 |
Center for Nonlinear Studies |
The increase in available computing resources allows data scientists in various fields to simulate increasingly complex events of interest. Historically, most visualization techniques have focused on the analysis of the output of simulations, and simulation output has been increasing in complexity. However, advances in computational power now enable domain scientists to address conceptual and parametric uncertainty by running simulations multiple times in order to sufficiently sample the uncertain input space, or the uncertain model space. While these approaches help address conceptual model and parametric uncertainties, the ensemble dataset records a distribution of possible values for each location in the domain. In this talk, we review some solutions to these complex visualization problems produced by these advanced simulations. For example, contemporary visualization approaches that rely solely on summary statistics (e.g., mean and variance) cannot convey the detailed information encoded in ensemble distributions that are paramount to ensemble analysis; summary statistics provide no information about modality classification and modality persistence. In this presentation, we review a number of techniques that address these model and parametric uncertainty analysis problems, and give examples of their usage.
Host: Curt Canada 665-7453