Visualization for Ocean General Circulation Model via Multi-dimensional Transfer Function and Multivariate Analysis

• Matsuoka Daisuke

  Earth Simulator Center, JAMSTEC

• Araki Fumiaki

  Earth Simulator Center, JAMSTEC

• Kida Shinichiro

  Earth Simulator Center, JAMSTEC

• Sasaki Hideharu

  Earth Simulator Center, JAMSTEC

• Taguchi Bunmei

  Earth Simulator Center, JAMSTEC

Abstract

  Visualization of numerical simulation data is necessary to intuitively understand natural phenomena or structures. However, today's massively parallel supercomputers output huge datasets from large scale high resolution simulations. Due to its size, it is difficult to extract informative data from these datasets. We suggest here the use of combining a feature extraction method and the creation of a transfer function in order to obtain an effective visualization method for ocean general circulation model data. Features such as ocean currents, vortices or water masses are extracted using multivariate analysis which clusters temperature, salinity fluid velocity, etc.<br>

Journal

• Transaction of the Japan Society for Simulation Technology

  Transaction of the Japan Society for Simulation Technology 4 (4), 168-175, 2012

  Japan Society for Simulation Technology

Keywords

• Ocean currents
• Visual data mining
• Multi-dimensional transfer function
• Multivariate analysis
• Cluster analysis

Details

• CRID: 1390001205301890176
• NII Article ID: 130004544054
• DOI: 10.11308/tjsst.4.168
• ISSN: 18835058
• ISSN: 18835031
• Text Lang: en
• Data Source:
  • JaLC
  • CiNii Articles
  • KAKEN
• Abstract License Flag: Disallowed