Fiber Surfaces for many Variables

Abstract

Scientific visualization deals with increasingly complex data consisting of multiple fields. Typical disciplines generating multivariate data are fluid dynamics, structural mechanics, geology, bioengineering, and climate research. Quite often, scientists are interested in the relation between some of these variables. A popular visualization technique for a single scalar field is the extraction and rendering of isosurfaces. With this technique, the domain can be split into two parts, i.e. a volume with higher values and one with lower values than the selected isovalue. Fiber surfaces generalize this concept to two or three scalar variables up to now. This article extends the notion further to potentially any finite number of scalar fields. We generalize the fiber surface extraction algorithm of Raith et al. [RBN∗19] from 3 to d dimensions and demonstrate the technique using two examples from geology and climate research. The first application concerns a generic model of a nuclear waste repository and the second one an atmospheric simulation over central Europe. Both require complex simulations which involve multiple physical processes. In both cases, the new extended fiber surfaces helps us finding regions of interest like the nuclear waste repository or the power supply of a storm due to their characteristic properties.

Multi-Field Visualisation Via Trait-Induced Merge Trees

“…Fiber surfaces have then been generalized to general multi-variate data by Raith and Blecha et al [3,31]. In their framework, userdefined geometries in attribute space are called interactors.…”

“…Namely, fiber surfaces and feature level sets (FLS) [18]. Initially, fiber surfaces have been introduced by Carr et al [9] as a method for bi-variate data and have later been extended by Blecha et al [3] to general multi-variate data. Fiber surfaces rely on the intersection of mesh cells and user-given polygons in attribute space.…”

Visualization of Climate Science Simulation Data

“…For the analysis of climate simulation data, this can be used to visualize, for example, relations between several hydrometeorological quantities and the wind field, in order to show the updraft of moisture as the primary power supply of a cumulonimbus cloud system in a high-resolution atmospherical simulation. 14…”