Skip to: Content
Skip to: Site Navigation
Skip to: Search

Clues from hurricane 'fingerprints'

Scientists decode hurricane 'records' left in trees and rocks to try to predict the strength of future storms.

(Page 2 of 2)

To gather that information, scientists first turned to sediment cores taken from marshes, lagoons, and lakes behind barrier beaches. This "is the most useful and proven technique to date," says Kam-biu Liu, a Louisiana State University paleoclimatologist who first applied the approach to the hurricane problem in 1989. As a tropical cyclone makes landfall, the storm surge washes beach sand into these bodies of water. The sand settles to the bottom, then gets covered with organic material that forms most of the muck at the bottom. The sand layers show up in the cores. Scientists can find out when the storm or storms struck by using radiocarbon dating techniques on the organic material above and below the sandy layer. Dr. Liu and Jeffrey Donnelly of the Woods Hole (Mass.) Oceanographic Institution have pioneered this approach along the Gulf Coast and in the Northeast and Puerto Rico.

Skip to next paragraph

Scientists have been able to trace storm histories back at least 5,000 years. After looking at data from four sites along the Gulf Coast, for example, Liu notes that hurricane activity in the region was fairly low during the first 1,200 years in the samples and the past 1,000 years or so. But for 2,800 years in between, activity was relatively high; a major hurricane struck each site as often as once every 200 years.

Researchers are now trying to figure out the atmospheric and oceanic drivers for such long-term swings.

Clues written in stone

Where some researchers hunt for hurricanes in the muck, others are turning to the record written in stone and wood. Both approaches take advantage of changes that severe storms bring to the ratio of oxygen isotopes in water they drop. Rain from tropical cyclones carries more of the lighter oxygen isotopes than rain from ordinary storms. University of Houston researcher James Lawrence – who had been working on the isotopic content of storm water – saw that the approach could be applied to hurricane studies and offered the idea up in a paper he published in 1998. That sent Boston College's Dr. Frappier and researchers at the University of Tennessee, among others, off to caves and forests.

By 2001, researchers were bringing tree-ring samples and stalagmite samples in for analysis. But, notes University of Tennessee tree-ring expert Henri Grissino-Mayer, the work was a lot harder than it looked. Sometimes the best specimens are underwater stumps. Then it's time to date the entire ring, each with a light portion representing early-season growth, and a dark portion representing late-season growth. Because the hurricane season is most intense from August on, this late-season growth is the main target. Then each of those segments gets a once-over for the oxygen-isotope content. The work is painstaking and requires a steady hand to extract the samples.

The stalagmite work is similar. Frappier says this summer's effort in Mexico has been largely a reconnaissance trip to find stalagmites in caves that may hold the information they seek.

In both cases the results from their samples, gathered in 2001, only hit the streets in the past year. The tree-ring results from the work by Tennessee's Claudia Mora and Dr. Grissino-Mayer, and colleagues from the University of South Carolina and the University of New Mexico, appeared last September.

Frappier's results appeared last February. The tree rings tracked hurricanes back some 220 years. And while the record from the stalagmite Frappier selected stretched back 23 years, it records storms by week and month, not just by season. Beyond these approaches, "we are also experimenting with the application of corals to this line of research, although it's not as far along as tree rings and cave deposits," Liu notes.

Still, "by putting together this team of people and tackling the problem from different angles using different techniques and looking at different time scales we will have a better understanding of the spatial and temporal variations in hurricane activity across the entire Caribbean Basin."