Donnelly and Woodruff began reconstructing the history of land-falling hurricanes in the Caribbean in 2003 by gathering sediment-core samples from Laguna Playa Grande on Vieques (Puerto Rico), an island extremely vulnerable to hurricane strikes. They examined the cores for evidence of storm surges—distinctive layers of coarse-grained sands and bits of shell interspersed between the organic-rich silt usually found in lagoon sediments—and pieced together a 5,000-year chronology of land-falling hurricanes in the region.

In examining the record, they found large and dramatic fluctuations in hurricane activity, with long stretches of frequent strikes punctuated by lulls that lasted many centuries. The team then compared their new hurricane record with existing paleoclimate data on El Niño, the West African monsoon, and other global and regional climate influences. They found the number of intense hurricanes (category 3, 4, and 5 on the Saffir-Simpson scale) typically increased when El Niño was relatively weak and the West African monsoon was strong.

“The processes that govern the formation, intensity, and track of Atlantic hurricanes are still poorly understood,” said Donnelly, an associate scientist in the WHOI Department of Geology and Geophysics. “Based on this work, we now think that there may be some sort of basin-wide ‘on-off switch’ for intense hurricanes.”

Donnelly and his colleagues have pioneered efforts to extend the chronology of hurricane strikes beyond what can be found in historical texts and modern meteorological records and previously applied their methods to the New England and the Mid-Atlantic coasts of the United States.

Their research area, Laguna Playa Grande, is protected and separated from the ocean during all but the most severe tropical storms. However, when an intense hurricane strikes the region, storm surges carry sand from the ocean beach over the dunes and into Laguna Playa Grande. Such “over-topping” events leave markers in the geological record that can be examined by researchers in sediment core samples.

The geological record from Vieques showed that there were periods of more frequent intense hurricanes from 5,000 to 3,600 years ago, from 2,500 to 1,000 years ago, and from 1700 AD to the present. By contrast, the island was hit less often from 3,600 to 2,500 years ago and from 1,000 to 300 years ago.

To ensure that what they were seeing was not just a change in the direction of hurricanes away from Vieques—that is, different storm tracks across the Atlantic and Caribbean—the scientists compared their new records with previous studies from New York and the Gulf Coast. They saw that the Vieques record matched the frequency of land-falling hurricanes in New York and Louisiana, indicating that some Atlantic-wide changes took place.

Donnelly and Woodruff, a doctoral student in the MIT/WHOI Joint Graduate Program, then decided to test some other hypotheses about what controls the strength and frequency of hurricanes. They found that periods of frequent El Niño in the past corresponded with times of less hurricane intensity. Other researchers have established that, within individual years, El Niño can stunt hurricane activity by causing strong winds at high altitudes that shear the tops off hurricanes or tip them over as they form. When El Niño was less active in the past, Donnelly and Woodruff found, hurricane cycles picked up.

The researchers also examined precipitation records from Lake Ossa, Cameroon, and discovered that when there were increased monsoon rains, there were more frequent intense hurricanes on the other side of the Atlantic. Researchers have theorized that frequent and stronger storms over western Africa lead to easterly atmospheric waves moving into the Atlantic to provide the “seedlings” for hurricane development.

Much media attention has been focused recently on the importance of warmer ocean waters as the dominant factor controlling the frequency and intensity of hurricanes. And indeed, warmer sea surface temperatures provide more fuel for the formation of tropical cyclones. But the work by Donnelly and Woodruff suggests that El Niño and the West African monsoon appear to be critical factors for determining long-term cycles of hurricane intensity in the Atlantic.

Donnelly and Woodruff published their latest results in the May 24 issue of the journal Nature.

The research by Donnelly and Woodruff was funded by the National Science Foundation, the Risk Prediction Initiative, the National Geographic Society, the WHOI Coastal Ocean Institute, and the Andrew W. Mellon Foundation.

Note: If no author is given, the source is cited instead.

• Hurricanes and Cyclones
• Severe Weather
• Weather

• Early Climate
• Fossils
• Paleontology

• List of major natural disasters in the United States
• Great Hurricane of 1780
• Hurricane Opal
• Hurricane Agnes
• List of Category 5 Pacific hurricanes
• National Hurricane Center