SLR validation of the IGS Repro3 orbits for ITRF2020

The International GNSS Service (IGS) analysis centers (ACs) for the third time issued the results of the reprocessing campaign (IGS Repro3) of all the GNSS network solutions backward starting from 1994. The Repro3 products provided the IGS contribution to the latest International Terrestrial Reference Frame realization ITRF2020. Unlike the previous reprocessing campaigns, the IGS Repro3 includes for the first time not only GPS and GLONASS but also the Galileo constellation. 

In this study, we show results from the GLONASS and Galileo orbit validation of different IGS ACs, as well as the combined orbits generated by the IGS Analysis Centre Coordinator (IGS ACC) at Geoscience Australia. Individual multi-GNSS orbit solutions were provided by Center for Orbit Determination in Europe (COD), European Space Agency (ESA), GeoForschungsZentrum (GFZ), Centre National d’Études Spatiales (GRG), Graz University of Technology (TUG) and Massachusetts Institute of Technology (MIT). Different IGS ACs use different orbit modeling strategies, e.g., estimating specific empirical orbit parameters and using or not using the a priori box-wing models for the Galileo satellites. The individual IGS Repro3 contributions have been combined by the IGS ACC combination software using a robust algorithm and a satellite-based weighting approach considering different qualities of GNSS orbits provided by different IGS ACs. We summarized the recent progress in GNSS precise orbit determination focusing on the impact of the orbit modeling aspects on the systematic effects in the SLR validation results, e.g., solar radiation pressure modeling and observable types.

For all the defined Galileo and GLONASS satellite subtypes, the combined solutions do not perform worse (to the 1 mm level) than the best individual AC solutions for a specific satellite type. In terms of the standard deviation of SLR residuals, ESA, MIT, and TUG deliver the best Galileo-FOC orbits and ESA provides the best Galileo-IOV orbit solutions. Finally, ESA and TUG provide the best GLONASS products. Searching for patterns in SLR residuals for different satellite-Sun-Earth geometries reveals that some issues in the orbit modeling for the Galileo-FOC and IOV satellites still need to be fully diminished. The pattern dominates in COD, GFZ, and GRG solutions, and is mostly diminished in ESA and TUG solutions. The characteristic pattern is also noticeable in the combined solutions, but with a reduced magnitude.