It was 10 October 1465 – the day of the hotly anticipated wedding of King Alfonso II of Naples. He was set to marry the sophisticated Ippolita Maria Sforza, a noblewoman from Milan, in a lavish ceremony. As she entered the city, the crowds gasped. Before them was a sight so strange and beautiful, it was like nothing they had ever seen before.
Alas, they weren’t staring at the bride to be – they were looking up at the sky. Though it was the middle of the day, the Sun had turned a deep azure, plunging the city into eerie darkness. Rumours began to spread – was it a solar eclipse? As the early dusk lingered on, others suggested it could be a consequence of the weather. After all, they’d had a particularly wet autumn and some claimed they had seen a thick, humid fog rise up into the sky.
This was just the beginning. In the months that followed, European weather went haywire. In Germany, it rained so heavily that corpses surfaced in cemeteries. In the town of Thorn, Poland, the inhabitants took to travelling the streets by boat. In the unrelenting rain, the castle cellars of Teutonic knights were flooded and whole villages were swept away.
Four years later, Europe was hit by a mini ice age. Fish froze in their ponds. Trees failed to blossom and grass didn’t grow. In Bologna, Italy, heavy snow forced locals to travel with their horses and carriages along the frozen waterways.
In fact, what Alfonso’s wedding party witnessed may have been more extraordinary than anyone imagined. Many thousands of miles away in the tropics, a giant volcano was making geological history. This was an eruption so big, it produced an ash cloud which enveloped the Earth and led to the coolest decade for centuries.
The blast itself would have been heard up to 2,000km (1,242 miles) away and created a tsunami which caused devastation hundreds of kilometres away. In terms of scale, it surpassed even the 1815 eruption of Tambora, which unleashed energy equivalent to 2.2 million Little Boy atomic bombs and killed at least 70,000 people. Traces of the eruption have been found from Antarctica to Greenland.
The thing is, scientists can’t find the volcano that did it. What’s going on?
This is a true geological mystery, which has left geologists scratching their heads for decades
That the ‘unknown eruption’ happened is undisputed – like most mega-eruptions, it vapourised vast quantities of sulphur-rich rock, which was blasted into the atmosphere and eventually snowed down on the poles as sulphuric acid. There it was locked into the ice, forming part of a natural record of geological activity that spans millennia. There’s no other event capable of doing this, short of an asteroid impact.
But establishing its existence is the easy part. What scientists don’t know is – well, pretty much everything else. This is a true geological mystery, one which has left geologists scratching their heads for decades.
It all began with a rumour and a coral-fringed island in the South Pacific. Back in the 1950s, archaeologists visiting Tongoa, Vanuatu heard stories of an ancient landmass which, many generations ago, linked it to the neighbouring island of Epi. It was known as Kuwae – and in the centre was a giant volcano.
Excluding obvious embellishments, such as a local man’s supernatural wrath over being tricked into an incestuous relationship with his mother, the legend boils down to this. One day, after several strong earthquakes, a cataclysmic eruption broke the island in half.
Many people escaped, fleeing to nearby islands by boat. Most of those who stayed perished, but one of those who survived was the young Ti Tongoa Liseiriki, who resettled Tongoa after the eruptions ceased and passed the story down the generations.
All that is left of the volcano today is a crater about half a mile deep (nearly a kilometre), hidden beneath the ocean – the Kuwae caldera – and a thick layer of ash on Epi and Tongoa, created by flows of superheated gas and rock which surged over the island at hundreds of miles per hour.
Scientists didn’t know about the 15th-Century eruption until the 1980s, when they discovered a spike of acidity from around that time in cores taken from polar ice. Suddenly, for those investigating Kuwae, it looked like they were onto something.
Early estimates of the eruption’s date were based on the number of tribal chiefs who had reigned since it happened and placed it at between 1540 and 1654 AD. Archaeologists even dated Liseiriki’s skeleton, which turned out to be from the 14th or 15th Century. So far, so good.
Pang figured he would find out exactly when Kuwae erupted by looking for a period of global cooling
Soon evidence began emerging from other sources, too. The languages from islands surrounding the crater all appeared to have common linguistic roots, suggesting they may have evolved from the same dialect many hundreds of years ago.
Then in 1993, a scientist from Nasa’s Jet Propulsion Laboratory took a look. While they’re still in the atmosphere, sulphate aerosols reflect sunlight back towards space. The dimming effect of eruptions can be so powerful, artificial ‘volcanoes’ have been suggested as a means of combatting climate change – so Pang figured he would find out exactly when Kuwae erupted by looking for a period of global cooling.
He searched far and wide for clues, combing ancient records for famines and even scrutinising the rings on oak-framed British portraits to look for stunted growth. Pang concluded that the eruption at Kuwae had occurred in 1453 – too early to disrupt Alfonso’s wedding, but at the right time to coincide with another particularly lethal year.
In Sweden, crops failed and grain stores emptied. Across Europe, trees stopped growing. In China, tens of thousands of people froze to death. Months after, it snowed non-stop for 40 days below the Yangtze River (the same latitude as Northern Mexico) and the Yellow Sea froze up to 20 km (13 miles) out from the shore. On the other side of the world, the Aztecs were faced with the greatest famine experienced in pre-history.
