Forward/reverse wind parcel trajectory path computation and visualization through a 2D or 3D, earth navigated, gridded Eulerian wind field has been developed within the VisAD library.
Parcel displacement computation methodologies include straightforward Euler method, HYSPLIT (mean flow), and Runge-Kutta (RK4). Time interpolation is either linear or piecewise cubic. High spatial and temporal resolution can be computationally challenging, particularly with 3D domains, so interpolation and parcel displacement are performed “lazily”, i.e. only on grid cells intersected by a trajectory path. Multiple CPU cores can be leveraged to compute in parallel when possible.
Multiple types of trajectory visual depictions include line, point, kinematic derived deformable ribbons, and shaded fixed width tubes. Interactive controls include adjustable time visibility window, trajectory density, and support for arbitrary starting locations. Trajectory depictions can be colored by an initial scalar field, e.g. a conservative parameter such as Potential Temperature, or non-conservative such as smoke or aerosol concentration, or other parameters following the motion like wind speed.
Users access this capability by driving the McIDAS-V or IDV gui for dataset selection which triggers a chain of events: internal metadata recognition to construct the appropriate VisAD Data Model object and rendered according to selections made associated display control widgets.
Examples of applications utilizing computation and visualization of wind parcel trajectories using McIDAS-V will be provided including NearCast, transport of conservative scalar such as Equivalent Potential Temperature, visualization of kinematic properties, isentropic analysis and forward propagation using forecast models in the vicinity of smoke or volcanic ash.