[M-GI26] Data assimilation: A fundamental approach in geosciences

[MGI26-P01] Development of LETKF system based on the JMA’s ASUCA-Var

*Koji Terasaki¹, Yasutaka Ikuta¹, Takuya Kawabata¹ (1.Meteorological Research Institute, Japan Meteorological Agency)

Keywords:Data assimilation, Ensemble Kalman filter, Numerical weather prediction

We have been developing an ensemble-based data assimilation system using the Japan Meteorological Agency’s (JMA) operational meso-scale model (ASUCA, Asuca is a System based on a Unified Concept for Atmosphere). The JMA operates the ASUCA-Var data assimilation system, which incorporates an incremental 4D-Var as an operational numerical weather prediction (NWP). We incorporated a local ensemble Kalman transform filter (LETKF) into ASUCA. We use observation operators of the ASUCA-Var to compare the results by assimilating the same observations in the ASUCA-Var. Our system consists of the ensemble members with a horizontal resolution of 15 km and the deterministic member with that of 5 km. The analysis increment is added to the deterministic member at the start of the assimilation window, and we perform a 5-km deterministic forecast, which is regarded as the analysis. The initial ensembles for the next data assimilation cycle are generated by adding the analysis ensemble perturbations to the deterministic analysis. We have performed preliminary data assimilation and forecast experiments for a typhoon case with 50 ensemble members. In this presentation, we will introduce the overview of the system and the latest results.