Performance of KIAPS-LETKF data assimilation system

Korea Institute of Atmospheric Prediction Systems (KIAPS) has successfully implemented Local Ensemble Transform Kalman Filter (LETKF; Hunt et al. 2007) data assimilation system to KIAPS-GM (Global Model) which has fully unstructured quadrilateral meshes based on the cubed-sphere grid that has a strong advantage on the flexibility and scalability in the future high performance computing environment without any singularity. Following a successful evaluation of KIAPS-LETKF data assimilation system with various OSSEs, assimilation of real observations has been performed. As a preliminary experiment, we have assimilated surface and radiosonde observations from NCEP preprocessed data, AMSU-A, IASI, and GPS RO data preprocessed by KPOP (KIAPS Package for Observation Processing), for a month of November 2012. Also, several vertical localization strategies will be evaluated for radiance observations with RTTOV as a forward operator. We will present a current state of KIAPS-LETKF data assimilation system with real observations, and discuss how we further improve the system.