Assimilation of hyper-spectral infrared radiances using Principal Component-based radiative transfer model

Modern hyper-spectral remote sensors, such as Cross-track Infrared Sounder (CrIS), Atmospheric Infrared Sounder (AIRS), and Infrared Atmospheric Sounding Interferometer (IASI), have two orders of magnitude more spectral channels as compared to traditional sounders. Data from these sensors contain information on atmospheric temperature, moisture, and trace gas vertical profiles, cloud properties, surface emissivity spectra, and surface skin temperature. On the other hand, it is a challenging task to explore information content contained in these high-spectral resolution spectra due to computational limitations. Assimilation of these hyper-spectral data has been built within the NCEP GSI data assimilation system based upon the Principal Component-based Radiative Transfer Model (PCRTM), developed at NASA Langley Research Center. The use of PCRTM enables the assimilation of full-spectrum radiances in a much more reduced computational cost. It also allows the assimilation of transformed data in PC-space (i.e., PC-scores). The analysis/forecast cycling experiments with the WRF model have been conducted to compare AIRS radiance assimilation using PCRTM and JCSDA's CRTM as observation operator. Very encouraging results were obtained by assimilating hyper-spectral radiances using PCRTM in addition to it computational benefit. Preliminary results of assimilating PC-scores will be also presented.