Extended Abstract (2.0M)4.4
Visualization of Radar Data in Three-dimensions
Arthur L. Doggett IV, Texas Tech University, Lubbock, TX; and X. Gilliam, K. Manross, and M. Gamel
To more fully explore storm scale features associated with severe thunderstorms, it has become necessary to incorporate radar data from all elevation angles of the volume scan and evaluate those data from a volumetric standpoint. This would allow better study of features such as the rear-flank downdraft, updraft core, mesocyclone, and downburst regions.
Most radar software packages do not provide for such volumetric representation of the data. The user is left to mentally align data from various scans and qualitatively construct three-dimensional storm structure. At best, cross sections and proxy variables such as VIL and max reflectivity provide some depth to the data sets. However, technology from fields such as medial imaging is available that can be used to provide more compressive depictions of the radar volumes scans.
In an attempt to develop better three-dimensional rendering, the NASA radar software library (RSL) is being used to access unprocessed Level-II NEXRAD radar data. Then, by using data processing and visualization software such as MATLAB and Vis-5D, it is possible to construct three-dimensional volumetric representation of storm structure from the raw radar data. Such representation of the data can provide a richer interpretation of storm evolution, structure, and dynamics. While the visualization is still rather crude, it does open the door to a wide variety of applications.
Additionally, it is hoped that the raw radar volume scans can be put into a data format that can be used by image processing algorithms and software. This would allow more sophisticated representations to be developed and expand the utility of the technique.
Session 4, Radar and Multi-Sensor Applications
Tuesday, 13 August 2002, 8:00 AM-10:00 AM
Previous paper Next paper