A tsunami hit southwestern Norway 5,400 years ago

A tidal wave between 8 and 10 metres high may have swept across Stone Age communities.

Researchers now see a clear link between the Garth tsunami and destroyed Stone Age settlements.
Published

In the 1990s, archaeologist Håkon Glørstad began to suspect something unusual: One of the most well-known Stone Age settlements in Rogaland county, southwestern Norway, appeared to have been suddenly covered with sand. 

Could this have been caused by a sudden flood? Perhaps an unknown tsunami?

Today, younger researchers have picked up where Glørstad left off, investigating the possibility of a tsunami striking Rogaland at the end of the Stone Age. 

"We now believe Glørstad was very likely correct in his suspicions," archaeologist Svein Vatsvåg Nielsen tells sciencenorway.no.

A long-standing mystery

It has been known for a long time that around 5,400 years ago, the southwest coast of Norway experienced a dramatic rise in sea level.

But what actually happened?

For a long time, researchers believed this was due to a slowly rising sea, possibly caused by an unknown geological process. However, they never fully understood what was behind it.

Few have thought of a tsunami.

In an article 20 years ago, forskning.no reported that archaeologists had uncovered 17,000 artefacts at a settlement in Galta, in Rennesøy municipality. Strangely, the site had ended up underwater. Back then, speculation pointed to a massive landslide in the Boknafjord or another nearby fjord, triggering a wave.

"There's both published and unpublished material that contains speculation about a major flood in Rogaland around 5,400 years ago," says Svein Vatsvåg Nielsen.

Svein Vatsvåg Nielsen has investigated the tsunami hypothesis.

Stone Age archaeology and tsunami research

Now, Nielsen and his German colleague Martin Hinz have reviewed much of the material and combined it with the latest archaeological findings from Stone Age settlements in Rogaland.

They also incorporated tsunami research from Shetland. 

"We now consider it certain that the sea level rise was brief. It also seems likely that a tsunami struck the Stone Age people in Rogaland – and probably also Vest-Agder – sometime between 3445 and 3396 BCE," says Nielsen. 

Nielsen and Hinz suspect that it may have been the Garth tsunami, named after the Garth Loch lake in Shetland. More on that further down.

A settlement on the Tananger Peninsula

Jåsund 2 is the name archaeologists have given to one of the most well-known Stone Age settlements in Rogaland. It is located on the Tananger Peninsula in Sola municipality, just outside Stavanger. 

Evidence suggests Jåsund 2 was flooded by a wave possibly as high as 10 metres. Nielsen and his colleagues are now fairly confident in this conclusion. 

"This Stone Age settlement was discovered deep beneath a more recent settlement, which was examined at the same time. The findings suggest the older site was suddenly encapsulated by a layer of sand and gravel," says Nielsen, adding:

"The layers are so substantial that the entire site must have vanished from the landscape. It also doesn't appear that people ever returned."

While excavating the Jåsund 2 site in Sola municipality, archaeologists initially uncovered a settlement from the Bronze Age. But buried beneath a thick layer of sand and gravel, they discovered a second, much older settlement dating back to the Stone Age. The layer sequence can be seen in the bottom picture.

Stone Age twigs confirm the tsunami

Not far from Jåsund 2, archaeologists have made another important discovery.

At a nearby beach ridge, researchers uncovered twigs that appear to have been abruptly deposited there.

Radiocarbon dating of the twigs shows they were deposited at the exact time geologists believe the Garth tsunami struck Shetland.

Stone Age people lived near the coast

In Norway, Stone Age communities were almost always located near the sea or other bodies of water.

Due to land uplift after the Ice Age, archaeologists often find these settlements several metres above the current terrain.

The people living on the Tananger Peninsula 5,400 years ago had not yet adopted agriculture. The traces they left behind show that they lived by hunting, trapping, and fishing. Archaeologists also find no signs of the large longhouses that began appearing in Eastern Norway – along with agriculture – around the same time.

"The tsunami may have hit just a few hundred years before agriculture reached Rogaland," says Nielsen.

The colossal Storegga tsunami

The most well-known tsunami to strike Norway during human habitation was the Storegga tsunami, 8,200 years ago.

It was massive, sweeping across Norway, Scotland, Shetland, and the Faroe Islands.

We now know what caused that tsunami: An enormous underwater landslide the size of Iceland broke off the edge of the continental shelf off the Norwegian coast. Some rock masses, up to a kilometre in diameter, slid from depths of 250 metres all the way down to 3,000 metres below sea level.

The resulting wave, possibly more than 20 metres high, surged inland, devastating the hunter-gatherer communities along the coast of western Norway.

Have three tsunamis struck Norway?

Geologist Stein Bondevik has played a key role in documenting what may be three different tsunamis during the time humans have lived in Norway.

These likely occurred around 8,200, 5,400, and 1,500 years ago.

The oldest, the Storegga tsunami, is the most well-known. Bondevik and his colleagues have found traces of it on Shetland, more than 20 metres above the Stone Age sea level.

They have also discovered evidence of the Garth tsunami, which may have hit southwestern Norway 5,400 years ago, on Shetland. The sediments left behind by this tsunami are similar to those from the Storegga tsunami. In this case, the wave run-up on land is over 10 metres. 

Mass graves on Shetland and the Orkney Islands, dating to the time of the Garth tsunami, have also been linked to this disaster. But whether this tsunami was caused by an underwater landslide, an earthquake on the seabed, or possibly a meteorite – we simply don’t know.

Finally, Bondevik and his team have found evidence of a third, smaller tsunami that occurred about 1,500 years ago. This wave appears to have reached heights of 5-6 metres.

Did the Garth tsunami open the door to farming?

Svein Vatsvåg Nielsen wonders whether the Garth tsunami may have helped trigger the shift to agriculture in southwestern and western Norway.

The people who lived in the region before the tsunami were highly adapted to a marine environment. They fished extensively, hunted, and foraged for food like nuts and shellfish. At Jåsund 2, researchers have even found wild boar teeth.

The Garth tsunami likely wasn’t a complete disaster for the people living there in the Late Stone Age. But it left a lasting impression for many generations.

"We still have many unanswered questions about the Garth tsunami," Nielsen tells sciencenorway.no.

Exciting to see the hypothesis confirmed

Håkon Glørstad, who first proposed the tsunami theory in the 1990s, is pleased that Nielsen and Hinz have now confirmed it in a new scientific article.

"At the time, much of what we uncovered in archaeological digs in Rogaland didn’t make sense," Glørstad tells sciencenorway.no.

"But I realised something dramatic must have happened, something involving water, that was unique to Rogaland. It had to be a fairly local phenomenon," he says. 

Glørstad had also studied the stratigraphic layers during the excavation of settlements, including the distinct layers seen at Jåsund 2.

That’s when he began to suspect that a sudden flood had struck the area.

———

Translated by Alette Bjordal Gjellesvik

Read the Norwegian version of this article on forskning.no

References:

Bondevik et al. 'Evidence for three North Sea tsunamis at the Shetland Islands between 8000 and 1500 years ago', Quaternary Science Reviews, vol. 24, 2005. DOI: 10.1016/j.quascirev.2004.10.018 (Abstract)

Nielsen, S.V. & Hinz, M. 'Twigs date the second Holocene transgression in southwestern Norway', Radiocarbon, 2025. DOI: 10.1017/RDC.2025.14 (Abstract)

Powered by Labrador CMS