Tsunami hazard assessment along the US East Coast from extreme sources in the North Atlantic Ocean: a limited probabilistic analysis

Coastal hazard is assessed along the US East Coast (USEC) through model simulation of 14 probable maximum tsunamis (PMTs) from sources located in the North Atlantic Ocean basin, in terms of: (i) standard inundation maps and other hazard products, and (ii) a limited probabilistic tsunami hazard analysis (PTHA) based on these. Tsunami sources are based on historical records or are hypothetical, including extreme seismic events in the Puerto Rico Trench (PRT)/Caribbean Arc area and in the Açores Convergence Zone (ACZ), near-field submarine mass failures (SMF) on or near the continental shelf break, and large scale volcanic flank collapses of the Cumbre Vieja Volcano (CVV), on La Palma in the Canary Archipelago. Various state-of-the-art models are used to simulate tsunami generation from these sources and propagation to the USEC, in a series of nested grids, by one-way coupling. Based on these results, hazard metrics are computed for each PMT and their envelope as: (i) a synoptic representation along the entire USEC (with results interpolated on the 5 m isobath in coarse 450 m grids), and (ii) high-resolution (10–30 m) inundation maps, for about 30% of the USEC to date, prioritizing the most exposed areas. While existing maps can be accessed in a repository, in this paper, we focus on assessing the synoptic tsunami hazard using 4 hazard metrics (surface elevation, flow velocity, momentum force and arrival time) and their combination into a Tsunami Intensity Index (TII) introduced in earlier work. A comparison of metrics for each PMT shows similar alongshore patterns of higher/lower impact, due to wave focusing and defocusing effects induced by the shelf bathymetry; on this basis, 14 focusing areas are identified from Maine to southern Florida, where tsunami hazard for any PMT is always relatively larger. The limited PTHA is then performed based on the PMT metrics, using estimated return period ranges for each source (in the $$T_r\in [500; 50,000]$$ T r ∈ [ 500 ; 50 , 000 ] year interval), from the literature and/or educated guesses. This analysis yields return periods along the USEC for 5 classes of metrics and the TII. The PTHA results are then interpolated to provide the same metrics along the USEC for $$T_r = 500$$ T r = 500 and 2,500 years, which are standard in geohazard assessment. PTHA results show, depending on low-high estimates of PMT return periods that: (i) up to $$T_r = 137-261$$ T r = 137 - 261 years, there is no probable tsunami hazard; (ii) for $$T_r =500$$ T r = 500 years, tsunami hazard would be expected to be low-medium and only larger near some focusing sites, with flooding metrics at par with a severe coastal storm; and (iii) for $$T_r =2,500$$ T r = 2 , 500 years, which is commensurate with the longest return periods considered on the US west Coast for large Cascadia events, the USEC would face high tsunami hazard, with widespread destruction near the focusing areas and severe impact elsewhere. As uncertainties are large on the estimated return period ranges used and, hence, on the results of this limited PTHA, while raising awareness for the potential tsunami hazard facing the USEC, this work stresses the need for conducting a more in-depth and comprehensive PTHA, including using a larger selection of tsunami sources and further supporting and refining the range of return periods selected for each of those. Such work has been initiated and will be the object of future papers.