The 2015 Taan Fiord Landslide and Its 633-Foot Tsunami: A Warning for Icy Bay

On October 17, 2015, 180 million tons of rock slid into Taan Fiord, an arm of Icy Bay, and set off a tsunami that stripped the forest from 8 square miles (20 square km) of Wrangell-St. Elias National Park and Preserve. The wave reached as high as 633 feet (193 m) above the fjord, the fourth-highest tsunami ever recorded. It had almost no human impact: nobody was near enough to be harmed, and the only damage to infrastructure was rocks scattered on a beach used for landing bush planes. The five authors call that good fortune that may not characterize similar events in the future.

How a landslide tsunami differs from an earthquake tsunami

Most research on tsunami impacts has looked at longer-period regional tsunamis, so it is unclear how much applies to landslide tsunamis.

Why Alaska's parks are prone

Landslides that raise large tsunamis most often happen where glaciers are retreating. The authors list 14 tsunamis with runups of 50 meters or more: 10 came from mountain landslides into fjords or lakes in glaciated terrain, and four were in Alaska parks. Four giant tsunamis in Alaska have been triggered by landslides: two at Lituya Bay, including 1958, one at Grewingk Lake in 1967 and the one in Taan Fiord.

Climate warming is likely driving more large landslides and creating deep water where large tsunamis can form, the authors say. Glaciers erode the base of mountain slopes, stress them and crack them, but they also fill the valley and prop up the slopes above and below. When they retreat, whether from climate change or ordinary glacial cycles, that support goes, and slopes sag and fractures widen. Thawing, or even warming, of permafrost can greatly weaken mountains too. And retreat exposes new bodies of water: the events at Grewingk Lake and Taan Fiord, and a 2016 landslide tsunami in Cowee Creek near Juneau, all took place in water that did not exist a few decades earlier, because ice filled the basins.

What happened at Taan Fiord

The slide followed a century of rapid glacial retreat, decades of ground cracking and creep, a month of above-average rain and a few seconds of mild shaking from a distant earthquake. It began as a rotational slide, probably along planes of weakness formed decades earlier, soon after the ice withdrew. Remote-sensing data show active movement back to the mid-1990s, aerial surveys recorded tens of meters of gradual motion, and other researchers noted ground cracking and deformation a decade before the failure. Most of the slide mass crossed the shoreline and the end of Tyndall Glacier and entered the fjord. Some of it ran across the fjord bottom and back up the far side, forming debris hillocks 30 to 70 feet (10-20 m) high.

Nobody saw it. Automatic seismic systems identified the slide within hours, satellite imagery showed the slide and tsunami days later, and scientists reached the site months later. A tide gauge at Yakutat, more than 90 miles (about 150 km) away, recorded a change in water level of 4 inches (about 0.1 m).

What it did to land, water and ships

On land the wave stripped the forest to heights that fell away down the fjord. Ground 60 feet (20 m) below the inundation limit was typically left with only torn roots and bits of soil, while trees near the limit were often toppled but still rooted. The wave moved boulders up to 16 feet (5 m) across. Deposits were typically 12 to 20 inches (30-50 cm) thick as far as 6 miles (10 km) from the source, and over 16 feet (5 m) thick where loose sediment was abundant. One small island that once held vegetation above the tide was scoured into a shoal. A hill about 800 feet (250 m) across, surrounded but not overtopped, had runups of over 160 feet (50 m) where the water struck it head-on and under 65 feet (20 m) on its sheltered side.

The slide also reshaped the fjord bottom, shallowing some areas by tens of meters, which can make nautical charts unreliable until new surveys are done. For ships the dangers are a breaking wave where its height is large compared with the water depth, and strong currents. In Taan Fiord the currents carried icebergs that gouged the soil and left some stranded above the tide. The authors say ships in such currents could be holed in collisions or swept broadside onto shoals and rolled, small boats could be caught in eddy lines and whirlpools like those of tidal rapids, and in the troughs between waves vessels might ground on obstacles that are normally harmless.

Not every rockslide is a threat: a 2012 slide at Lituya, though very large, stopped more than 6 miles (10 km) from tidewater and raised no tsunami. Even so, the authors expect the hazard to rise as glaciers retreat and fjords and lakes grow toward steep, newly uncovered slopes.

A warning for Icy Bay

The authors say Taan Fiord warns that Icy Bay may see similar, potentially more deadly events. Lituya Bay produced at least five giant tsunamis over three centuries, and Icy Bay could build a similar record. Icy Bay was once filled with glaciers, and four steep-walled fjords began to open at its head in the 1960s. The 2015 tsunami was the largest in that short window but probably not the only one: at least two sites along the fjord have fresh landslide deposits reaching into the water, a ridge opposite the 2015 slide is laced with fissures that may fail, and the steep slopes of the bay's other fjords have not been surveyed for landslide hazards.

Kayakers, bear hunters, commercial and sport fishers and even cruise ships visit the bay. When the tsunami struck, workers were at Icy Bay Lodge 20 miles (32 km) away, beyond the damaging waves. Worse outcomes are easy to imagine, the authors say: a slide at a busy time of year, a small slide or tsunami reaching a beach campsite, or a large slide in another fjord arm, which would hit the outer bay, where more people are.

What the authors recommend

Decades of movement before the failure meant the Taan Fiord slide could have been anticipated, they argue, had anyone been analyzing it. Continuous GPS stations on a slide mass or satellite radar interferometry can reveal the acceleration that warns of failure, and the motion at Taan Fiord was large enough to see even in low-resolution Landsat imagery. Modeling likely could have shown that the impacts would be severe inside the fjord but reach only a short way into the main bay.

For parks they propose systematic field surveys and remote monitoring of potential slide sources, modeling that outlines likely impact areas, and a comparison of those results with infrastructure, tourist destinations and ship-tracking data to find the most vulnerable places. Development should sit outside areas of expected flooding, with evacuation routes for anyone inside them, and planners should remember that a tsunami can carry riprap boulders and bury areas near stream deltas in sediment. Modeling future events is hard: results depend on where a slide might start, which takes many runs, the currents tsunamis create are rarely documented directly, and tsunami models differ substantially in what they produce. The authors say they are working on a more detailed model.

Sources

• Bretwood Higman (Ground Truth Trekking), Marten Geertsema (University of Northern British Columbia), Dan Shugar (University of Calgary), Patrick Lynett (University of Southern California) and Anja Dufresne (Rheinisch-Westfälische Technische Hochschule Aachen University, Germany), The 2015 Taan Fiord landslide and tsunami, Alaska Park Science 18(1), National Park Service, 2019; rewritten in hubnx's own words.