A Hidden Wave Emerges

On March 27, 1964, at 5:36 pm, southern Alaska shook intensely. Sixty years later, the magnitude 9.2 Great Alaska Earthquake remains the second-largest ever recorded.  At the time, the immense seismic potential of this part of the country was not fully recognized. The shaking triggered massive underwater landslides, generating deadly tsunamis in many coastal communities. In Anchorage, damage resulted from shaking, ground subsidence, and landslides. Despite its coastal location, no tsunami was observed.

For decades, the absence of significant tsunamis in Cook Inlet supported the idea that its length and shallow slope offered protection from such events. However, recent findings from Alaska tsunami researchers suggest that a tsunami did indeed reach upper Cook Inlet on that night in 1964.

Three Alaska tsunami scientists—Elena Suleimani, Barrett Salisbury, and Dmitry Nicolsky—worked together to reassess the tsunami hazard throughout the upper Cook Inlet region ( full report here ).

1964 As the theory of plate tectonics was emerging, understanding of earthquake and tsunami dynamics was limited. There were no tools to estimate when and where tsunamis might strike communities.

Late 1990s Numerical modeling tools for tsunami dynamics were becoming increasingly reliable. “When a tsunami has been generated by an underwater earthquake, numerical modeling is the only way to describe its propagation over real ocean topography, and it is the only way to forecast the arrival times at different locations along the coastline,” explains Elena Suleimani, a tsunami scientist with the Alaska Earthquake Center.

2000–2020  Geologic research over the past two decades in Southcentral Alaska reveals that there have been even larger earthquakes there in the past. Geologists examining soil layers around the 1964 rupture zone (Copper River Delta, Girdwood, and the Kenai Peninsula) show that the 1964 earthquake was not the biggest to hit the Gulf of Alaska. 

2011 The location of earthquake rupture is a critical component of tsunami generation. The magnitude 9.1 Tohoku earthquake and ensuing tsunami offshore of Japan were shocking in their devastation, and also because scientists were caught off guard. The earthquake ruptured in an unexpected place, at the shallow edge of a subduction zone. This triggered more ocean floor movement, causing larger tsunami waves, than was believed possible.

In 1964, the greatest seafloor uplift occurred under Prince William Sound and Kodiak Island, causing complicated wave interactions that influenced how the tsunami entered Cook Inlet. In the future, if maximum seafloor uplift occurs under the entrance to Cook Inlet, a more powerful tsunami could reach Anchorage.

Because tsunamis can strike within minutes of an earthquake, awareness and preparedness are crucial for anyone who lives, works, or travels along Alaska’s coast.

Tsunamis result from either significant land displacement into the ocean (from above or below the water) due to events like landslides or volcanic eruptions, or from uplift of the seafloor caused by earthquakes. The latter are known as tectonic tsunamis, and can have wide-reaching effects.

Source #1: Earthquakes. Along the Alaska–Aleutian subduction zone, the oceanic Pacific Plate scrapes under the North American continental plate, causing some of the world’s biggest earthquakes. The 1964 earthquake ruptured nearly 500 miles of this subduction zone. The magnitude 9.2 quake is the second-largest instrumentally recorded earthquake in the world.

Source #2: Landslides. Many spots along Alaska’s rugged coast are poised for landslides above or below the ocean’s surface. Sudden mass failures could generate tsunamis that reach coastlines in minutes. Alaska has experienced several of the world's largest landslide-generated tsunami waves.

Upper Cook Inlet has some of the most dramatic tides in the world, with water levels changing nearly 30 vertical feet between high and low tides. The inlet stretches roughly 220 miles from the Gulf of Alaska to the Municipality of Anchorage, narrows from 80 to 9 miles, and has an average depth of 330 feet. At Anchorage, Cook Inlet splits into Knik Arm and Turnagain Arm. Low tide exposes miles of glacially fed mudflats. 

