Geocenter motion based on GNSS: a comparison between low-degree surface load coefficients and network shift parameters
摘要
We apply two methods of geocenter recovery to GNSS data: (1) the center of mass (CM) approach, which employs mass load theory and the temporal displacements of GNSS stations, and (2) the network shift (NSH) approach, which estimates the geometrical shift of the GNSS orbit origin with respect to the reference frame origin realized by well-established station coordinates. We compare both methods based on the network of 372 stations and 19 years (2002–2020) of GNSS solutions from the Center for Orbit Determination in Europe, as part of the International GNSS Service repro3 campaign, incorporating GPS, GLONASS, and Galileo observations. The geocenter time series are estimated with spherical harmonics expanded to degrees ranging from 1 to 8 and decomposed into interannual, seasonal, and intraseasonal components. We found a pronounced reduction in signals associated with draconitic errors in GNSS data employing the CM approach. The Z-component amplitudes of these signals are, on average, approximately three to even six times smaller than those of the NSH approach. Furthermore, the CM solution exhibits signal stability in all geocenter motion components, while the NSH solution shows increased variability, particularly when more GLONASS satellites are included in the solutions. A comparison of the seasonal geocenter motion with external geodetic and geophysical estimates indicates that, in most cases, the solution based on the CM approach provides greater consistency than the direct estimates obtained from the NSH approach. However, the Y component in the CM approach exhibits more than 1 mm less seasonal oscillations compared to most estimates.