Paleo Home Research Data Education What's New Features Paleo Perspectives Site Map Mirror Sites National Climatic Data Center, Asheville, North Carolina navigation bar


Image of survivors of hurricane




Animation by Jeffrey Donnelly, Brown University. Used by permission.




Image of Typhoon Odessa





Preliminary Summary of

Hurricanes are one of the most destructive of natural disasters, causing extensive damage to coastal regions such as the Gulf of Mexico and eastern seaboards of the United States each year. Although the prediction of hurricanes has improved greatly in recent years, our understanding of the natural range of hurricane characteristics such as severity and frequency, and the controls on hurricanes, is incomplete, in part because of the limited length of the instrumental record. However, knowledge about hurricanes can be greatly augmented with paleoclimatic proxy data. A term used for the study of past hurricanes through the use of paleoclimatic data is "paleotempestology."

A recent workshop (March 2001), organized by Cary Mock, Michael Chenoweth, Kam-biu Liu, and Ricardo Garcia Herrera, and held at the University of South Carolina, brought together researchers who have focused their studies on modern and paleohurricanes. Workshop topics addressed included hurricanes and modern climate , forcing factors, reconstructions of hurricane events from historical documents, hurricane chronologies from lake sediments, the potential for documenting hurricane events from tree rings and corals, ecological impacts and responses to hurricanes (see Harvard Forest Data Catalog: Hurricane Studies and LTER Database), and the practical applications of this work.

A variety of sources of paleoclimatic proxy data have been used to investigate paleohurricanes. One of the most promising sources of paleohurricane data appears to be archived in historical documents. Of key importance are ships' logs for the Atlantic Ocean, which recorded wind direction and force, rain, and some recording barometric pressure. Logbooks from the British Navy are an excellent resource, documenting conditions at sea and at port at up to an hourly resolution. Land-based records, such as newspapers, are also used together evidence and characteristics of individual hurricane events. The Spanish Colonial Archives represent a wealth of information from the era of the Spanish Conquest, with records going back to the late 15th century. Detailed histories of U.S. hurricane events have been compiled for individual states (Texas, Louisiana, Virginia , New England and others) and individual events (e.g., 1813, 1878, 1898).

Other sources of high resolution (annual to monthly) paleohurricane data may be obtained from corals and tree rings. A coral record from Puerto Rico reflects variations in both the Atlantic and Pacific SSTs and may provide information about the long term role of both ENSO and the Atlantic in hurricane occurrence. In the Caribbean, coral chemistry within annual growth bands is being investigated to determine if changes in isotopic signature can be used to document hurricane events. Rainfall from hurricanes has been found to have a distinct isotopic signature which can be detected in coral chemistry. Fluorescence intensity of corals may also document hurricane conditions. Tree rings can reflect damage or changes in ring growth patterns due to the impact of hurricanes, so an analysis of ring patterns in trees in impacted areas may be used to document hurricane events. The hurricane isotopic signature found in corals may be detectable in tree rings as well, although this has not yet been tested.

At long time scales and lower resolution, lake and marsh sediments provide chronologies of the most severe hurricane events. Lakes, lagoons, and marshes located just beyond coastal barrier dunes can record hurricane events in their sediments when severe hurricanes cause the sea to flow over barrier dunes and wash sandy sediments into these water bodies. (See animation in left column and refresh or reload page to restart). The sand layers are interspersed with layers comprised of organic material which can be dated, using radiocarbon, cesium or lead210 dating methods. The location of the sand layers, relative to the lake edge, can also provide information on the strength of the events. Lakes and marshes have been sampled from the Gulf coast up to Cape Cod. Results from studies suggest a hyperactive period of hurricane activity from about 1000 to 3400 years B.P., and a quite period prior to 3400 B.P., perhaps influenced by the location of the Bermuda High. New England lake sediments, some of which are varved (contain annual layers) also offer promising chronicles of hurricane events, with records going back the 12th and 13th centuries

Other researchers discussed dynamics and possible forcing mechanisms for hurricanes and hurricane tracks. On short time scales the role of upper tropospheric trough plays an important role and the influence of ENSO events was also acknowledged. Less is understood about low frequency controls, but variations in the Atlantic appear to be key. The Northern Atlantic Oscillation was discussed as well as a perhaps more important Atlantic multidecadal oscillation. The part SSTs play and the location and strength of the Bermuda High were also topics of discussion. A better understanding of the low frequency characteristics of hurricanes and what influences low-frequency changes must come from the development and interpretation of paleohurricane records.

The issue of funding for and uses for paleohurricane data came up and was addressed in the context of the Bermuda Biological Stations's Risk Prediction Initiative. This initiative has brought together scientists and users of paleohurricane data - primarily the insurance industry. The partnership has helped fund paleohurricane research at a number of institutions.

One result of the workshop was a sharing of information via links to web pages provided by many of the participants. Most of the links on this web page are contributions from workshop participants and can be used to obtain more detailed information about the topics summarized above.

Image of NOAA logo NOAA Paleoclimatology Program
Contact Us
18 February 2000

Return to Paleohurricane Resource Center homepage