The Huu-ay-aht people kept the memory alive for more than three hundred years. Their storytellers described a winter evening when the shaking began after the village had gone to sleep. "It threw down their houses and brought great masses of rock down from the mountains," one account relates. Then the ocean followed. "I think it was at nighttime that the land shook. I think a big wave smashed into the beach. The Pachena Bay people were lost." The accounts are specific about one detail: nobody in the village survived. A young woman named Anacla aq sop, who happened to be staying at Kiix?in on the more sheltered Barkley Sound, became the last living member of her community. Her name would eventually be given to the village that the Huu-ay-aht later built at Pachena Bay -- Anacla.

Half a world away, Japanese scribes recorded what they could not explain. Around midnight on January 27, 1700, a tsunami struck Kuwagasaki, destroying thirteen houses and setting off fires that burned twenty more. It caused a shipwreck and damaged rice stores. There had been no earthquake in Japan. For nearly three centuries, this remained an orphan wave -- a tsunami without a parent. In 1996, a team of researchers linked the Japanese records with the geological evidence accumulating along the Pacific Northwest coast: layers of sand interrupting marsh sediments, fragments of wood preserved in soil, drowned forests. Using wave heights and arrival times from the Japanese accounts, scientists worked backward and concluded the tsunami originated from a magnitude 8.7 to 9.2 earthquake off Vancouver Island. The date, the time, the season -- everything matched the Huu-ay-aht oral tradition.

Dig into the marshes around Pachena Bay and the evidence is literally layered. Coastal marsh plants grow, accumulate, and are suddenly buried under a blanket of sand washed ashore by a tsunami. The marsh slowly rebuilds. Another layer of plant material grows. Then another layer of sand appears. The deeper scientists excavate, the more of these cycles they find, each one a record of a past megathrust earthquake. These geological signatures, combined with the oral histories and the Japanese written records, allowed researchers to establish that massive subduction earthquakes have struck this coast every 250 to 850 years. The convergence of three independent lines of evidence -- Indigenous oral tradition, Japanese historical documents, and geological stratigraphy -- represents one of the most compelling examples of interdisciplinary earthquake science anywhere in the world.

Today, the Huu-ay-aht First Nations village of Anacla sits at Pachena Bay, 13 kilometers south of Bamfield within Pacific Rim National Park. Residents still live primarily in lower-lying areas near the shore, but they evacuate to higher ground at the administration building when tsunami warnings are issued. More than three centuries have passed since the last great Cascadia rupture -- well within the 250-to-850-year recurrence interval that the geological record reveals. The bay remains beautiful, sheltered enough for a community, exposed enough to remind anyone who knows the history that the seafoam in the name could return with devastating force. The difference now is that the knowledge held by Huu-ay-aht storytellers for generations has been validated, quantified, and woven into the scientific understanding of one of the planet's most dangerous fault zones.