Impact of terrestrial reference frame realizations on altimetry satellite orbit quality and global and regional sea level trends: A switch from ITRF2008 to ITRF2014

Abstract

A terrestrial reference frame (TRF) is a basis for precise orbit determination of Earth-orbiting satellites, since it defines positions and velocities of stations, the tracking data of which are used to derive satellite positions. In this paper, we investigate the impact of the International Terrestrial Reference Frame realization ITRF2014, as compared to its predecessor ITRF2008, on the quality of orbits, namely, on root-mean-square (rms) fits of observations and orbital arc overlaps of three altimetry satellites (TOPEX/Poseidon, Jason-1, and Jason-2) in the time interval from August 1992 to April 2015 and on altimetry products computed using these orbits, such as single-satellite altimeter crossover differences, radial and geographically correlated mean sea surface height (SSH) errors and regional and global mean sea level trends. The satellite orbits are computed using satellite laser ranging (SLR) and Doppler Orbitography and Radiopositioning Integrated by Satellite (DORIS) observations of a global network of stations. We have found that using ITRF2014 generally improves the orbit quality as compared to using ITRF2008. Thus, the mean values of the rms fits of SLR observations decreased (improved) by 2.4% and 8.8% for Jason-1 and Jason-2, respectively, but are almost not impacted for TOPEX/Poseidon when using ITRF2014 instead of ITRF2008. The internal orbit consistency in the radial direction (as derived from arc overlaps) is reduced (improved) by 6.6%, 2.3%, and 5.9% for TOPEX/Poseidon, Jason-1, and Jason-2, respectively. Single-satellite altimetry crossover analyses indicate reduction (improvement) in the absolute mean crossover differences by 0.2mm (8.1%) for TOPEX, 0.4mm (17.7%) for Jason-1, and 0.6mm (30.9%) for Jason-2 with ITRF2014 instead of ITRF2008. The major improvement of the mean values of the rms of crossover differences (0.13mm; 0.3%) has been found for Jason-2. Multi-mission crossover analysis shows slight improvements in the standard deviations of radial errors: 0.1%, 0.2%, and 1.6% for TOPEX, Jason-1, and Jason-2, respectively. The standard deviations of geographically correlated mean SSH errors improved by 1.1% for Jason-1 and 5.4% for Jason-2 and degraded by 1.3% for TOPEX. The change from ITRF2008 to ITRF2014 orbits only has minor effects on the estimation of regional and global sea level trends over the 22-year time series from 1993 to 2015. However, on interannual timescales (3-8 years) large-scale coherent trend patterns are observed that seem to be connected to drifts between the origins of the tracking station networks. This leads to the changes in interannual global mean sea level of up to 0.06mmyr ĝ1 for TOPEX, 0.05mmyr ĝ1 for Jason-1, and up to 0.12mmyr ĝ1 for Jason-2, i.e., up to 4% of the corresponding sea level signal based on altimetry for timescales of 3 to 8 years. The respective changes in the regional sea level trend on these timescales are up to 0.4mmyr ĝ1 in the time span from April 1993 to July 2008 and up to 1.0mmyr ĝ1 in the time span from July 2008 to April 2015.