HR: 11:12h
AN: SF12A-05 [Abstracts]
TI: Multiresolution Precipitation Visualization of Weather Radar Volume Scans
AU: Gerstner, T
EM: gerstner@ins.uni-bonn.de
AF: University of Bonn, Department for Applied Mathematics
Wegelerstr. 6, Bonn, 53115
Germany
AU: * Meetschen, D
EM: dmeetsch@uni-bonn.de
AF: University of Bonn, Department of Meteorology
Auf dem H\"ugel 20, Bonn, 53121
Germany
AU: Simmer, C
EM: csimmer@uni-bonn.de
AF: University of Bonn, Department of Meteorology
Auf dem H\"ugel 20, Bonn, 53121
Germany
AB:
Weather radars measure the backscatter from atmospheric hydrometeors. The
three-dimensional structure of a complete radar scan contains information
which can improve rainfall rate estimates which is crucial for hydrological
modeling and simulations.
For the understanding and conveyance of the data, interactive volume
visualization tools within visual geographic information systems are required.
However, due to the large size of the data it is difficult to achieve real-time
visualization performance. Furthermore, the data has to be put in a context
for better comprehension, e.g. by additionally rendering the local terrain
or satellite image data which further increases the amount of data which has to
be displayed. In addition, the simultaneous display of volume and surface data
leads to algorithmic visualization difficulties.
In this talk, we address these problems by adaptive multiresolution algorithms
(see [1]). Multiresolution methods allow a fast coarse display of the data for
overview images or interaction, while higher resolution is used when zooming
into the data or for detail images. The algorithms are adaptive in the sense
that the resolution does not have to be uniform everywhere but may be variable
in space. For example, in smooth areas a lower resolution can be sufficient to
provide a good representation of the data while in areas with greater variation
a higher resolution is used. The adaptive refinement is controlled by suitable
error indicators which can be specified by the user.
The employed volume visualization algorithm is based on the simultaneous
extraction of several transparent isosurfaces. In this way, interactive
visualization performance of time-varying three-dimensional radar data can be
achieved without the usage of special-purpose hardware (apart from a standard
graphics card).
[1] T. Gerstner, D. Meetschen, S. Crewell, M. Griebel, and C. Simmer. A Case
Study on Multiresolution Visualization of Local Rainfall from Weather Radar
Measurements. In H. Pfister and M. Bailey, editors, Proceedings IEEE
Visualization 2002, pages 533-536. IEEE Computer Society Press, 2002.
DE: 6952 Radar atmospheric physics
DE: 6969 Remote sensing
SC: Special Focus: Advances in Data Acquisition, Management, Analysis and Display [SF]
MN: 2004 AGU Fall Meeting