Combined orbits and clocks from the IGS 2nd reprocessing

In early 2015, the Analysis Centers (ACs) of the International GNSS Service (IGS) completed their second reanalysis of the full history of globally distributed GPS and GLONASS data collected since 1994. The suite of reprocessed AC solutions includes daily product files containing station positions, Earth rotation parameters, satellite orbits and clocks. This second reprocessing—or repro2—provided the IGS contribution to ITRF2014; it follows the successful first reprocessing, which provided the IGS input for ITRF2008. For this poster, we will discuss the newly combined repro2 GPS orbits and clocks. We also revisit our previous analysis of orbit day-boundary discontinuities with several significant changes and improvements: 1) Orbit discontinuities for the contributing ACs were studied in addition to those for the IGS repro2 combined orbits. (2) Apart from homogeneous reprocessing with updated analysis models, the main difference compared to the IGS Final operational products is that NOAA/NGS inputs were not submitted for the IGS reprocessing, yet they contribute heavily in the operational orbits in recent years. (3) Also, during spring 2016, the ESA modified their orbit model so that it is no longer consistent with the one used for reprocessing. A much longer span of orbits was available now, up to 11.2 years for some individual satellites, which allows a far better resolution of spectral features. 4) The procedure to compute orbit discontinuities has been further refined to account for extrapolation edge effects, improved geopotential fields, and to allow for spectral analysis of a longer time series of jumps. The satellite position time series used are complete enough that linear interpolation is necessary for only sparse gaps. So the key results are based on standard FFT power spectra (stacked over the available constellation and lightly smoothed). However, we have also computed Lomb-Scargle periodgrams to provide higher frequency resolution of some spectral peaks and to permit tests of the effect of excluding eclipse periods.