissure eruptions before, including Eldgjá in the 10th century, which was even bigger than Laki, and the Reykjanes peninsula near the capital has been erupting repeatedly since 2021, though on a tiny scale by comparison. What volcanologists can say is that a Laki-type eruption is not a once-in-history freak. It is something Iceland’s geology does, every few centuries. North America has its own record of giants, which this site covered in the five largest volcanic eruptions in North America, but none quite like this one: an eruption that barely exploded and still darkened a continent.

The historian Katrin Kleemann, whose 2023 book A Mist Connection traces how people across Europe experienced and explained the eruption, talks here about her research:

What the fog left behind

The Laki story ends up being about three different kinds of knowledge. There was the pastor in Kirkjubæjarklaustur, who watched the lava and the animals day after day and wrote down exactly what he saw, so that scientists two centuries later could reconstruct the eruption from his notes. There was the naturalist in Selborne, keeping a weather journal that turned a strange summer into dated evidence. And there was the old diplomat in Passy, who had no data from Iceland at all but put the pieces together, a dimmed sun, a dry fog, a hard winter, and asked whether the same thing had happened before.

All three were right about something, and none of them could have known how far the haze had reached, from the starving farms of southern Iceland to the Nile. It took ice cores, satellites, climate models and parish registers to see the whole shape of it. When the next great fissure opens, the instruments will be watching from the first day. Whether anyone is ready for the fog is another matter.

Further reading