Human Origins / Archaeology

The strangest and most significant find, however, was a soil layer down by the beach that looked like nothing MacGillivray had ever seen in four decades as a field archaeologist. A horizontal band of gravel about a foot thick was stuffed with a mad jumble of broken pottery, rocks, lumps of powdery gray ash, and mashed-up animal teeth and bones. Perhaps an exceptionally violent storm had inflicted this chaos, MacGillivray considered, but he began to suspect that a tsunami was the more likely culprit.

The statue at Mallia may have been
smashed and burned during an
uprising against the Minoan elite.

Image courtesy of Evan Hadingham

MacGillivray invited Hendrik Bruins to Palaikastro. The Dutch-born geoarchaeologist and human ecologist had a reputation as a skillful analyst of the thorny dating controversies that beset archaeology in the Middle East, but figuring out the chaotic layer overlooking the beach presented a novel scientific challenge. “Identifying a tsunami deposit is a completely new field,” Bruins explains. “Until the early 1990s, earth scientists didn’t even recognize that tsunamis do more than just destroy the coast—they leave distinctive deposits behind as well. I needed to do a lot of different tests to convince myself, as well as my colleagues, that we were dealing with a tsunami and not something else, like debris from a storm surge.”

Another building looked like it had been flattened. Could a wave have done this?

Bruins sent thin sections of the chaotic deposit to micropaleontologist Chaim Benjamini, a colleague at Ben-Gurion University of the Negev in Israel. Benjamini identified the tiny round shells of foraminifera and fragments of red coralline ­algae; these marine organisms suggested that the ocean, rather than a river or a flash flood, had been involved. If the marine organisms had been scooped up from below sea level and dumped on the elevated promontory, something much bigger than a storm surge must have pounded the coast of ancient Crete.

The strange pattern of gravel deposits in the town offered further evidence of a deep oceanic disturbance. Then there were lumps of gray ash in the beach layer, “resembling unstirred instant-soup lumps at the bottom of a cup,” according to Bruins. He sent samples of these lumps to two state-of-the-art geochemistry labs in Germany, which analyzed the sample’s geochemical signature. The results of both tests were identical: a perfect match between Theran ash and the “soup lumps” on the beach.

Finally, there was the question of when all this disruption occurred. Bruins sent fragments of cattle bones and seashells from the chaotic layer to the radiocarbon dating lab at the University of Groningen in the Netherlands. Because of well-known problems in calibrating dates from 3,500 years ago, he knew the lab would be unable to pin down the exact calendar age of the samples, but the uncalibrated measured age of the cattle bones closely matched the latest equivalent dates for the cataclysm on Thera.

All the clues pointed to one answer: A giant wave had struck Palaikastro Bay while freshly fallen ash from Thera was still lying about, inundating the town for miles inland and streaking it with strange patterns of ash. But could even a giant wave be big enough to wipe out an entire civilization?

MacGillivray consulted Costas Synolakis, an energetic Greek-born earth scientist at the *University of Southern California, where he pioneered the predictive computer model used by the Pacific Tsunami Warning Center in Hawaii. Synolakis’s first attempts to model tsunamis in the early 1990s began as a solitary exercise. Everything changed after the Indian Ocean tsunami of 2004. Synolakis visited Banda Aceh, the city in northwestern Sumatra closest to the epicenter of the undersea quake, where hundred-foot waves had destroyed a city of more than 150,000 people in minutes. “It was a surreal, absurdist landscape,” he says. “It took an effort of imagination to conceive that people had ever lived there.” Almost overnight, Synolakis’s expertise in computer modeling of tsunamis became a focus of worldwide scientific and media attention.