Visualization of multi-scale data sets in self-organized criticality sandpile model

Abstract

To study the three-dimensional evolution of a CA model of a sandpile as a function of control parameters, a computational steering approach was used. Sandpile profiles are calculated in real time and displayed in three dimensions simultaneously with the evolution plots of state variables. This represents a generic problem of multiscale structure in data sets derived from nonlinear systems. The software modules were developed in OpenGL/sup TM/ and the Glut/sup TM/ library and will ultimately be implemented in semi-immersive VR on ImmersaDesk/sup TM/. To achieve sufficient speed in real time display, the full data set is represented graphically in a compact form.