Weighted mean temperature(Tm)is a key variable in the inversion process of GNSS precipitable water.Based on the measured sounding data from the 18a of the three sounding stations in Xuzhou,Sheyang and Nanjing in Jiangsu Province(2000-2017),this paper analyzes the correlation between Tm and surface temperature Ts,sur-face pressure P,surface vapour pressure e and the logarithmic form of surface vapour pressure lne,establishes the Tm empirical models of Jiangsu Province by using the least square method:the single-factor model Tm1 and multi-factor models Tm2 and Tm3,and evaluates the accuracy of the three localization models by using the sounding data from 2018 to 2019 as the data for future forecast.The results show that the accuracy of the three localization models is all higher than that of the Bevis model,the accuracy of the multi-factor model Tm3 is higher than that of the multi-factor model Tm2 and the single-factor model Tm1,and that the accuracy of the model is 16.8%higher than that of the Bevis model.
关键词
加权平均温度/探空站数据/单因子模型/多因子模型/精度评估
Key words
Weighted mean temperature/Sounding station data/Single-factor model/Multi-factor model/Ac-curacy evaluation
维普
万方数据