A review of geodetic and geodynamic satellite Doppler positioning

Abstract

Since 1967, when the U.S. Navy Navigation System was released to the general public, the accuracy of geodetic and geodynamic satellite Doppler positioning using satellite Doppler has improved steadily. This improvement is due mainly to refinements in hardware, refraction, and orbital modeling. Point positioning, using a minimum of 30 passes and the U.S. Defense Mapping Agency (DMA) precise ephemeris, has reached submeter accuracy. Relative positioning among several simultaneously observing stations separated by distances up to 250 km and using a minimum of 30 passes has a typical accuracy below 0.5 m (1σ). Current hardware design and errors in modeling of delays in the wet portion of the troposphere limit the achievable accuracy in relative positioning at the 0.1‐ to 0.5‐m (1σ) level, unless errors in receiver phase and tropospheric refraction delays are continuously measured (calibrated). Also, a crystal oscillator of lower short‐term stability (≥ 5 × 10−12/100 s) increases the Doppler data noise and causes a positive correlation persisting over several minutes. Geodetic and geodynamic Doppler data reductions usually involve meticulous modeling for instrumental, atmospheric, and orbital errors as well as careful Doppler data editing. A short‐arc orbit computation employed in such satellite Doppler reductions requires a gravitational potential expanded to at least degree and order 10 if the orbit shape accuracy is to be comparable to DMA precise ephemeris. Satellite Doppler positioning has had a significant impact on geodetic datum definition and the establishment of national and continental networks of unprecedented accuracy. Satellite Doppler methods, apart from providing one of the most accurate continuous methods of polar motion monitoring, have also been found useful in other geodynamical and ice motion studies. This review concentrates mainly on system performance, models, and computational methods.