Creating a New Suite of Statistical Models that Predict the Onset of an Eyewall Replacement Cycle

On September 12, 2008, Hurricane Ike was taking aim at the coast of Galveston, TX. Once a strong Category 4 hurricane, Ike had now weakened considerably to a Category 2 storm. However, Ike had also undergone multiple “eyewall replacement cycles” and therefore, while it was not a particularly intense hurricane, it was now exceptionally large and capable of bringing devastating storm surge to the Galveston area. Because of this, and the fact that coastal residents often equate danger only in terms of hurricane intensity and not size, the warnings issued by NOAA’s National Hurricane Center included remarkably extreme language. To convey the high danger of this storm, the warnings included the statement:

“All neighborhoods ... and possibly entire coastal communities ... will be inundated during the period of peak storm tide ... Persons not heeding evacuation orders in single-family one- or two-story homes will face certain death.”

“Eyewall replacement cycles”, in which a second concentric eyewall forms around the original eyewall and eventually supplants it, are common in strong long-lived hurricanes. Like Hurricane Ike, Hurricane Katrina also weakened but grew in size as the result of an eyewall replacement cycle, and the large area of strong wind led to an enormous storm surge that devastated the Gulf coast. Contrarily, Hurricane Charley made landfall as a stronger but much smaller hurricane (see figure) and the smaller area of strong wind contributed substantially to lessening the associated damage.

Hurricanes can grow gradually over time, but eyewall replacement cycles usually lead to a rapid expansion of the area of strong winds, as well as large fluctuations in peak intensity. Because of the associated rapid changes in size and intensity, skillful forecasting of eyewall replacement cycles is essential to protect life and property. But until very recently there was no official guidance available for forecasting these events. To address this deficiency, James Kossin of NOAA’s National Climatic Data Center (NCDC) along with Matthew Sitkowski, a graduate student at the University of Wisconsin-Madison, used large archives of geostationary and microwave satellite data, in combination with aircraft reconnaissance data, to create a new suite of statistical models that predict the onset of an eyewall replacement cycle and the associated changes in hurricane intensity and size.

One of these new forecast models is now part of the official suite of operational products available to forecasters at NOAA’s National Hurricane Center, and the others are currently being transitioned to operational testing. This project is part of the NOAA Joint Hurricane Testbed (JHT) Project, which is part of the overarching Hurricane Forecast Improvement Project (HFIP) under the NOAA Weather-Ready Nation strategic initiative. The formulation and testing of these new eyewall replacement cycle forecast-models required extensive use of large datasets compiled from many years of data from NOAA’s operational geostationary and polar-orbiting satellites.



Kossin, J. P., and M. Sitkowski, 2009: An objective model for identifying secondary eyewall formation in hurricanes. Mon. Wea. Rev., 137, 876-892.

Kossin, J. P., and M. Sitkowski, 2012: Predicting hurricane intensity and structure changes associated with eyewall replacement cycles. Wea. Forecasting, 27, 484-488.

Sitkowski, M., J. P. Kossin, and C. M. Rozoff, 2011: Intensity and structure changes during hurricane eyewall replacement cycles. Mon. Wea. Rev., 139, 3829-3847.

Sitkowski, M., J. P. Kossin, C. M. Rozoff, and J. Knaff, 2012: Hurricane eyewall replacement cycles and the relict inner eyewall circulation. Mon. Wea. Rev., in press.