青海省一次降水过程的数值模式预报及误差分析  

Numerical Model Forecast and Error Analysis of A Precipitation Process in Qinghai

作　　者：石秀云 陈红松 刘成丰 祁雨佳 SHI Xiu-yun;CHEN Hong-song;LIU Cheng-feng;QI Yu-jia(Qinghai Geermu Meteorological Bureau,Geermu Qinghai 816099,China)

机构地区：[1]青海省格尔木市气象局,青海格尔木816099

出　　处：《青海农林科技》2021年第1期77-83,88,共8页Science and Technology of Qinghai Agriculture and Forestry

摘　　要：基于高空实况和数值模式资料,对2020年5月5-6日期间发生在青海省的一次降水天气过程就降水实况、各模式降水落区、降水及温度预报偏差等方面进行了分析,结果表明:EC模式预报相对稳定,GRAPES-GFS与实况更靠近,NCEP模式预报仍然偏弱。对于降水落区预报来说GRAPES-GFS的24h预报降水落区和降水量级更准确,对于降水和最高、最低气温预报准确率,整体来看GRAPES3km准确率最高,其次是GRAPES-GFS,再次是EC-thin,NCEP准确率最低。Based on the actual data of upper air and numerical model data,the precipitation weather process in Qinghai during May 5-6 of 2020 is analyzed in terms of the actual precipitation,precipitation area of each model,precipitation and temperature forecast deviation.The results showed that EC model forecast is relatively stable,EGGF-GFS is closer to the actual situation,and NCEP model forecast is still weak.For the forecast of precipitation falling area,the 24-hour GRAPES-GFS forecast of precipitation falling area and precipitation level is more accurate.For the forecast accuracy of precipitation,maximum and minimum temperature,the GRAPES3km has the highest accuracy on the whole,followed by GRAPES-GFS,then EC-thin,NCEP has the lowest accuracy.

关 键 词：降水 数值模式 降水落区 降水及温度预报偏差 

分 类 号：P456.7[天文地球—大气科学及气象学]