October 28, 2025
The tiny figures in this drone photo are standing near one of the stranded coral boulders, showing how far inland the tsunami carried it. Credit: Michaela Spiske
Sometime between 1381 and 1391, an earthquake exceeding magnitude 8.0 rocked the northeastern Caribbean and sent a tsunami barreling toward the island of Anegada.
Flooding scattered debris across the island, depositing coral boulders hundreds of meters inland. The corals died but their skeletons remain. More than six centuries later, scientists are learning that these skeletons hold clues about tsunami history. Computer models showed the flooding likely resulted from a tsunami caused by a large earthquake in the nearby Puerto Rico Trench.
Now, in an open-access paper recently published in Geophysical Research Letters, led by Hali Kilbourne, an associate research professor at the University of Maryland Center for Environmental Science, researchers narrow the tsunami time frame to the last decades of the 14th century. The researchers expect this finding to support ongoing efforts to prepare for future Caribbean tsunamis.
“Knowing when a tsunami happened can help make it more real for people who live in the area, so they are more likely to take precautions, Kilbourne said.”
“If you’re designing a school or a hospital near the coast, you want to know whether there’s a chance that a very big earthquake could occur, and you want to design that building to withstand it,” said co-author Brian Atwater, a University of Washington affiliate professor of Earth and space sciences and research geologist with the United States Geological Survey.
Anegada is the northernmost of the British Virgin Islands, sitting just south of the Puerto Rico Trench, where the Caribbean and North American plates converge. Most of the islands are protected by a broad, shallow continental shelf. Waves lose energy as they roll across the expanse, decreasing the chances of a tsunami hitting Caribbean shores. Anegada is different — the seafloor slopes steeply toward the deep trench, making the island more hazard-prone.
Written records from the northeastern Caribbean go back five centuries, but none provide evidence for a tsunami from the Puerto Rico Trench. Geology allowed the researchers to evaluate tsunami history on a longer timescale.
Researchers began surveying the region after a massive earthquake and tsunami struck the Indian Ocean in 2004, killing more than a quarter of a million people.
The disaster surprised everyone, including researchers, prompting officials in the U.S. to take a closer look at coastal hazards on the Atlantic seaboard. Uri ten Brink, one of the project leads and a research geophysicist at Woods Hole Coastal and Marine Science Center, asked Atwater to check for signs of similar activity on Anegada. Atwater spent years in Indonesia after the tsunami.
The evidence uncovered on Anegada drew various research teams to the island and produced a series of discoveries.
In the most recent study, the researchers present a time frame for the medieval tsunami based on how old the coral was when it died.
They calculated age by measuring two radioactive elements — uranium and thorium — that decay at known rates. These measurements were made on samples from the inside of the coral skeletons, due to weathering and potential contamination. The researchers then tallied annual growth bands in the coral skeletons to determine how many years the corals lived.
“Corals have annual density bands, much like tree rings,” Kilbourne said. “We were able to count how many years passed between the top density bands and the sections we used for dating.”
Kilbourne can also gather valuable environmental data from the coral skeletons, which store information about temperature and salinity.
“The dating work we've presented in this paper indicates that many of these corals lived in the 1200s and 1300s, a time before written records or thermometers in the Caribbean,” Kilbourne said. “It can help us understand the context for modern climate changes and help test ideas about how the climate system functions on longer timescales, such as decades to centuries – time periods that are difficult to study with modern instrumental climate records because the records are not long enough.”
Additional co-authors include Jennifer Weil-Accardo and Nathalie Feuillet at Paris Institute of Earth Physics; Pierre Deschamps at Aix-Marseille University; Yuan-yuan Xu at the University of Delaware; Chuan-Chou Shen at National Taiwan University and Robert B. Halley at Colorado Mesa University and the United States Geological Survey.
This research was funded by the U.S. National Science Foundation, the University of Paris-IPGD, the French National Research Agency, Academica Sinica, the Higher Education Sprout Project of the Taiwan Ministry of Education, the National Taiwan University Core Consortiums Project, the Taiwan National Science and Technology Council.
