Forecasting of Tropical Cyclone Motion Using an EOF (Empirical Orthogonal Function) Representation of Wind Forcing.
NAVAL POSTGRADUATE SCHOOL MONTEREY CA
Pagination or Media Count:
Empirical Orthogonal Function EOF analysis is used to represent the environmental wind forcing of selected western North Pacific tropical cyclone tracks from 1979-1983. The EOF analysis is applied separately to the zonal and meridional wind components at 700, 400 and 250 mb on a 527-point grid with 288.7 km zonal and meridional spacing that is relocated with the storm center. The 527 EOF coefficients for each level and component are computed for a sample of 682 cases. The coefficient vectors are truncated to the first 35 coefficients based on a Monte Carlo selection criterion. These coefficients account for at least 82 percent of the variance in each field. The EOF coefficients, along with storm movement during the past 24 hours, position, date and intensity, are then used as potential predictors in a regression analysis forecast scheme for tropical cyclone motion. The EOF-based regression equations are tested on the dependent data cases. The mean 72-hour track forecast error is between 450 and 500 km. Therefore, it appears that this regression scheme has potential for operational applications.