Scientists probing beneath the Alaska seafloor have discovered a fault system that could pose huge risks for generating future tsunamis.
The discovery, led by scientists from Columbia University's Lamont-Doherty Earth Observatory, identifies a potential tsunami hazard at an area off the Alaska Peninsula known by geologists as the Shumagin Gap. The newly mapped fault system appears to be geologically similar to the system that produced the 2011 Japanese tsunami and the nuclear disaster at the Fukushima power plant, the scientists said.
The discovery is explained in a study published in the journal Nature Geosciences.
"Results from our study show that geological structures particularly prone to tsunami generation exist in this area and these structures should be considered in future hazard assessments for this region," lead author Anne Becel of the Lamont-Doherty Earth Observatory said in an email.
The hazard is not so much that the Shumagin Gap will have a big earthquake but that a quake in a neighboring area would generate a big tsunami along the features that they mapped, she said.
Both the Shumagin Gap and the offshore area that produced the Japanese tsunami are notable for characteristics that had previously suggested lowered risks for big earthquakes. They are both "creeping" subduction zones, areas where geologic plates move steadily and produce a lot of small quakes, not abrupt large ones. In the Shumagin Gap, geologic evidence shows that an oceanic plate is sliding beneath a continental plate.
Previous conventional wisdom held that creeping subduction zones, by releasing geologic tension gradually, pose less of a hazard than "fixed" subduction zones, where geologic plates strike against each other, only to suddenly give way, producing a large earthquake on the Earth's surface.
The Columbia-led mapping project adds to some other new information suggesting that creeping subduction zones are not as safe as once believed.
One of the surprises emerging from the Japanese earthquake and Tohoku tsunami disaster was its location — in a creeping subduction zone, the Columbia scientists said. There, a geologic fault was known to exist but the implications were not fully understood until the disaster. The fault had partially detached a portion of the tectonic plate and that portion fell during the quake, unleashing a tsunami that produced a tsunami that exceeded 130 feet in some places.
The new study of the Shumagin Gap details a similar fault condition off Alaska — and its authors caution that other places in the world might also have similar conditions.
The study uses data collected by scientists who sailed the area on a research vessel equipped with sonic devices to measure features beneath the seafloor.
The main fault they found, they said, stretches about 90 miles, running roughly parallel to land, and dips underground more than 20 miles to the point where two tectonic plates — the Pacific plate and the North American plate — overlap and move against each other. On the seafloor, they said, the fault is marked by steep underwater slopes about 15 feet high. They also found a "splay fault," a secondary branching fault, that slices into the top plate and creates a partial prism at its edge.
The Shumagin Gap, about 600 miles southwest of Anchorage, got its name decades ago because it is a place where there had been no "great" earthquakes, magnitude 8 or larger, in historic times, said Rob Witter, a U.S. Geological Survey scientist who has studied the area.
Before the "creeping" subduction theory took root, the thinking in past decades was that the Shumagin Gap was due for a large earthquake, said Witter, who was not involved in the mapping project that produced the new study. But even though smaller earthquakes have occurred in the area, a magnitude 8 quake has not yet been recorded there, he said.
The geologic mapping has yielded important information that should be considered in earthquake-preparedness planning, he said.
In that part of Alaska, "nobody's really used that high-tech mapping before," he said.
The main fault they found is considered a "normal" fault, "indicative of stretching or extension of the continental crusts," Witter said. He likened it to one of the cracks that would emerge on a stretched Snickers candy bar.
That and the splay fault indicate some potential trouble in the future, he said.
"If there were to be a great earthquake in the Shumagins, there's a high potential that there would be a big tsunami," he said.
Witter, though not part of the Columbia-led study, was part of a USGS project that showed potential hazards exist in a nearby area that also lies along the creeping subduction zone.
Witter and his USGS colleagues surveyed Sedanka Island, in the eastern Aleutians near Unalaska, and found evidence on land of six major tsunamis in the past 1,700 years. The tsunamis left ocean sands layered in the island's soil, as well as a big driftwood log found half a mile inland.
Those findings, published in 2016, complicated some earlier ideas about creeping subduction zones.
The Alaska subduction zone is a complex geologic feature. In the Shumagin Islands, it is clearly creeping, but elsewhere it runs through intermittently locked and creeping areas, Witter said. And the creeping feature is limited along certain places where the two plates overlap, he said. Northeast of the Shumagins, around Chirikof Island at the southwestern tip of the Kodiak archipelago, for example, the plates "appear to be locked solid," he said.
Much about the Shumagin Gap still puzzles scientists.
Witter and his colleagues have searched in vain for on-land signs of a destructive 1788 earthquake described by Russians of the time. That earthquake and resulting tsunami, as reported by the contemporary Russians, destroyed a village on Unga Island in the Shumagins. But searches of three places in the Shumagins have yet to turn up any sands or sediments that would have been deposited by a tsunami.
Results of those searches are detailed in a 2014 study.
It's possible that the 18th century Russian accounts were a little inaccurate, Witter said. There is evidence of a big quake and tsunami at that time in the Kodiak area, and there might have been a different but smaller Shumagin-area quake that was confused with the first, he said.