Trajectory actor, together with a transfer function.

Description

The trajectory actor draws Lagrangian particle trajectories (path lines) or streamlines into a scene, either as 3D tubes or as particle positions.

• Selected features:
  • Supports 3-D visual analysis of pre-computed particle trajectories or particle-clouds.
  • A converter is available for LAGRANTO trajectories. Also, converters for some of the output of the atmospheric dispersion models HYSPLIT and NAME are available (please contact us).
  • Trajectories can be drawn as 3D tubes, coloured by pressure elevation and (new in Met.3D 1.7) any other variable that is provided as auxiliary input data or has been traced/sampled along trajectories computed in Met3D.
  • Shadows of the trajectories may be rendered to the surface.
  • The individual positions of the trajectory particles can also be drawn, either for the entire trajectories or for individual time steps.
  • Animations of particle positions are possible. Also, trajectory tubes can be drawn for the part of the trajectory between trajectory start and current time.
• Required input data:
  • For visual analysis of pre-computed trajectories or particle-ensembles a file with trajectory data is required in a format similar to NetCDF-CF (please contact us for details/examples)
  • For rendering the colours of pre-computed trajectories or particle-ensembles according to pre-computed auxiliary data variables pre-computed auxiliary data variables, these data need to be included as a variable in the NetCDF-CF file with an attribute "auxiliary_data" set to "yes" and the same dimensions as the trajectory vertex information (please contact us for details/examples).
  • For computation of trajectories (streamlines or pathlines) in Met3D, gridded 3-D NWP data with 3 wind velocity components has to be provided 
• GUI interaction:
  • If you are interested in visualizing trajectories or particle-ensembles in Met3D, please do not hesitate to contact us for support. 

How to...

...connect to a trajectory dataset?

• Note that you first need to load a dataset of precomputed trajectories, or configure on-the-fly computation of trajectories. See Actors that manage trajectory data.
• When you first create a trajectory actor, you will be prompted to select a trajectory dataset.
• Alternatively, in the actor properties, select "select data source" - a dialog will pop up from which you can select from the currently available trajectory datasets.

...configure how the trajectories are visualized?

Trajectory data can be visualized in several ways:

• Tubes (as in the sample image above), either showing the entire trajectories (i.e. all time steps) or only part of the trajectories.
• Particles at a specified time step.
• Tubes and particles combined.

To display trajectory data, you need to:

• Select a valid bounding box.
• Select a valid transfer function to have the tubes/particles be coloured according to their pressure elevation or to a selected traced variable. See 1D transfer function (Colour map).
• Define whether the tubes/particles should be coloured according to pressure elevation or according to a traced variable.
• In the user interface, this will look similar to this:
  
• To switch between tubes and particles, select the corresponding setting in the "render mode" property.
• If the dataset provides auxiliary variables that have been traced along the trajectories, you can switch colouring using the "colour according to" and "auxiliary variable" properties. You also need to select a suitable transfer function that fits the range of the variable values. See 1D transfer function (Colour map).
• You can configure tube and particle radius using the properties "tube/sphere radius".

...filter (e.g., domain-filling) trajectories according to their ascent?

This functionality is mainly of interest to visually analyze Warm Conveyor Belts (WCBs). Also see our publication on this topic.

• In the "trajectory filters" property group, enable "min. req. ascent" (minimum required ascent) and set the desired elevation and time intervals. This will only show trajectories that ascent at least PP hPa in HH hours (e.g., a typical value for WCBs is 500 hPa in 48 hours).
• Example:
  

...work with trajectories and streamlines that are computed "on-the-fly" in Met.3D?

Prerequisites for computing trajectories or streamlines on-the-fly:

• You need to configure a dataset that computes the trajectories, see Actors that manage trajectory data.
• You need to add an additional actor to the scene that will act as a source for seed points for the computed lines (i.e., points at which trajectories/streamlines are started). The following actors can be used:
  • Movable poles - seed points will be placed at specified elevations along the pole; the user can interactively move the pole to explore the flow field.
  • Vertical sections - seed points will be placed at specified elevations and at regular horizontal intervals along the vertical section.
  • Horizontal sections - seed points will be placed along regular horizontal intervals on the pressure elevation of the section.
  • Volume bounding box - seed points will be placed at regular intervals in all three spatial dimensions in the box.

To configure trajectory/streamline computation:

• The required properties are available in the "trajectory computation" property group.
• In the "seed points" group, select "add seed actor" to add an actor that will provide the seed points. In the example below, a pole actor is chosen, and a number of vertical levels for the seeds are specified.
• Select whether trajectories or streamlines shall be computed ("type").
• Select the integration method (Euler integration as well as Runge-Kutta 4th-order integration is available).
• Select the interpolation method - "as LAGRANTO v2" will use the same interpolation as the LAGRANTO model, alternatively, simple trilinear interpolation can be used.
• For trajectories, select the integration time (e.g. 48 hours from the start time).
• For streamlines, select a segment length (for the numerical integration) and a number of segments that shall be computed.
• Example:
  

To interactively explore the flow field:

• Simply switch to the "actor interaction mode" (double click in the scene view), grab the pole (or other seed actor), and move it around.