GRAPES-GFS模式中系统误差的订正方法研究  

Researchs on Systematic Error Correction Method in GRAPES-GFS Model

作　　者：佟铃[1] 彭新东[1,2] 崔晶 

机构地区：[1]成都信息工程大学大气科学学院高原大气与环境四川省重点实验室 [2]中国气象科学研究院灾害天气国家重点实验室 [3]96427部队

出　　处：《成都信息工程学院学报》2015年第3期264-270,共7页Journal of Chengdu University of Information Technology

基　　金：国家973资助项目(2012CB417200)

摘　　要：采用基于历史资料的模式距平积分订正方法 ANO方法,利用ERA-interim和NCEP FNL资料,对GRAPES-GFS-subdomain模式的预报结果进行订正,消除部分模式系统误差,到达提高天气预报质量的目的。试验说明ANO方法不仅在环流场的预报订正有较为显著的效果,其中位势高度和温度的距平相关系数ACC和均方根误差RMSE订正幅度最为明显。对于2011年7月中旬一次降水的关键条件水汽的强度以及移动路径也有更加接近于真实情况的订正效果,在降水过程前的预报结果中,温度与水汽通量的移动路径订正效果明显,体现出具有实际应用的价值。Using the Anomaly Numerical Correction method with observation（ ANO） based on historical data and anomaly integration,a numerical calibration to the prediction of GRAPES-GFS-subdomain model was performed with the ERA-interim reanalysis and FNL data,aiming at eliminating part of the model system error and improving the weather forecast quality. Experiments show that ANO has significant effects. While the Anomaly Correction（ ANO） and Root Mean Square Error（ RMSE） of the geopotential height and temperature are the most obvious. Not only the prediction error of the circulation field but also the strength of vapor and moving path which are the key conditions of precipitation in JULY 2011 are closer to the correction results of the real situation. Refering to the forcasting results before the precipitation,the effective correction of the temperature and the path of the vapor flux demonstrated the value of practical application.

关 键 词：计算机技术 数值天气预报 数值预报订正 ANO方法 历史资料 GRAPES-GFS模式 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]