应用GRIST 模式对“23·7”华北极端降水的模拟试验  

Simulation experiment on“23·7”extreme precipitation in North China using the GRIST model

作　　者：陈湉茹 张祎 孙溦[4] 李妮娜[5] 刘鸿波 徐幼平 CHEN Tianru;ZHANG Yi;SUN Wei;LI Nina;LIU Hongbo;XU Youping(2035 Future Laboratory,PIESAT Information Technology,Co Ltd,Beijing 100195;State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics,Institute of Atmospheric Physics,Chinese Academy of Sciences,Beijing 100029;Beijing Research Institute,Nanjing University of Information Science and Technology,Beijing 100094;Chinese Academy of Meteorological Sciences,Beijing 100081;National Meteorological Center,China Meteorological Administration,Beijing 100081;Beijing Institute of Applied Meteorology,Beijing 100029)

机构地区：[1]航天宏图信息技术股份有限公司2035未来实验室,北京100195 [2]中国科学院大气物理研究所大气科学和地球流体力学数值模拟国家重点实验室,北京100029 [3]南京信息工程大学北京研究院,北京100094 [4]中国气象科学研究院,北京100081 [5]国家气象中心,北京100081 [6]北京应用气象研究所,北京100029

出　　处：《暴雨灾害》2024年第3期276-287,共12页Torrential Rain and Disasters

基　　金：国家自然科学青年基金项目(42305169)。

摘　　要：全球-区域一体化预测系统(Global-Regional Integrated forecast SysTem,GRIST)的变网格公里尺度非静力模式(以下简称GRIST模式)兼具全球和区域模式的特点,可在无需侧边界条件下进行中期时间尺度预报,有望在局部区域达到与高分辨率的区域模式预报接近的效果。为了检验该模式对极端天气事件的模拟能力,以2023年7月29日—8月2日华北地区特大暴雨过程(以下简称“23·7”极端降水)为例,通过回报试验对GRIST变网格模式对“23·7”极端降水的模拟能力进行了评估;同时利用CMPA降水资料和ERA5再分析资料,采用空间相关系数、TS和BIAS评分等客观分析指标,将国内外GRIST变网格模式对极端降水的模拟效果与目前国内外5种主要业务模式即ECMWF、CMA-GFS全球模式和CMA-MESO、CMA-SH、CMA-BJ区域模式的预报效果进行了比较。结果表明:(1)GRIST及5种业务模式均能模拟或预报出“23·7”极端降水过程的发生。GRIST对该过程中降水最强时段(7月30日08:00—8月1日08:00,北京时,下同)的模拟效果最佳,与观测降水的空间相关系数最高可达0.85,各量级降水的TS评分均达区域模式的预报水平。(2)GRIST模式能准确模拟“23·7”极端降水过程的大尺度环流特征,包括西太平洋副热带高压的位置和范围、北向的水汽输送等,对于表征局地环流演变特征的风暴相对螺旋度的分布和强度也有较好的模拟效果。(3)GRIST模式对不同尺度环流形势的准确模拟,有助于其对降水落区和降水演变特征的精准刻画。GRIST(Global-Regional Integrated forecast SysTem),the kilometer-scale variable-resolution(VR)model,combines the virtues of both global and regional models.It can provide medium-term forecast products without lateral conditions while it is expected to reach a forecast level similar to that of high-precision regional models.To evaluate GRIST’s ability to simulate extreme weather events,this study uses the VR version of GRIST to conduct a hindcast experiment on the extremely heavy rainfall event that occurred in North China from July 29 to August 2,2023.The simulation results are evaluated against satellite-gauge merged precipitation analysis(China Merged Precipitation Analysis,CMPA)and global reanalysis(the fifth generation of the European Center for Medium-Range Weather Forecasts atmospheric reanalysis of the global climate,ERA5)data,and are also compared with five other operational numerical models(including global models,ECMWF and CMA-GFS,as well as regional models,CMA-MESO,CMA-SH,and CMA-BJ).Results indicate that both GRIST and five operational models can reproduce the occurrence of this heavy rainfall event at various degrees.From 08 BT 30 July to 08 BT 1 August,the period with the strongest precipitation during the event,the simulation performance of GRIST is optimal.The spatial correlation coefficient between GRIST and CMPA can reach up to 0.85,and the TS scores of precipitation at various intensities are at the levels of regional model forecasts.GRIST accurately simulated the large-scale circulation characteristics during this process,encompassing the location and extent of the Western Pacific Subtropical High,northward water vapor transport.Furthermore,it demonstrated a commendable capability in simulating the distribution and intensity of storm-relative helicity,a key indicator of the local circulation.The precise representation of circulation conditions at various scales by GRIST contributes significantly to its ability to characterize precipitation areas and the evolution of precipitation.

关 键 词：数值模拟 极端降水 变分辨率模式 全球模式 

分 类 号：P435[天文地球—大气科学及气象学]