Abstract
During the past several years a number of techniques have been developed for forecasting the motion of tropical cyclones over areas of the Atlantic, Caribbean, and Gulf of Mexico. Many of these have been tested on an operational basis. These forecasts have been collected and verified and probability ellipses have been constructed. While direct comparisons between the forecast methods are difficult (because of inhomogeneity of sample size, differences in times of availability of the forecasts, etc.), it has been found that two of the objective techniques tested result in slightly better forecasts than the others. The accuracy of the forecasts depends somewhat on the geographical area and the forecast error varies almost linearly with the length of the forecast period.
