Abstract
We assess the accuracy of JPL's estimated OSTM/Jason-2 Global Positioning System (GPS)-determined orbits based on residuals to independent satellite laser ranging (SLR) data, compared with orbits produced by different software from different data (SLR/DORIS), Geophysical Data Record version C (GDR-C) orbits, and altimeter crossover tests. All of these tests are consistent with sub-cm radial accuracy: high elevation SLR residual standard deviation lies at 6.8 mm, RMS differences from GDR-C in the radial component typically fall below a cm, and altimeter crossovers from JPL orbits have a variance 89 mm2 smaller than altimeter crossovers from GDR-C orbits. Although RMS differences between radial components of different orbit solutions typically lie below a cm, we observe systematic dependences on both time and geography.
The improved precision and accuracy of JPL's OSTM/Jason-2 orbit solutions rely on a new algorithm for applying constraints to integer carrier phase ambiguities. This algorithm is sufficiently robust to improve solutions despite half-cycle carrier phase identification issues in OSTM/Jason-2's BlackJack receiver. Although Jason-1 receiver performance differs, our algorithm should extend to Jason-1 processing (during the time span of nominal GPS receiver operations).
Acknowledgements
We would like to thank the POD teams at CNES, GSFC, and Center for Space Research (CSR) for providing comparison orbits and technical exchanges. Our progress relies on this valuable interaction. The results described in this paper were enabled by a long-term investment in GPS infrastructure and technology by the NASA Earth Science Division. Additional support was provided by the NASA Physical Oceanography Program and the OSTM Project at JPL. We also thank Robert Decarvalho for his help in analyzing orbit differences. The work described in this paper was carried out at the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.