AMSU资料变分同化及在暴雨数值模拟中的应用研究

在中尺度数值模拟中,利用中国气象科学研究院数值天气预报创新基地开发的GRAPES三维变分同化系统,对AMSU-A/B微波遥感资料进行了同化试验,研究了这种资料在我国夏季暴雨数值预报中的作用。以2003年7月4日的一次暴雨过程为例,分析了同化结果及模拟结果,结果表明:(1)单独同化AMSU-A资料主要改进了初始温度场,而单独同化AMSU-B资料主要改进了初始湿度场;(2)无论是同化AMSU-A资料还是同化AMSU-B资料,对暴雨预报都有一定的改进作用,但是同化AMSU-B资料的改进作用更明显;(3)同时同化AMSU-A/B资料比只使用其中一种资料的模拟效果好,可以更好地改进模拟的暴雨落区及强度,使结果与实况更加接近。使用AMSU资料,对我国夏季暴雨数值预报有改进作用。

Variational Assimilation of AMSU Data and Its Application to Numerical Simulation of Heavy Rainfall

GRAPES three-dimensional variational data assimilation system is used in this paper to assimilate AMSU data,then a process of a heavy rainstorm on 4~5 July 2003 is simulated through the WRF mesoscal model(version 2.0),in order to find the impact of AMSU data on numerical forecast of heavy rainfall in our country.By comparing and analyzing the results of assimilation and simulation respectively,the main conclusions to be drawn from this paper are as follows:(1) AMSU-A data contributes mainly to adjusting the temperature of initial fields,and AMSU-B data adjusts its humidity more distinctly;(2) From the adjusting extent to the results of hard rain simulation,the contribution of AMSU-B data is superior to one of AMSU-A data,and the effect of AMSU-A data seems to be less;(3) After assimilating AMSU-A/B data synchronously,the results of torrential rain simulation are improved more obviously than that of only using AMSU-A data or AMSU-B data in the areas and intensity of heavy rains.The analyses above show that AMSU data may advance the quality of the initial fields and increase the accuracy of heavy rainfall NWP in our country.