Development of GRAPES_Meso system and its performance

In China, 2001, a state-key Research&Development project was launched to develop a unified numerical weather prediction (NWP) system, this project was co-sponsored by MOST and CMA. One of the major objectives of the project is to develop new NWP systems for both operational and research uses. Global&Regional Assimilation and PrEdiction System (namely GRAPES) is the achievement of this project. GRAPES is designed as a multi-scale unified model. The main features of GRAPES_meso includes: (1) full compressible dynamical core with hydrostatic or non-hydrostatic approximation in options ; (2) a semi-implicit and semi-Lagrangian scheme; (3) height terrain following vertical coordinate; (4) full physical package; (5) 3DVAR/4DVAR for data assimilation; (6) modularized and parallelized code architecture; (7) latitude-longitude grid. Now we have a meso-scale system (GRAPES_meso) which cover the East Asia and make 60h-prediction twice a day operationally, with a horizontal resolution of 15km and totally 36 vertical levels. A statistic analysis of real-time forecast performance will be given, and analyses of some high impact weather cases compared with observation will also be discussed.