|
Activity Number:
|
170
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #304804 |
|
Title:
|
Multivariate Analysis in Planetary Science: Understanding Jupiter's Atmosphere
|
|
Author(s):
|
Irina Kukuyeva*+ and Amy Braverman and Padma Yanamandra-Fisher and Jan de Leeuw and Amy Simon-Miller
|
|
Companies:
|
University of California, Los Angeles and Jet Propulsion Laboratory and Jet Propulsion Laboratory and University of California, Los Angeles and NASA Goddard Space Flight Center
|
|
Address:
|
8125 Math Sciences Bldg., Box 951554, Los Angeles, CA, 90095,
|
|
Keywords:
|
Principal Component Analysis ; Independent Component Analysis ; Jupiter
|
|
Abstract:
|
The structure and dynamics of Jupiter's atmosphere are still unknown. In recent years a wealth of data has been collected from both satellite and ground-based instruments that promises to begin to shed light on the largest gas planet. Data sources include spectra obtained from the Hubble Space Telescope (HST) and NASA's Infrared Telescope Facility (IRTF) on Mauna Kea. In this talk we will review some of the scientific questions and recent progress in using multivariate statistical methods to answer them. We compare results from applying Principal Component Analysis (the standard in the planetary community) and Independent Component Analysis to reduce dimension and uncover data structure. In particular, we assess and compare the stability of these two methods, and comment on their scientific reliability. This talk builds on the talk presented at last year's JSM by the same authors.
|
