National Oceanic and
Atmospheric Administration
United States Department of Commerce


FY 2016

Tsunami Hazard Assessment Based on Wave Generation, Propagation, and Inundation Modeling for the U.S. East Coast

Titov, V., C. Moore, M. Spillane, Y. Wei, E. Gica, and H. Zhou

NUREG/CR-7222, U.S. Nuclear Regulatory Commission, Washington, D.C., Available online at the U.S. NRC Library (2016)

This report describes a comprehensive study of tsunami hazard assessment for the Atlantic coast of the United States (U.S.) based on potential tsunami scenarios.

The study makes use of the Pacific Marine Environmental Laboratory (PMEL) pre-computed database of over a thousand synthetic tsunami sources to identify potentially hazardous tsunami events for the eastern U.S. coastline, in particular the area of Virginia Beach, Virginia. The historical Lisbon 1755 tsunami event is used to validate the simulations by comparing the computed results with the evidence of tsunami impact along the Caribbean arc.

As a result of this investigation, a segment of the Caribbean seismic arc located north of Puerto Rico between the U.S. Virgin Islands and Hispaniola and known as the Puerto Rico Trench, is identified as the most hazardous tsunami source for the U.S. eastern coastline. For potential seismic events of magnitudes between Mw 8.6 and Mw 8.9 the modeled run-up heights are between 3.5 and 5 m in Virginia Beach. In addition to the seismically generated tsunami hazard, the impact of potential tsunamis generated by the possible future collapse of the flank of the Cumbre Vieja volcano in La Palma (Canary Islands), and by the Currituck landslide on the Atlantic continental shelf of the U.S. are also investigated. For the landslide events, the Eulerian-Lagrangian hydrocode, iSALE is used to compute the generated landslide and the solution is coupled to three different tsunami simulation models. Special attention is paid to wave dispersion effects by comparing results from these three different simulations using the shallow water wave equations, the weakly non-linear Boussinesq equations, and the strongly-nonlinear Boussinesq equations. The Method of Splitting Tsunamis (MOST) code is used to compute the non-dispersive shallow water wave solution with numerical dispersion adjusted to match that prescribed by linear theory in deep water.

The results of this study show that dispersive effects tend to be weak when the tsunami propagates over shallow areas of the continental shelf, with good agreement between simulations computed with MOST and with the dispersive Boussinesq-type models.

Feature Publications | Outstanding Scientific Publications

Contact Sandra Bigley |