PPP与NCEP再分析资料在水汽反演的研究

Study on Precipitible Water Vapor Inversion Based on PPP and NCEP Reanalysis Data

对基于PPP水汽反演时站点气象数据(温度、气压)缺失的现象进行了研究,提出一种基于NCEP数据的插值方法。选择4个提供气象文件的IGS站点,将插值方法得到的温度、气压与站点实测数据比较,以实测气象数据反演得到的PWV为真实值,比较不加入气象数据、加入插值气象参数反演得到的PWV与真实值的差异。结果表明,站点插值法得到的气压与真实值差值平均RMS分别为1.39 mbar,插值温度与实测温度差值的平均RMS分别为3.83℃;不加入气象数据反演的PWV与真实值差值平均RMS为2.34 mm,而加入插值气象后,反演PWV与真实值差值平均RMS为0.37 mm,说明插值法可大大提高PWV反演精度,该方法在缺乏实测气象数据时是一种行之有效的补充方式。

We laid emphasis on the phenomenon that lack of meteorological data when precipitible water vapor(PWV) inversion of the station based on PPP, and proposed a new interpolation method based on NCEP reanalysis data. We chose 4 IGS stations, which all supplied meteorological file and compared the meteorological data through interpolation with the observed meteorological data. Taking the PWV which inversed by observed meteorological data as true value, we compared true value and the PWV, which inversed without meteorological data and with interpolation data, and evaluated the accuracy. The results show that the mean RMS of the difference between interpolation pressure and observed pressure is 1.39 mbar. The mean RMS of the difference between interpolation temperature and observed pressure is 3.83 ℃.The mean RMS of the difference between PWV which inversed without meteorological data and true value is 2.34 mm, while PWV which inversed with interpolation meteorological data and true value is only 0.37 mm, which means the interpolation method can improve the accuracy obviously and this method can be a very useful supplement when lack of meteorological data.