对流层模型评估及其在GNSS精密单点定位中的应用

以水汽辐射计（WVR）精确测定的天顶方向延迟值作为参考，评估Saastamoinen、GPT2、EGNOS、UNB3M四种常用对流层模型在上海地区的改正精度；并将WVR观测值及以上4种对流层模型计算的对流层延迟值作为真值应用到GNSS精密单点定位（PPP）中，评估其对定位精度的影响。比较发现，GPT2模型的对流层改正精度比其余3种要好，其天顶干延迟（ZHD）的偏差均值与中误差分别为-0.11 cm、±0.75 cm，天顶湿延迟（ZWD）的平均偏差与中误差分别为-2.34 cm、±7.67 cm；和传统的PPP结果相比，采用WVR对流层观测值的定位精度提高了16%。

Performance Evaluation of Troposphere Models and Application in GNSS Precise Point Positioning

The troposphere delay is one of the most important sources of error in GNSS positioning. In this paper, the zenith tropospheric delay measured by water vapor radiometer (WVR) is used to evaluate the performance of four types of widely used troposphere models: Saastamoinen, GPT2, EGNOS and UNB3M, in Shanghai. Furthermore, the troposphere delays obtained from WVR and the four troposphere models are applied in GNSS precise point positioning (PPP). Data processes show that GPT2 shows the best performance in troposphere delay correction, as compared with the other three models; the mean ZHD bias of GPT2 is -0.11 cm with RMS of 0.75 cm; the mean ZWD bias is -2.34 cm with RMS of 7.67 cm. The position accuracy of PPP results from the ZHD and ZWD observed by WVR is improved by 16%, as compared with the conventional PPP model, in which the tropospheric parameter is estimated.