Cook Inlet is so long and shallow, and the tidal changes are so large, that opposite ends of Cook Inlet can have opposite tidal stages. The extent of flooding in communities depends on when the wave arrives in the tidal cycle. 

In this hypothetical example, the earthquake causes subsidence, or down-dropping of land (shown by the black dashed line in the figure to the right). The earthquake occurs just before high tide in Anchorage. When the tsunami arrives many hours later, the tide is going out. The outgoing tidal flow (toward the Gulf of Alaska) fights against the incoming tsunami. Because the tidal range is so large and dynamic, the tsunami is effectively "swallowed" by the tide.

In this hypothetical example, like in the last, the earthquake causes down-dropping of land in the pper Cook Inlet region. However, the earthquake occurs as the tide is falling in Anchorage, such that when the tsunami arrives several hours later it coincides with the next high tide. The combination of high tide and tsunami wave produces inundation above the normal high tide level. 

Underwater landslides generated tsunamis almost immediately, in some cases before ground shaking stopped. In those communities, many people died because they did not know to leave the coastline immediately after they felt the earthquake. In other places, for example Kodiak, the earthquake-generated tsunami didn’t arrive for nearly half an hour, allowing time for evacuation. 

When geologists dig deeper, they find evidence of repeated major earthquakes, revealed in the unique layers of mud and soils around Cook Inlet. In the early 2000s, scientists dug into tidal mud around Girdwood and Anchorage to decipher evidence of past large earthquakes. Geologists can also use ash layers—such as from the 1912 Novarupta/Katmai volcanic eruption—as well as isotopes and other markers, to help date earthquakes even deeper in history.

The Ghost Forest of Girdwood and the soils it sprouts from tell the stories of past devastating earthquakes. Dead trees dot the land that dropped several feet below sea level, immersing the vegetation in salt water. Over a period of several years, silt brought in by the tides buried the tree stumps and grasses, creating a peat layer.

In a similar fashion, older layers of peat and tree stumps provide geologic evidence that 1964-style great earthquakes have occurred repeatedly in the past. Microscopic fossils in these deposits, such as diatoms that lived in intertidal zones, show that the transitions from freshwater to saltwater environments occurred suddenly during tectonic events, rather than gradually due to long-term sea-level rise. 

Learn more about how people search for evidence of past tsunamis:

The 2011 Tohoku tsunami taught us a great deal about the importance of earthquake location. On March 11, 2011, a magnitude 9.1 earthquake struck off the coast of Japan. Despite being one of the most tsunami-ready countries in the world, Japan was still surprised by the size of the tsunami. Water flooded over sea walls and caused widespread damage. The location of the earthquake rupture was unexpected, bringing new insight into how large subduction-zone earthquakes can occur.

Two major lessons from the 2011 Tohoku earthquake and tsunami guided tsunami researchers as they created the upper Cook Inlet tsunami hazard assessment. 

Earthquake rupture can spread in unexpected ways. Highly sensitive GPS instruments can measure small changes in ground levels, tracking where tectonic plates are locked and storing energy. In the Tohoku earthquake, rupture occurred basically where a locked patch released and triggered a neighboring patch, creating a larger rupture zone. Scientists didn’t necessarily expect this type of rupture path.

Earthquake rupture can occur closer to the seafloor than previously thought. In the Tohoku earthquake, the rupture spread into a shallow section of the subduction zone, nearly reaching the seafloor. The amount of seafloor uplift determines the size of a tsunami, so a shallow, large earthquake causes a larger tsunami. The shallow rupture of the Tohoku earthquake also affected the pattern of subsidence (or how the land drops across a region), causing more tsunami flooding than was anticipated.  

What the next large tsunami in upper Cook Inlet might look like

Using all of the developments in tsunami science since 1964, the team came up with a real, but rare, worst-case scenario to guide tsunami preparedness in upper Cook Inlet.

Like the water and earth at the heart of tsunami science, our insight into these powerful forces is always growing and changing. Each tsunami teaches us something new.