| Precipitable water vapor plays a key role in the global hydrologic cycle and in climatic change.Mastering the spatial and temporal distribution of precipitable water vapor and its multi-scale variation characteristics are important for revealing some atmospheric and climate changes.Firstly,it is necessary to obtain precipitable water vapor information with high accuracy,high temporal and spatial resolution.Global Positioning System(GPS)has become one of the important means of remote sensing precipitable water vapor information nowadays.Intensive GPS stations provide a lot of data for the study of precipitable water vapor distribution and multi-scale changes.The use of GPS to obtain water vapor requires accurate weighted mean temperature(T_m)and pressure(P_s)information.For GPS stations with insufficient meteorological data,how to accurately determine these two meteorological parameters becomes the key to ground-based GPS precipitable water vapor research.In this study,the relationship between T_m and temperature(T_s),and water vapor pressure(e_s)was systematically studied.RT_m and PT_m regression models related to ground meteorological parameters were established.RT_m model is related to T_s,and PT_m model is related to T_s and e_s.Compared with RT_m model,the accuracy of PT_m model is improved by about 1.0 K.Considering that the residuals of the two T_m regression models have seasonal signals related to the position of the station,RT_m-I and PT_m-I models with a 2.5°×2°(lon×lat)resolution that take into account the seasonal variation characteristics are established based on the RT_m and PT_m models.By comparing with the sounding station data and GGOS data,the RT_m-I and PT_m-I models have good applicability in China,and the accuracy is better than the Bevis T_m-T_s regression model.The ERA-Interim reanalysis data with spatial resolution of 0.5°×0.5°is used to interpolate the pressure and temperature of the GPS continuous stations of the Crustal Movement Observation Network of China(CMONOC).The accuracy of the interpolated meteorological data was verified by the measured meteorological data of the sounding station and the meteorological station,and the influence of the interpolated results on the GPS precipitable water vapor was quantitatively studied.The difference between the GPS precipitable water vapor calculated by the interpolated meteorological parameters and the GPS precipitable water vapor calculated by the meteorological parameters of the meteorological station is less than 0.8 mm.The GPS precipitable water vapor calculated by interpolating meteorological parameters is consistent with the integrated precipitable water vapor obtained by radio sounding.Using the RTm-I model,interpolated ground meteorological parameters and zenith total delay,the precipitable water vapor data of GPS continuous station in the CMONOC from 2015 to 2017 were obtained.the spatial distribution and multi-scale variation characteristics of precipitable water vapor in China were studied.The results show that the annual mean precipitable water vapor distribution over China roughly shows a decreasing trend from southeast to northwest,and the minimum annual mean precipitable water vapor appears in the Qinghai-Tibet Plateau area under the influence of topographic characteristics.Influenced by the monsoon circulation in China,and the maximum precipitable water vapor annual amplitudes are located in the lower reaches of the Yangtze Rive.Semiannual amplitude is smaller than the annual amplitude,the maximum and minimum semiannual amplitudes are located in the mideastern and southwestern China,respectively.The diurnal amplitude of precipitable water vapor is not obvious,and the peak of abnormal diurnal variation usually occurs at 20:00 Beijing time. |
PDF Full Text Request