HR: 0830h
AN: NG11A-0167    [PDF]
TI: Visualization of Complex Multiscale Phenomena at Subduction Zones
AU: * Rudolph, M L
EM: max@msi.umn.edu
AF: Department of Geology, Oberlin College, Oberlin, OH 44074 United States
AU: Gerya, T V
EM: gerya@msi.umn.edu
AF: Institut f\"{u}r Geologie, Mineralogie und Geophysik, Ruhr-Universit\"{a}t Bochum, Bochum, 44780 Germany
AU: Gerya, T V
EM: gerya@msi.umn.edu
AF: Institute of Experimental Mineralogy, Russian Academy of Sciences, Chernogolovka, Moscow, 142432 Russian Federation
AU: Yuen, D A
EM: davey@krissy.geo.umn.edu
AF: Minnesota Supercomputing Institute and Department of Geology and Geophysics, University of Minnesota, Minneapolis, MN 55455 United States
AU: DeRosier, S
EM: sderosie@msi.umn.edu
AF: Minnesota Supercomputing Institute and Department of Geology and Geophysics, University of Minnesota, Minneapolis, MN 55455 United States
AB: Increases in the computational power of modern supercomputers have led to a gap between the resolution of digital simulations and the resolution of conventional display devices. This problem is compounded by the increasing size of datasets from simulations and the bandwidth constraints of the Internet. In our simulations, we model a subduction zone using at least 30-50 million markers that track data across many different fields such as temperature, viscosity, density, and chemical composition. We found commercially available software to be insufficient for our visualization needs and so we were driven to develop a new set of tools tailored to high-resolution, multi-aspect, multiscale simulations and adaptable to many other applications in which large datasets are prevalent. In order to address this gap in visualization techniques, we have developed solutions for remote-visualization and for the visualization of locally-stored data. Our remote visualization solution is a web-based, zoomable image service (WEB-IS) that requires minimal bandwidth while allowing the user to explore our data through time, across many properties, and through different spatial scales. For local visualization, we propose the use of bandwidth-intensive, high-resolution display walls with at least 10 million pixels to perform parallel visualization in order to best understand causal and temporal relationships between multiple properties in a simulation.
UR: http://tomo.msi.umn.edu/~max/web-is/
DE: 3200 MATHEMATICAL GEOPHYSICS (New field)
DE: 3210 Modeling
DE: 3299 General or miscellaneous
DE: 8121 Dynamics, convection currents and mantle plumes
SC: Nonlinear Geophysics [NG]
MN: 2003 Fall Meeting