GRAPES模式对2005年登陆强台风预报检验分析

VALIDATION OF TYPHOON FORECASTS BY GRAPES MODEL

针对2005年4个登陆强台风,在台风移动路径、登陆时间、地点、强度等方面,对GRAPES模式的预报能力进行检验评价,并进行了误差原因分析。检验结果显示,GRAPES模式台风中心位置预报24 h平均误差131 km,48 h平均误差252 km。模式预报台风登陆的时间、地点接近实况,预报比较准确;台风登陆时的强度预报偏弱。模式48 h预报准确性略低于24 h。误差原因分析表明,需要加强对非常规观测资料的同化应用,改进模式初始条件;提高模式分辨率,使得模式对中小尺度系统的预报模拟能力增强;改进模式地形的精细处理方法,提高模式对地形影响涡旋移动的模拟能力。

Four landed typhoon cases in 2005 were selected for a numerical simulation study with GRAPES model. The preliminary assessment results of the performance GRAPES model, including the predictions of typhoon track, landfall time, location and intensity etc., are presented in this paper. And the sources of errors were also analyzed. It is showed that 24-hour distance forecast error of the typhoon center by GRAPES model is about 131 km and 48-hour error is 252 km. The models were relatively more skilful for the forecasts of landfall time and locations and the intensity forecasts. On average, the 24-hour forecasts are slightly better than the 48-hour ones. An analysis of data impacts indicates that the assimilation of unconventional observation data is essential in the improvement of the model simulation. It is also illustrated that the model could be improved by increasing model resolution to simulate the mesoscale and fine scale systems and by improving methods of terrain refinement processing.