CoAuthors

Jean-Damien Desjonquères (CNES, France); Nicolas Picot (CNES, France); Jean-Christophe Poisson (CLS, France); Pierre Thibaut (CLS, France); Sophie Peyridieu (CLS, France)

In 2015, the Jason-3 mission will ensure the continuity with Jason-2, providing long-term and reliable measurements of the ocean surface topography. The main instrument onboard Jason-3 is the radar altimeter Poseidon-3B which inherits from Poseidon-3 embarked onboard Jason-2. Thanks to the exponential increase in computing power, it is now possible to replace the Gaussian impulse response approximation in the Brown model by its real measurement. The operational retracking algorithm evolves thus towards a numerical form which allows to account for the real altimeter Point Target Response (PTR) and provides estimates of the highest quality, directly usable at 20Hz. Among many benefits of this solution, the most significant are the estimation improvements in weak waves regions, taking account of the natural long-term instrumental drift of the PTR or the possibility to perform a maximum likelihood estimator instead of a least square estimator.
In this study, a complete assessment of the numerical retracking is performed on Jason-2 data. Its performances are analyzed in terms of noise, bias, dependencies and spectral content for the different geophysical estimates. Finally, a special attention is brought to the determination of the skewness coefficient which may be slightly different depending on how the impulse response is considered (real or Gaussian approximation).