Computational Analysis for Discovery on the Plasma Waves Observed by Scientific Satellites

Abstract

Akebono is a Japanese scientific satellite, which was launched in 1989 for precise observations of the magnetosphere. Akebono is successfully operated for more than 11 years and the accumulated data amounts to about 1 Tbytes of digital data and about 20,000 audiotapes of analogue data. The VLF instruments on- board the Akebono satellite are designed to investigate plasma waves from a few Hz to 17.8 kHz [1], [2]. In the field of conventional geophysics, research is mainly performed by the following way; (1) a scientist discovers observation result as an evidence of a new theory, or (2) a scientist tries to explain a very unique observation result theoretically. In the analysis of physical phenomenon from data observed by one satellite, it is quite dificult to tell whether we see a temporal change, a spatial change or mixture of them. Hence it is indispensable for investigating these phenomena to examine as large amount of data as possible, although they can be clarified to some extent by event study. Our aim is to develop new computational techniques for extracting the attributes of the plasma waves from the enormous data sets of Akebono and to discover epoch-making knowledge.