摘要
利用华南高分辨率模式分析了不同水汽分析场对一次华南前汛期暴雨模式预报效果的影响。分别使用来自NCEP-GFS(Global forecast system)和ECMWF-IFS(Integrated forecast system)两个不同预报系统提供的分析场进行模拟发现,使用IFS模式提供的初始场进行预报能够得到更为合理的降水预报效果,并发现初值中850 hPa高度以下的水汽对本次过程的降水预报结果具有显著影响。由于在华南地区两种初始场内低层的水汽分布存在较大差异,导致了模式预报过程中出现了不同的降水发生机制,最终导致降水预报出现明显的偏差。在GFS分析场的基础上,进一步利用GRAPES-3DVar系统对常规观测资料进行再次同化,发现同化对本次降水过程的前期预报有改进,但是对24 h累计降水的预报改进并不明显。
A high-resolution model of south China is used to study the influence of the initial field on numerical forecast of torrential rain during the annually first flood season in south China. Forecasts with different analysis fields from NCEP-GFS （Global Forecast System） and ECMWF-IFS （Integrated Forecast System） models were compared and it is found that the simulation of precipitation was more reasonable with the IFS initial field, and the moisture under 850 hPa was most sensitive to the precipitation in model. The difference of moisture in the initial field over the South China Sea and Guangxi （ the source of moisture during this precipitation event） is re- markable, which causes different precipitation mechanism finally. As routine observations were assimilated into the GFS analysis field with GRAPES-3DVar, the earlier stage of precipitation forecast was improved while the improvement of 24 hours accumulated precipitation was not so obvious, which is restricted by the distribution of observations and technique of data assimilation. Some possible ways to improve the quality of numerical forecast of rainstorm that happened in south China were discussed.
出处
《热带气象学报》
CSCD
北大核心
2016年第2期155-162,共8页
Journal of Tropical Meteorology
基金
广东省科技计划项目(2011A032100006)"广东省突发灾害性天气应急技术研究中心"
广东省科技计划项目(2012A061400012)共同资助
公益性行业(气象)科研专项(GYHY20120610)"热带环流系统的数值天气预报技术研究"
关键词
初始场
华南前汛期暴雨
水汽
同化
initial field
precipitation during the pre-flood season in South China
moisture
assimilation
作者简介
通讯作者：徐道生，男，浙江省人，助理研究员，主要从事数值预报研究。E-mail：dsxu@grmc．gov．cn
