数据融合的对流层天顶延迟估计方法

为提高对流层天顶延迟(Zenith tropospheric delay，ZTD)的估计性能，提出了基于数据融合的ZTD估计方法。估计干延迟采用Saastamoinen模型，估计湿延迟采用Askne模型，地表气象测量设备提供给两模型所需的气压、温度以及水汽压，Askne模型所需的加权温度、温度变化率和湿度变化率由全球气压和温度2w(Global pressure and temperature 2w，GPT2w)模型提供。当气象测量设备不可用时，上述所有气象参数均来自于GPT2w模型。利用国际GPS服务(International GPS service, IGS)提供的数据进行验证，结果表明：当地表测量设备存在时，所提方法较Saastamoinen模型提高了8mm；当全部气象参数来自GPT2w时，本方法较GPT2w+Saastamoinen模型提高了8.1mm；对于季节分明的测站，误差趋势同样具备季节性；对于海拔高和气候干燥的的测站，估计误差较小。

Estimating Zenith Tropospheric Delay Based on Multi Source Data#br#

To improve the estimation performance of zenith tropospheric delay (ZTD), the way based on multi-source data is developed in this paper. The hydrostatic and wet component are estimated by Saastamoinen model and Askne model, respectively. The surface meteorology employed by those two models are measured by related devices, global pressure and temperature 2w (GPT2w) model provides the weighted mean temperature, lapse rate of temperature and water vapor decrease factor for Askne model. Once surface meteorological measurements are absent, all meteorological parameters are estimated by GPT2w. Proposed models are tested through the data provided by international GPS service (IGS), results indicate that when surface meteorological parameters are available, the annual accuracy is improved by 8mm than Saastamoinen model. Meanwhile, annual bias of other scheme is decreased by 8.1 mm than GPT2w+Saastamoinen model. For most stations, the bias shows seasonal characteristics, high elevation and dry weather can bring less bias.