Hurricane predictions more accurate thanks to improved ocean-air model developed by URI scientists
Todd McLeish, 401-874-7892
KINGSTON, R.I. -- July 8, 2005 -- As hurricane Dennis has intensified to a Category 4 storm and threatens to make landfall along the U.S. Gulf coast, one of the tools being used by the National Hurricane Center to predict the storm’s path and intensity was recently improved by University of Rhode Island researchers.
The hurricane prediction model in use, which factors in the effect of the ocean on hurricanes, was developed by Isaac Ginis, a physical oceanographer at URI’s Graduate School of Oceanography, and is coupled with an atmospheric model created by the National Oceanic and Atmospheric Administration’s Geophysical Fluid Dynamics Laboratory (GFDL). The coupled GFDL/URI Hurricane Prediction Model, one of a dozen used by the National Hurricane Center, was the most accurate forecasting model during the 2003 and 2004 hurricane seasons. Improvements made by Ginis and his colleagues following last year’s hurricane season will make it even more accurate this year.
“This year we hope the improved model will be able to predict even more accurately the track and intensity of hurricanes that threaten the United States,” said Ginis. “A more accurate prediction means we can narrow the areas that come under a hurricane watch or warning and reduce the cost of unnecessary hurricane preparations.”
Ginis was the first scientist to demonstrate the significant role the ocean plays in the formation, path and intensity of hurricanes. Using data collected from aircraft, satellites and ocean buoys, the coupled hurricane-ocean model has helped to significantly improve hurricane predictions in the Atlantic Ocean and Gulf of Mexico in the last five years.
“The most important factor in forecasting hurricane intensity is water temperature,” he said. “Hurricanes develop because of the heat from the ocean, which is why they only develop in the summer and they die out once they hit land or move north over colder water.”
Hurricane force winds create ocean currents that cause a mixing of warm surface waters and cooler subsurface waters, leaving a wake of cool water at the surface behind most hurricanes. This cool wake plays an important role in moderating the intensity of hurricanes. The development of the coupled hurricane-ocean model for the National Hurricane Center is Ginis’ major contribution to the field of hurricane forecasting.
During the last two years, Ginis and his colleagues at URI have been improving his model by examining the boundary between the air and ocean.
“Surface waves create friction or drag which has the effect of slowing the surface winds in hurricanes,” explained Ginis. “We used to think that when stronger winds created higher waves, the drag would increase. But it turns out that’s not true. When winds reach about 75 miles per hour and higher, the hurricane seems to just skim across the top of the waves and is less impacted by the surface roughness of the waves.”
As a result of this improved understanding of the interaction between the ocean’s surface and hurricane winds, Ginis and his team converted the research results into the operational model. Now the GFDL/URI 2005 model is expected to be even more accurate at forecasting very intense hurricanes, such as those with winds exceeding 100 miles per hour.
While Ginis and his URI colleagues keep a close eye on this hurricane season, they are already planning additional improvements to the GFDL/URI model that they hope to implement in time for the 2006 season. Ginis’ team was recently awarded a $190,000 grant from NOAA to refine their forecasting model that factors in additional data about ocean waves.
“When a hurricane moves over the ocean, the highest waves are formed at the front right portion of the storm while the back left portion has the lowest waves,” he said. “This creates different drag in different parts of the storm due to the different sea states, and that has an impact on the hurricane’s track. This information will be factored into the model next year.