赤外ハイパースペクトルサウンダーのデータ解析と予報モデル同化利用における主成分分析の応用—現状と課題—

抄録

Recent studies on the application of principal component analysis (PCA) to infrared hyperspectral sounder (HSS) data are reviewed in order to promote the utilization of infrared HSS data in remote sensing of the atmosphere and data assimilation for numerical weather forecasting. Infrared HSS sensors are contained in some earth observational and geostationary/polar-orbiting meteorological satellites, primarily to observe the vertical profile of the atmospheric temperature and water vapor content. Because of the large number of channels, the volume of HSS data is enormous and is expected to increase, however, this volume makes data transfer and processing difficult. Data compression techniques, including PCA, are expected to be an effective in reducing the volume of HSS data while maintaining as much observation information as possible. Near real-time spectral data of some HSSs are or will be disseminated in the format of principal component scores (PCS). This article summarizes research examples of the usage of PCA for infrared HSS data and examines the issues involved in developing schemes for it.