Simulating the effects of wave dispersion, surface roughness, and debris size on pre-Columbian extreme-wave deposits of Anegada, British Virgin Islands

Abstract:

It is commonly difficult to distinguish geologically between an ancient tsunami and an ancient storm. Numerical simulations of tsunamis and storms may aid in drawing this distinction, which is needed for assessing hazards from extreme waves and their sources. To make the simulations rigorous we are examining how wave dispersion, surface roughness, and debris size influence the expected geological effects of tsunami waves and storm waves. We will be calibrating the results to geology on Anegada, a northeast Caribbean island where extreme waves of unknown source scattered coral and limestone boulders hundreds of meters inland about six centuries ago.

We are using hydrostatic (MOST and HYSEA) and nonhydrostatic (NEOWAVE) models to study the dispersive characteristics of tsunami waves, and a Boussinesq model (BOSZ-2D) to resolve the formation of tsunami-like bores from storm swells. In all models, we are applying friction coefficients that vary with the underlying terrane, which includes a reef crest, a smooth sandy flat between reef and shore, wooded karstic limestone, and mud-floored ponds fringed with mangroves. In addition, the simulations keep track of sedimentary clasts of various sizes that are derived from the reef crest and reef flat, for comparison with the observed boulder fields. Further, the storm scenario for model validation will resemble one of the three hurricanes—Donna (1960), Earl (2010), and Irma (2017)—for which flooding and sedimentation at Anegada have been documented.

We are evaluating four potential explanations for the Anegada boulder fields. Two are near-field tsunamis—from normal faulting on the outer wall of the Puerto Rico Trench, from thrust faulting along the boundary of North America and the Caribbean plates. The others are a far-field tsunami from a Lisbon source, and tsunami-like bores from an unusual storm comparable to the 2013 Super Typhoon Haiyan. In model runs thus far, only the near-field tsunamis inundate the most inland parts of the boulder fields. The storms considered are as large as category 5 but do not yet generate tsunami-like bores. It is hoped that the eventual findings will clarify near-field earthquake and tsunami hazards in the Virgin Islands and on the north shore of Puerto Rico.