Validation of GNSS orbits using SLR observations Original Research Article

Satellite Laser Ranging (SLR) observations allow for a completely independent validation of orbits derived using microwave measurements as provided by the International GPS Service (IGS). The orbit validation is based on the difference between the observed range (SLR measurements) and the computed range (assuming that the satellite positions are know from the IGS). The resulting range residuals are primarily an indicator for the radial accuracy of the microwave orbits. This validation method is well-known and several validation results have been published in the last 10 years. The IGS analysis center CODE (Center for Orbit Determination in Europe) provides on a routine base daily range residuals for all GNSS satellites observed by both, SLR and microwave techniques. We present results of our recent range residual analysis for two GPS satellites (PRN G05, G06) and three GLONASS satellites (PRN R03, R22, R24). Microwave orbits provided by IGS and CODE are used as well as SLR normal point observations of 13 globally distributed sites. The resulting range residual time series of about 3.3 years, starting in January 2001, for GPS satellites and of about 10 months, starting in June 2003, for GLONASS satellites are analyzed. The comparison of the range residuals with previous analysis results shows the impact of several improvements, i.e., the increasing quality of the microwave orbit as well as the use of new laser retroreflector arrays on GLONASS satellites launched after 1995. The validation of the CODE orbits shows a standard deviation of 2.7 cm for the GPS satellites and of 4.9 cm for the GLONASS satellites. A bias of about −5.8 cm for GPS satellites and of −2.3 cm for GLONASS satellites between the SLR measurements and the microwave orbits was determined. The cause of these biases is not yet fully understood. Pass-specific systematics in the range residuals were found, but it was not possible to assign them to station- or satellite-specific error sources. Deficiencies in microwave orbit modeling might probably cause pass-specific patterns.