| Abstract |
Traditional visualization methods for time-varying volume data use volume visualization for the
investigation of individual time steps of a simulation run, possibly animated over time to show the
temporal evolution. For comparative visual analysis of multiple simulation runs within an ensemble
such an approach is neither effective nor efficient. Instead, aggregated information needs to be
extracted and visually conveyed at an overview level.
We present an interactive visual analysis approach for spatio-temporal simulation ensembles
using multiple coordinated views of different aggregation levels. Our approach allows for an
overview of the ensemble’s data value distributions, the detection of main events within the
simulation ensemble, a comparative view of the evolution of all simulation runs within the
ensemble, the investigation of simulation parameters on the simulated output, the detection of
patterns and outliers within the ensemble, and the investigation of spatial distributions of
detected features. We demonstrate the effectiveness of our approach by applying it, among others,
to particle simulation ensembles based on smoothed particle hydrodynamics.
|
