THE GFDL HURRICANE PREDICTION SYSTEM AND ITS PERFORMANCE IN THE 1995 HURRICANE SEASON

The Geophysical Fluid Dynamics Laboratory (GFDL) Hurricane Prediction System was adopted by the U.S. National Weather Service as an operatio nal hurricane prediction model in the 1995 hurricane season. The frame work of the prediction model is described with emphasis on its unique features. The model uses a multiply nested movable mesh system to depi ct the interior structure of tropical cyclones. For cumulus parameteri zation, a soft moist convective adjustment scheme is used. The model i nitial condition is defined through a method of vortex replacement. It involves generation of a realistic hurricane vortex by a scheme of co ntrolled spinup. Time integration of the model is carried out by a two -step iterative method that has a characteristic of frequency-selectiv e damping.The outline of the prediction system is presented and the sy stem performance in the 1995 hurricane season is briefly summarized. B oth in the Atlantic and the eastern Pacific, the average track forecas t errors are substantially reduced by the GFDL model, compared with fo recasts by other models, particularly for the forecast periods beyond 36 h. Forecasts of Hurricane Luis and Hurricane Marilyn were especiall y skillful. A forecast bias is noticed in cases of Hurricane Opal and other storms in the Gulf of Mexico. The importance of accurate initial conditions, in both the environmental flow and the storm structure, i s argued.