Tuesday, 29 April 2008: 1:30 PM
Palms H (Wyndham Orlando Resort)
Amir Mohammad Yadghar, The Boundary Layer Wind Tunnel Laboratory, University of Western Ontario, London, ON, Canada; and C. Miller
In order to interpret the wind speed profiles obtained using GPS dropwindsondes in hurricanes one first needs to understand the dynamic behaviour of dropwindsonde as it falls through the hurricane boundary layer, particularly with regard to establishing whether the flow characteristics should be viewed from a Lagrangian or an Eulerian viewpoint. In an attempt to achieve this, we model the fall of a dropwindsonde through a simulated boundary layer with known mean and turbulent flow characteristics in order to obtain a link between the observed' dropwindsonde profiles and the known boundary layer characteristics as viewed in a conventional Eulerian framework.
Wind tunnel testing is done first to establish the aerodynamic characteristics of the dropwindsonde body alone and the dropwindsonde body plus parachute. Combining this information with the full non-linear equations of motion for a falling body and an appropriate numerical method for generating an artificial random wind field using a second order autoregressive process to form auto- and cross-correlation matrices allows a model of the behaviour of the falling dropwindsonde in a turbulent atmospheric boundary layer to be developed. This allows the effects of the linearization of the wind finding equations and the conditions under which they may not be valid to be explored. In addition, relationships between both individual and composited observed' dropwindsonde profiles and the simulated boundary layer characteristics can be obtained to assist the understanding of actual observed dropwindsonde profiles.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner