Source and progression of a submarine landslide and tsunami: The 1964 Great Alaska earthquake at Valdez

Abstract

Like many subduction zone earthquakes, the deadliest aspects of the 1964 M = 9.2 Alaska earthquake were the tsunamis it caused. The worst of these were generated by local submarine landslides induced by the earthquake. These caused high runups, engulfing several coastal towns in Prince William Sound. In this paper, we study one of these cases in detail, the Port Valdez submarine landslide and tsunami. We combine eyewitness reports, preserved film, and careful posttsunami surveys with new geophysical data to inform numerical models for landslide tsunami generation. We review the series of events as recorded at Valdez old town and then determine the corresponding subsurface events that led to the tsunami. We build digital elevation models of part of the pretsunami and posttsunami fjord-head delta. Comparing them reveals a~1500 m long region that receded 150 m to the east, which we interpret as the primary delta landslide source. Multibeam imagery and high-resolution seismic reflection data identify ã 400 m wide chute with hummocky deposits at its terminus, which may define the primary slide path. Using these elements we run hydrodynamic models of the landslide-driven tsunamis that match observations of current direction, maximum inundation, and wave height at Valdez old town. We speculate that failure conditions at the delta front may have been influenced by manmade changes in drainage patterns as well as the fast retreat of Valdez and other glaciers during the past century.

Nearly Instantaneous Tsunamis Following the Mw 7.5 2018 Palu Earthquake

“…Deltas, formed from unconsolidated material, are very sensitive to failure by an external trigger such as seismic shaking (Girardclos et al, 2007;Hasiotis et al, 2006;Hilbe & Anselmetti, 2014;Leithold et al, 2018). Furthermore, the potential for delta failures to generate significant tsunamis during strong shaking events (Vanneste et al, 2018) has been documented at sites with diverse climatic conditions, such as Alaska (USA; Field et al, 1982;Parsons et al, 2014) and Haiti (Hornbach et al, 2010).…”

Submarine Deposition of a Subaerial Landslide in Taan Fiord, Alaska

“…There are significant differences in the deposits of this STS landslide with purely submarine landslides in other Alaskan fjords (Table ; see Brothers et al, ; Haeussler et al, , ; Lee et al, , ; Parsons et al, ; Ryan et al, ). As discussed earlier, submarine landslide deposits tend to have a broad lobate front, in contrast to the more pointed front of the Taan Fiord deposit.…”

Effects of a Macro-Roughness Element on Tsunami Wave Amplification, Pressures, and Loads: Physical Model and Comparison to Japanese and US Design Equations

“…This circumstance was partially because the US has experienced fewer damaging tsunamis than other regions in the Pacific Ocean. The most destructive tsunami affecting the West Coast of North America to date was the 1964 Great Alaska Earthquake Tsunami, which caused damage and loss of life in Alaska, California, Oregon, and British Columbia [Parsons et al, 2014]. However, many areas of the US remain vulnerable to damage by earthquake-generated tsunamis.…”

“…However, individual damage depends on local building structure and design, even under the same hydrodynamic conditions. Moreover, wavestructure interaction is difficult to model [Petukhin et al, 2012], and many models remove buildings to allow for more efficient calculations [see Westerink et al, 2008;Parsons et al, 2014;Yao et al, 2014;Grilli et al, 2015]. Such simplifications can result in errors in hydrodynamic outputs.…”

“…For example, after Hurricane Ike affected the Bolivar Peninsula (2008), modeled wave-heights were shown to be greater than those measured by USGS wave gauges by over 1 m in several locations, particularly at locations far from the shoreline; this overprediction affected wave and surge fragility functions derived for pile-elevated residences [Tomiczek et al, 2014]. Tsunami models have also reported that inundation predictions can be overestimated due to coarse grid resolution and smoothed topographies [Parsons et al, 2014;Oishi et al, 2015]. A more thorough understanding of wave propagation through urban environments will enhance the validity and robustness of tsunami simulations.…”