Temperature and Water Vapor Channel Selection of FY-3E HIRAS II for Application in Numerical Weather Prediction

Hyperspectral infrared atmospheric sounder II (HIRAS II) onboard in Fengyun-3E (FY-3E) early-morning-orbit meteorological satellite is the second-generation hyperspectral infrared (IR) sounder of Chinese Meteorological Administration (CMA). HIRAS II has three consecutive spectral bands named long-wave (LW), medium-wave (MW), and short-wave (SW) band with 1671 more channels than its predecessor HIRAS I. Channel selection is implemented to get the optimal subset for application in numerical weather prediction (NWP) and to explore the contribution of each spectral band. The iterative channel selection method using information content (IC) is applied in temperature and water vapor channel selection based on temperature and humidity sampled profiles respectively. In total, 150 channels were chosen, consisting of 35 LW temperature channels, 65 MW temperature channels, and 50 MW water vapor channels. The temperature channel selection shows that the selected channels in LW band present most information in upper troposphere, while the selected channels in MW band present most information in the mid and lower troposphere. It is found that the humidity of atmosphere profile has large impact on the water vapor channel selection. As the increase of profile humidity, the maximum contribution layer of channel IC shift upward and the degrees of freedom for signal (DFS) profile become sharper and narrower, which results in the variation of channel selection.