Large-Scale, Realistic Cloud Visualization based on Weather Forecast Data

R. Hufnagel, M. Held (Austria), and F. Schröder (Germany)


Visualization, Natural Phenomena, Cloud Rendering, Weather Model Data, Cloud Classification.


Modern weather prediction models create new challenges but also offer new possibilities for weather visualization. Since weather model data has a complex three-dimensional structure and various abstract parameters it cannot be pre sented directly to a lay audience. Nevertheless, visualiza tions of weather data are needed daily for weather presen tations. One important visual clue for the perception of weather is given by clouds. After a discussion of weather data and its specific demands on a graphical visualization we present an approach to visualizing clouds by means of a particle system that consists of soft balls, so-called meta balls (Dobashi et al. 2000). Particular attention is given to the special requirements of large-scale cloud visualiza tions. Since weather forecast data typically lacks specific information on the small-scale structure of clouds we ex plain how to interprete weather data in order to extract in formation on their appearance, thereby obtaining five vi sual cloud classes. Based on this cloud extraction and classification, modeling techniques for each visual cloud class are developed. For the actual rendering we extend and adapt the metaball approach by introducing flattened particles and derived metaball textures. As shown by our implementation our approach yields a large-scale, realistic, 3D cloud visualization that supports cloud fly-throughs.

Important Links:

Go Back