Articles | Volume 8, issue 3
Research article
16 Mar 2015
Research article |  | 16 Mar 2015

Implementation of a GPS-RO data processing system for the KIAPS-LETKF data assimilation system

H. Kwon, J.-S. Kang, Y. Jo, and J. H. Kang

Abstract. The Korea Institute of Atmospheric Prediction Systems (KIAPS) has been developing a new global numerical weather prediction model and an advanced data assimilation system. As part of the KIAPS package for observation processing (KPOP) system for data assimilation, preprocessing, and quality control modules for bending-angle measurements of global positioning system radio occultation (GPS-RO) data have been implemented and examined. The GPS-RO data processing system is composed of several steps for checking observation locations, missing values, physical values for Earth radius of curvature, and geoid undulation. An observation-minus-background check is implemented by use of a one-dimensional observational bending-angle operator, and tangent point drift is also considered in the quality control process. We have tested GPS-RO observations utilized by the Korean Meteorological Administration (KMA) within KPOP, based on both the KMA global model and the National Center for Atmospheric Research Community Atmosphere Model with Spectral Element dynamical core (CAM-SE) as a model background. Background fields from the CAM-SE model are incorporated for the preparation of assimilation experiments with the KIAPS local ensemble transform Kalman filter (LETKF) data assimilation system, which has been successfully implemented to a cubed-sphere model with unstructured quadrilateral meshes. As a result of data processing, the bending-angle departure statistics between observation and background show significant improvement. Also, the first experiment in assimilating GPS-RO bending angle from KPOP within KIAPS-LETKF shows encouraging results.

Short summary
KIAPS has been developing a next-generation numerical weather prediction model and an advanced data assimilation system. We recently implemented a GPS-RO data processing system and then tested the processed data within an ensemble-based data assimilation system. The systems for the data processing and the data assimilation are fully coupled and cycled with a numerical weather prediction model. As a result, data processing and assimilation work properly to give reasonable innovation.