Validation of a long-range trajectory model using gas balloon tracks from the Gordon Bennett Cup 95

In September 1995, 18 gas balloon teams competed at the Gordon Bennett Cup, a long-distance ballooning event. The landing positions, travel times of all teams, and detailed information on the tracks of four teams are available. A special version of the trajectory model FLEXTRA (flexible trajectories) is used that allows the heights of calculated trajectories to be adjusted to the respective balloon heights at every computation time step. The comparison of calculated and observed balloon trajectories allows a validation of the trajectory model. In this case study, the agreement between calculated and balloon trajectories was good, with average relative transport errors of less than 20% of the travel distance after 46 h of travel time. Most of the trajectory errors originate from interpolation errors and from amplifications of small position disturbances in divergent wind fields. Trajectory ensembles, taking into account stochastic errors occurring during the trajectory calculations, are shown to be very reliable in assessing the uncertainties of the computed trajectories. In the present study, the balloon tracks were enveloped by the ensemble trajectories most of the time, suggesting that errors in the analyzed wind fields were relatively small.