GEOLOGIC EVIDENCE FOR A TSUNAMI SOURCE NEAR THE PUERTO RICO TRENCH

Coral boulders of medieval age at Anegada, British Virgin Islands, calibrated to local geologic effects of far-field tsunamis and hurricanes, provide tangible evidence for the generation of a tsunami by faulting along the eastern Puerto Rico Trench. Anegada is fringed on the north and east by a coral reef 100-1200 m offshore; founded on Pleistocene reefal carbonate; rimmed by sandy Holocene beach ridges; and bermed with coral-rubble on a rocky stretch of its north shore. CORAL BOULDERS: Scores of coral heads up to 2 m in diameter were moved across the north shore in medieval time. Some crossed the line of the modern storm berm, continued over a limestone rise 4 m above sea level, and came to rest on lower ground hundreds of meters farther south. Others traversed beach ridges, and two of these boulders are now 1.5 km from the fringing reef. Most are the brain coral Diploria. Some retain enough of their originally rounded, dimpled shape to have been deposited live. The likely time of emplacement is AD 1200-1450. This range is based on radiocarbon dating of outer growth bands of 18 heads from 5 separate areas. The youngest of the ages is 890±25 ¹⁴C yr BP, and the ΔR assumed is 0 to –200 ¹⁴C yr. CALIBRATIONS: A sand sheet securely dated to 1650-1800 represents either the largest known far-field tsunami in the Caribbean (1755 Lisbon) or some other tsunami or unusual storm that surpassed the Lisbon tsunami in its local geologic effects [refs 1-4]. Because the sheet extends 1.5 km inland, we had expected the coral boulders to be coeval with it. Instead they yielded only the pre-Columbus ages cited above. Hurricanes Donna (1960) and Earl (2010) rank below the 1650-1800 overwash in their Anegada effects. Their lasting deposits are limited to sandy spillover fans that extend a few tens of meters inland. nearby tsunami sources: Anegada faces normal faults that fissure the outer rise 200 km to the north. Thrust-earthquake potential may vary along northeast Antilles Subduction Zone [5]. No large slides are evident on the submarine slope that descends from Anegada northward to the Puerto Rico Trench [6]. RELATED ABSTRACTS at fall AGU: Coral-boulder ages (T41A-2562), storm and tsunami modeling (OS32A-05). REFERENCES: [1-4] Natural Hazards 63 (1), 51-149; [5] GRL 39, L10304; [6] Eos 85 (37), 349.