GRAPES_TYM改进及其在2013年西北太平洋和南海台风预报的表现

THE IMPROVEMENTS OF GRAPES_TYM AND ITS PERFORMANCE IN NORTHWEST PACIFIC OCEAN AND SOUTH CHINA SEA IN 2013

2013年国家气象中心对GRAPES_TYM进行了改进,包括集成GRAPES-Meso模式相关改进(即基础模式升级)、对流参数化过程由Simplified Arakawa-Schubert(简称SAS)升级为Meso-SAS,并对涡旋初始化方案进行优化。7个典型个例试验统计分析表明,基础模式升级可使72 h平均路径误差减小10%,在基础模式升级的基础上对流参数化方案的升级可使72 h平均路径误差减小20%,涡旋初始化方案的优化可使72 h平均路径误差进一步减小10%。基础模式的升级和对流参数化方案的升级对GRAPES_TYM的预报路径系统右偏有明显改进;基础模式升级对强度预报的影响不明显,Meso-SAS的应用对12~48 h强度预报的改善效果较显著,而台风初始化方案的优化可以减小6~24 h预报时段内的强度预报误差。2013年全年台风回算结果表明,升级后的GRAPES_TYM其48~72 h后的路径预报误差较准业务系统减小15%~20%,最大风速预报误差减小4%~16%。

The regional model GRAPES_TYM for TC intensity and track prediction was developed based on the GRAPES-Mesoin 2009 and was put into operational running in 2012. The results of retrospective forecasting in 2011 and results of operational application in 2012 indicated that there exist systematic northward bias of tracks and positive bias of intensity forecast. The above systematic biases were reduced in 2013 through the improvement of the following three aspects： （1） Integration of the improvement from GRAPES-Meso; （2） upgrading of SAS to MESO-SAS; （3） optimization of vortex initialization scheme.Statistical analysis of 7 cases show that the integration of the improvement from GRAPES-Mesocould reduce 72 h average track error by 10%, but the effectswere not obvious for the intensity prediction. Upgrading of SAS to Meso-SAS could reduce 72 h average track error by another 20%, together with the obvious improvement of 12-48 h intensity prediction. Optimization of vortex initialization scheme further reduces 72 h averagetrack error by 10%and the intensity prediction errors of 6-24 h. The results from the whole cases in 2013 show that 48-72 h average track errors could be reduced by 15%-20% and the average intensity errors of maximum wind speed could be reduced by 4% -16%.