Abstract
The Geodetic Satellite (Geosat) Follow-On (GFO), Ocean Topography Experiment (TOPEX), and Poseidon altimeter white-noise levels have been evaluated using a technique based on high-pass filtering of 1-Hz sea surface height time series. High-pass filtering removes the geoid and oceanography signals while revealing the random noise. This filtering technique is simpler to use than the repeat-track method, gives essentially the same results, and makes it easier to analyze much larger amounts of data to investigate subtle variations in noise levels. The new noise-level measurements provided here all show stable noise-process characteristics from cycle to cycle, with a linear dependence of the noise level upon significant wave height (SWH). The GFO altimeter noise level is estimated to be 2.5 cm for an SWH of 2 m. The Poseidon noise level is estimated at 2.0 cm for the same value of 2 m SWH. The TOPEX altimeter noise level is 1.8 cm when the dual-frequency ionospheric correction is included; when this noisy correction is not used, the level is reduced to 1.5 cm. Although the dual-frequency ionospheric correction provides an average improvement over the “Doppler orbitography and radiopositioning integrated by satellite” (DORIS) correction, high-frequency noise enters into the dual-frequency correction via noise from the Ku- and C-band ranges. Because the variations in ionospheric refraction are a relatively long wavelength global effect (with strong dependence on latitude), the dual-frequency ionospheric correction should be low-pass filtered before use, and this correction should not be included when estimating the high-frequency noise level of the altimeter.
Corresponding author address: Dr. Ngan Tran, CLS/Space Oceanography Division, 8-10 rue Hermes, 31526 Ramonville St-Agne, France. Email: tran@cls.fr
