Rapid Construction Of Precise Tropospheric Model

Tropospheric delay is the major factors affecting the accuracy of GNSS positioning.The low accuracy of the tropospheric delay empirical model is difficult to meet the demand of real-time precise point positioning.It is of great significance to build a wide-area,real-time and precise tropospheric model using real-time GNSS observa-tions.With the establishment of Crustal Movement Observation Network of China(CMONOC)and National BDS Augmentation Service System(NBASS),a large number of real-time reference station data is available now.At the same time,this is also one of the keys to achieving wide-area real-time precise positioning services.Be-sides,with the development of GNSS meteorology,ground-based GNSS ob-serva-tions can not only be used to obtain two-dimensional tropospheric information such as tropospheric correction and precipitable water vapor,but also can be used to invert the vertical spatial structure of the troposphere atmosphere,to provide new ideas for the construction of GNSS tropospheric models.In this paper,we mainly study the rapid construction of precise tropospheric de-lay model by using ground-based GNSS observations.The purpose is to construct a wide-area real-time high-accuracy tropospheric correction model and to achieve rapid construction of a three-dimensional wet refractive index field using the GNSS ob-ser-vations.We have focused on the basic theory of tropospheric delay derived from pre-cise GNSS processing,the construction method of real-time tropospheric delay cor-rection model,and the realization and verification of large-scale tropospheric three-dimensional tomography.The main research work and results of this paper are summarized as follows:(1)Based on the ERA-Interim reanalysis data provided by ECMWF,the tem-poral and spatial distribution characteristics of ZTD over China are studied.Based on this,a real-time high-accuracy tropospheric correction model RtZTD with considerta-tion of the height uniform and spatial variability of ZTD over China.The RtZTD model was evaluated using GNSS observations from CMONOC.The results show that the model can provide tropospheric delay correction of bias=0.27cm,RMS=1.48cm,and the reference accuracy given by the model can also be used to ob-jectively represent the accuracy distribution of the grid model.(2)The GNSS observations of NBASS are used to verify that the RtZTD model can effectively improve the convergence performance of BDS/GPS PPP.The test re-sults show that compared with the GPT2w model,the RtZTD model can improve the con-vergence time of the BDS-only PPP in vertical and horizontal directions by about 32%and 65%,respectively,and can improve the convergence time by 29%in the ver-tical direction for the GPS-only PPP.(3)The feasibility of three-dimensional wet refractive index inversion using GNSS observations is verified.Based on the GNSS observations from CORS net-work in Guangdong Province,ERA5 numerical prediction products were used as the initial values of the wet refractive index field.The different reconstruction algorithms were used in the tomographic.The results show that the Adaptive Simultaneous Itera-tion Reconstruction Technique(ASIRT)can achieve higher tomographic precision with average bias=-0.06 mm/km and RMS=8.07 mm/km,which can effectively correct systematic biases of forecast products and improve the refractive index field accuracy.