Archaeologists in Norway have discovered dozens of arrows—some dating back 6,000 years—melting out of a 60-acre ice patch in the county’s high mountains.
Expeditions to survey the Langfonne ice patch in 2014 and 2016, both particularly warm summers, also revealed copious reindeer bones and antlers, suggesting that hunters used the ice patch over the course of millennia. Their hunting technique stayed the same even as the weapons they used evolved from stone and river shell arrowheads to iron points.
Now the research team is revealing the finds in a paper published today in the journal Holocene. A record-setting total of 68 complete and partial arrows (and five arrowheads) were ultimately discovered by the team on and around the melting ice patch–more than archaeologists have recovered from any other frozen site in the world. Some of the projectiles date to the Neolithic period while the most “recent” finds are from the 14th century A.D.
While the sheer number of historical projectiles is stunning, the Langfonne discoveries are also upending generally accepted ideas in the relatively new specialty of ice-patch archaeology, and yielding new clues as to ice’s potential to preserve or destroy evidence from the past over the course of thousands of years.
An icy ‘time machine’?
Since archaeologists started systematically surveying melting ice sites 15 years ago, ice patches from Norway to North America have yielded almost perfectly-preserved artifacts from long-ago time periods. In isolation, the individual finds contain information about craftsmanship and long-ago hunting traditions.
Langfonne, in fact, was one of the first ice patch sites to come to light, after a local hiker discovered a 3,300-year-old leather shoe sitting next to the edge of the ice patch in the summer of 2006 and reported it to archaeologist Lars Pilø, now a researcher at the Innlandet County Council Cultural Heritage Department and a co-author of the new study.
Ever since that discovery alerted Pilø to the possibility of artifacts preserved in mountain ice patches, researchers in Norway and beyond—there are similar sites in Canada’s Yukon, the Rockies in the U.S. and the Alps in Europe—have wondered if the distribution of objects on and around the ice might tell them about how and when the ice patch sites were used and how they grew over time.
Unlike glaciers, which are essentially slow-moving frozen rivers, ice patches are fixed deposits of snow and ice that may grow and shrink over time. Sites like Langfonne, researchers assumed, resemble a patch of snow at the end of winter: As temperatures increase, artifacts trapped inside melt out in the order they were deposited.
“The idea was, ice is like a time machine. Anything that lands on it stays there and is protected,” Pilø says.
That meant the oldest items would be found in the deepest core of the ice patch, in the same way that archaeologists working with artifacts buried in soil assume lower layers of dirt contain older artifacts. And because the ice patches were thought to grow steadily with each winter’s snowfall, more recent finds would be closer to the edges of the patch.
If ice patches froze artifacts exactly where they were lost, archaeologists theorized, those items could help reconstruct what people did there in the past, how big the ice patches were at specific points in prehistory, and how fast they grew and shrank over time.
The Langfonne arrows seemed like a way to test the time-machine theory.
The arrows and reindeer bones confirmed earlier suspicions that Norway’s high mountain ice patches were reindeer-hunting hotspots: When the cold-loving creatures retreated to the ice to avoid biting insects during the summer months, people followed with bows, arrows, and hunting knives.
But after radiocarbon dating all the arrows and gathering dozens more dates from reindeer remains they found on the ice, the researchers realized that, at Langfonne at least, the time-machine theory was unreliable. Researchers expected that the oldest items would be trapped in place from the day they were lost and preserved just as well as artifacts buried in the ice in later centuries. But the oldest artifacts at Langfonne, which date back to the Neolithic, were fragmented and heavily weathered, as though they’d been churned by the ice or exposed to sun and wind for years.
Arrows from later periods, like the 1,500-year-old arrow that used a sharpened mussel shell harvested from a river at least 50 miles away, looked as though they were shot just yesterday. “That raises the suspicion something happened inside the ice” that exposed and re-froze the older items, Pilø says.
And the arrows didn’t seem to be emerging in any particular order, as you’d expect if the ice formed perfect layers over time. Arrows made thousands of years apart were lying not far from each other along the ice edge. “The idea that you find the oldest evidence when the ice patch is at its smallest—that isn’t really true,” says Montana State Parks archaeologist Rachel Reckin, who was not part of the research team. “It looks like gravity and water are moving artifacts down a great deal.”
Co-author Atle Nesje, a glaciologist at the University of Bergen, says that thousands of years ago, warm summers probably exposed older artifacts, which were carried to the edge of the ice patch by streams of meltwater before freezing again. The weight of ice pressing down on lower layers might have caused them to shift, carrying their frozen contents with them. Or lightweight wooden arrow shafts might have been blown across the surface by fierce winds before getting lodged in rocks or getting covered again by snow. Arrows lost in the snow more recently, meanwhile, might have stayed in place.
Because old arrows might be washed down by meltwater and then re-freeze, the spot where they were found could be a long way from where they originally landed. That meant using radiocarbon dated arrows to map the size of the ice patch in the past was a dead end. “Glaciologists and ice patch archaeologists were hoping that artifacts could give us an idea of the size over time, but that’s not the case,” Reckin says.
Wolverines and Vikings
Researchers were pleasantly surprised that the Langfonne arrows, once dated, could provide useful clues to how people used the ice patch over time. During certain periods, for instance, the team found lots of reindeer bones but very few arrows. That suggests people weren’t hunting on the ice; instead, reindeer were probably being killed by wolverines, which bury their carcasses in the snow to eat later.
Between A.D. 600 and 1300—roughly the Viking Age—radiocarbon dating revealed a different type of activity on the Langfonne patch. “There’s a lot of arrow finds, but hardly any reindeer material,” Pilø says. “That’s not a coincidence.” Humans were hard at work removing slain reindeer from the ice, harvesting their fur and antlers to sell as trade goods.
The rapidly changing understanding of the ice and the secrets it holds matches the speed at which the ice is disappearing. “I’ve been studying Norwegian glaciers for the last 40 years. It’s a lot of change,” says Nesje. “It’s quite scary to see how fast the ice patches can melt away, from one day to another.”
Based on lichen growth on the rocks around the ice patch, Nesje estimates that Langfonne today is half the size it was in the late 1990s—and a tenth of its extent during the Little Ice Age, a centuries-long dip in global temperatures that lasted from about 1300 AD into the 1800s.
The steady melting means archaeologists have to move fast while preserving as much information as possible. “Time is of the essence, and we’re trying to be good scientists while doing the best we can with the data we have,” Reckin says. “Every piece of this puzzle that helps us understand the complexity of these processes is really helpful.”