Two operational synoptic surveillance missions were conducted by the National Oceanic and Atmospheric Administration into Hurricane Humberto (2001). Forecasts from two leading dynamical hurricane track forecast models were improved substantially during the watch and warning period before a projected landfall by the assimilation of the additional dropwindsonde data obtained. Tests with a barotropic model suggest that further improvements may be obtained by the use of the Ensemble Transform Kalman Filter for assimilating these additional data into the model.