Unlocking the secrets GNSS Satellite Orbits: the breakthroughs behind IGS Repro3 for ITRF2020
In anticipation of the launch of ITRF2020, the IGS (International GNSS Service) ACs (Analysis Centers) have unveiled the outcomes of their third reprocessing endeavor, known as IGS Repro3. This extensive campaign involves reprocessing GNSS (Global Navigation Satellite System) network data dating back to 1994. A total of eleven analysis centres have contributed to the IGS Repro3 initiative by providing their multi-GNSS products, including the Centre for Orbit Determination in Europe (COD), the European Space Agency (ESA), GeoForschungsZentrum (GFZ), Groupe de Recherche en Géodésie Spatiale (GRG), Jet Propulsion Laboratory (JPL), Massachusetts Institute of Technology (MIT), National Geodetic Survey (NGS), Natural Resources Canada (NRC, EMR), Graz University of Technology (TUG), Université de la Rochelle (ULR) and Wuhan University (WHU). What makes IGS Repro3 distinct from its predecessors is its inclusion of not only GPS and GLONASS data but also data from the Galileo satellite constellation for the very first time. The IGS Repro3 initiative also crowned long-standing research into improving the modelling of GNSS system orbits under the umbrella of the IGS multi-GNSS Pilot Project, especially GLONASS and Galileo, and improving the models accompanying the determination of GNSS orbits.

In our study, we describe the quality of GNSS satellite orbits provided by individual analysis centres and combination orbits provided by the International GNSS Service Analysis Centre (IGS ACC) coordinator at Geoscience Australia. The quality of the combined orbit products has been verified using satellite laser observations. On this basis, we discussed the current developments and challenges in modelling GNSS orbits and summarised the GNSS community open issues in mulit-GNSS combination orbits.

IGS Repro3 orbit combination
The current operational IGS ACC combination software, known as the legacy combination software, has provided combined orbits and clocks since 1993. To combine the multi-GNSS orbits of the Repro3 campaign, a new orbit combination software was developed by the IGS ACC at Geoscience Australia based on similar algorithms to the current software. The multi-GNSS constellation comprises multiple distinct systems, each necessitating individual treatment due to the unique characteristics differentiating each constituent. In this study, thanks to the direct collaboration with Salim Masoumi from Geoscience Australia, we provide extensive discussion on the methodology of the current IGS combination software keeping an eye on the future task of attaching all GNSS constellations to the final portoflio of the IGS flagship products.

Improving SLR validation methodology
Given the use of the SLR technique as the main tool telling us about the quality of individual orbit solutions, we initially looked at a methodology for performing orbit validation using SLR.
After examining the internal consistency of the SLR data, we demonstrated the novel of handling detector-specific biases in the SLR validation process. This approach led to a reduction of up to 15% in the standard deviation of SLR residuals for Galileo FOC satellites, resulting in a more internally consistent SLR dataset that facilitated the detection of orbit modeling issues.

IGS Repro3 orbit quality
Comparing the standard deviations of SLR residuals for the best individual solution versus the combined solution, the results were 13/13, 13/12, 15/17, 17/17, 18/19, and 15/16 for Galileo-FOC, -FOCe, -IOV, GLONASS-K1B, -M, and -M+, respectively. This suggests that the combined solution is of equivalent quality to the best individual AC solutions, underscoring the efficacy of the orbit combination methodology.

To comprehensively evaluate the strengths and weaknesses of the individual solutions and their impact on the combined orbit, the study dissected SLR residuals in various Sun-Earth-satellite geometries (see example for Galileo in the graph below). The analysis revealed that spurious signatures in SLR residuals resulted from differences in orbit modeling. While solutions provided by ESA and TUG demonstrated exceptional quality in modeling Galileo orbits with minimal signatures in SLR residuals, some individual solutions still affected the combined orbit product to some extent. The SLR validation results indicated that ESA orbit products were the most precise among individual solutions for most satellite groups, closely followed by TUG for Galileo-FOC orbits. COD delivered the best GLONASS-K1B orbits, with marginal differences compared to ESA and TUG. Importantly, the IGS combined orbits consistently performed within 10% of the best individual solution, and for GAL-FOCe and GLONASS-K1B, the combined solution proved to be the best among all available options. Therefore, the combined orbit product emerged as the top choice, benefiting from the strengths of the best individual solutions for each satellite type.

Outcome for the orbit combination
Interestingly, there was a minor disparity between the weights assigned to individual ACs in the combination and the quality of their products as assessed externally by SLR. For instance, COD received the greatest weight for Galileo-FOC satellites, but SLR analysis revealed ESA superior accuracy in orbit modeling. This highlights the need to consider SLR validation results as an additional factor in the combination process when developing future combination methodologies.
In conclusion, the IGS combined product serves as a valuable resource for ACs to assess their processing techniques and pinpoint potential weaknesses. Furthermore, it supports ongoing orbit combination efforts by other groups, facilitating cross-comparison and validation of independent multi-GNSS orbit combinations. This collaborative effort contributes to advancing the precision and reliability of GNSS orbit determination techniques.

The conclusions received will be implemented in subsequent tasks carried out by the IGS combination task force, in which we actively participate.

Please find more details in the original paper:
Zajdel, R., Masoumi, S., Sośnica, K. et al. Combination and SLR validation of IGS Repro3 orbits for ITRF2020. J Geod 97, 87 (2023). LINK

This work was funded by the National Science Centre, Poland (NCN) Grant UMO-2019/35/B/ST10/00515 and Wroclaw University of Environmental and Life Scieces (UPWr) Grant N060/0002/23.

Poczta / Logowanie do systemu
GISLab - Laboratorium GIS
Stacja permanentna GNSS 'WROC'
Stacja permanentna GNSS 'WROC'
Wroclaw University of Environmental and Life Sciences
Grunwaldzka 53
50-357 Wroclaw

NIP: 896-000-53-54, REGON: 00000 18 67

Phone +48 71 3205617
Fax +48 71 3205617