大气加权平均温度对GNSS PWV精度的影响分析  

Influence of weighted average temperature on PWV accuracy

作　　者：郭敏[1,2] 张捍卫 李鹏杰[1] GUO Min;ZHANG HanWei;LI PengJie(School of Surveying and Land Information Engineering,Henan Polytechnic University,Jiaozuo 454003,China;State Key Laboratory of Geodesy and Earth\s Dynamics,Innovation Academy for Precision Measurement Science and Technology,CAS,Wuhan 430077,China)

机构地区：[1]河南理工大学测绘与国土信息工程学院,焦作454003 [2]中国科学院精密测量科学与技术创新研究院大地测量与地球动力学国家重点实验室,武汉430077

出　　处：《地球物理学进展》2023年第4期1455-1465,共11页Progress in Geophysics

基　　金：国家自然科学基金(41905027,42074002,41931075);河南省科技攻关项目(182102210315);中国科学院精密测量科学与技术创新研究院大地测量与地球动力学国家重点实验室开放基金(SKLGED2020-3-7-E)联合资助。

摘　　要：大气加权平均温度T_(m)是GNSS探测大气可降水量PWV(Precipitable Water Vapor)的关键参数.目前,加权平均温度模型主要包括线性模型和非线性模型.本文基于2011—2015年期间的编号54511北京探空测站的有效探测资料,建立T_(m)与T_(s)的线性和非线性(一阶傅里叶函数、一元二次函数)关系;利用2016年探空站实测资料对所建模型及常用模型进行对比分析,从RMSE、Bias及波动范围评价参数发现T_(m_G)模型精度高于常用模型,而再分析资料ERA-Interim建立的加权平温度T_(m)_ERA模型和新非线性T_(m)模型精度相差甚小,且误差概率分布趋近于正态分布;因此,新建模型能有效避免了通用Bevis全球模型在特定区域导致的区域性精度偏差问题,尤其在探空站缺乏的区域,可以采用ERA-Interim产品建立T_(m)模型.通过对不同T_(m)模型获取IGS站BJFS的PWV结果与相应时间54511探空站的实测PWV数据进行检验,结果表明不同T_(m)模型引起的PWV的偏差Bias范围在[-5,5]mm,均方根误差RMSE的差异甚小,Bias概率趋于正态分布,稳定性较强,尤其T_(m)_ERA、非线性加权平均温度T_(m_F)、T_(m_P)模型引起的PWV的Bias正态分布更强.The atmospheric weighted average temperature is a key parameter for Global Navigation Satellite System(GNSS)to detect atmospheric Precipitable Water Vapor(PWV).Although there are many weighted average temperature models,they can be divided into linear and nonlinear models.Based on the effective sounding data of Beijing 54511 sounding station from 2011 to 2015,the linear and nonlinear(firstorder Fourier function,unary quadratic function)relations between T_(m)and T_(s)are established in this paper.Based on the measured data of 2016 sounding station with parameters RMSE,bias and fluctuation range,the newly established model and common models are evaluated,then the results show that the accuracy of the new T_(m_G)model is higher than the accuracy of other models,that T_(m)_ERA model based on ERAInterim reanalysis data and the newly established nonlinear T_(m)model have little difference in accuracy,that the error probability distribution tends to normal distribution.In this way,regional biases brought by global models such as universal Bevis et al.(1992)are avoided.Then,the PWV of IGS station BJFS is obtained based on different T_(m)models,and the PWV measured by 54511 radiosonde station at the corresponding time is tested.The results show that the Bias range of PWV caused by different T_(m)models is in[-5,5]mm,the difference of root mean square error RMSE is very small,and the Bias probability tends to normal distribution and strong stability.Especially T_(m)_ERA,nonlinear weighted average temperature T_(m_F)、T_(m_P)Bias normal distribution of PWV caused by the models is stronger.

关 键 词：大气加权平均温度 线性和非线性加权平均温度模型 大气可降水量PWV 精度评定 

分 类 号：P407[天文地球—大气科学及气象学]