Pang was so confident in his research, he even suggested an exact date for the eruption – 22 May, when it had interrupted a great battle raging at the city of Constantinople. The great “fire” that had been set by Turkish attackers was really just the reflection of ash in the twilight.
The final clincher appeared to come when a team of French geologists visited the island to take a closer look. Based on the size of the crater, they estimated that Kuwae’s eruption had released vast quantities of magma, enough to fill the Empire State Building 37 million times over, and shot debris 30 miles (48 km) into the sky. This would have included three times more sulphate than the eruption at Tambora – more than enough to have a global impact on the climate.
Crucially, by radiocarbon dating trees which perished during the eruption, the team put its date at between 1420 and 1430. Though this doesn’t match the spike in ice cores perfectly, it was deemed close enough.
There’s no doubt that there are volcanic deposits, but their extent isn’t what you’d expect from a truly massive eruption – Karoly Nemeth, environmental scientist
Scientists began to refer to the 15th-Century eruption as the “Kuwae” event – the theory was gathering momentum. Alas, the clarity didn’t last. Barely more than a decade after the French study on Kuwae’s eruption, a team of scientists returned to the islands surrounding the crater.
Led by Karoly Nemeth, an environmental scientist from Massey University, New Zealand, they searched for signs of an explosive eruption capable of altering the global climate. Spoiler alert: they didn’t find any.
Truly gigantic eruptions have columns at least 25km (15.5 miles) tall – which allows them to inject sulphur directly into the upper atmosphere and means they tend to spread debris over a wide area. To find out how spectacular Kuwae’s was, all they had to do was look for the fallout. “There’s no doubt that there are volcanic deposits, but their extent isn’t what you’d expect from a truly massive eruption,” he says.
In fact the evidence points to the volcano having been relatively small, no more than 1,000m high, which is less than a quarter of the height of Tambora before it erupted. Nemeth concluded that instead of breaking apart in a single, massive burst, the volcano actually erupted several times, on a much smaller scale. It was a devastating blow, or at least you would think it was, but the scientific community didn’t really take any notice.
Then in 2012 an ice core from Antarctica revealed a big surprise. The sample was from a site known as Law Dome, which is famous for its unusually high snowfall. It’s a favourite spot for climate research, since the thick ice makes it easier to distinguish individual, annual layers. The team constructed the most accurate record of every major volcanic eruption for the last 2,000 years yet.
They found that the 15th-Century spike in the record wasn’t caused by a single eruption after all, but two. Besides, the earliest of these was in 1458 – much later than the eruption at Kuwae. The authors urged others to be cautious in making the link. The riddle of the missing volcano wasn’t over yet. A year later, another study confirmed their results.
How could scientists have got it so wrong?
In the end, the only evidence linking the three is circumstantial
It all comes down to the way ice cores are dated. They can’t provide an exact date, just a sequence of yearly events, one after the other. To estimate when an event – such as the ‘unknown eruption’ – occurred, the frozen record is calibrated with historical information, such as the accounts of climate chaos in 1453.
“It’s this kind of incredible circular argument,” says Nemeth. The eruption was thought to have happened in 1453, because that’s when the planet cooled down, which is also why the eruption at Kuwae was thought to have happened then… and so on.
In the end, the only evidence linking the three is circumstantial. “It all hinges on a misinterpretation of historical information,” says Martin Bauch, a historian from the University of Leipzig in Germany, who investigated the evidence for a global cooling event in 1465.
When evidence for the ‘unknown eruption’ was discovered in ice cores, people were expecting to find the site, so when the eruption at Kuwae was confirmed, so those researching Kuwae simply connected the dots. “It was all an accident,” says Nemeth.
But if the eruptions didn’t happen at Kuwae, where did they happen?
In order to have a truly global impact, it’s likely that the eruptions happened in the tropics. That’s because above the tropics, rising air can pull the volcanic cloud higher into the atmosphere where it lingers for many years. This debris is also likely to spread out over a wider area, since high winds tend to pull air from the equator to the poles.
“If you look on the map that’s a huge area, we’re talking about essentially many hundreds of volcanoes in the Pacific region and so on,” says Nemeth. Other scientists have suggested looking to the island arcs, including remoter parts of Indonesia, Melanesia, Polynesia, and Micronesia.
Since Tambora (which is responsible for an equivalently sized eruption) was 4,300m high before it erupted, they were probably on the large size. For obvious reasons, the culprits have probably long vanished underwater. The mystery of the 15th Century eruptions isn’t over yet.
And what of the cold snap in 1465? As it happens, this is just five years off one estimated eruption date (1460). It’s hardly conclusive proof, but this is within the expected margin of error. Who knows, in the end, the interrupting volcano may give Alfonso’s wedding the fame it deserves.
Join 800,000+ Future fans by liking us on Facebook, or follow us on Twitter.
If you liked this story, sign up for the weekly bbc.com features newsletter, called “If You Only Read 6 Things This Week”. A handpicked selection of stories from BBC Future, Earth, Culture, Capital, and Travel, delivered to your inbox every Friday.