CoAuthors

Florence Birol (UPS/Legos, France); Denis Blumstein (CNES/Legos, France); Nicolas Fuller (CNRS/Legos, France)

Altimetric waveforms near the coast are difficult to analyze because of the complexity of echoes returned from land and coastal waters. Most tracking algorithms are optimized for the open ocean; others try to fit the observed signal to one or more signal templates suitable for a particular setting and then estimate the altimetric range from the deduced parameters. We present new developments in which the goal is to use the reflective ground properties as the main parameters for a radar signal inversion problem over. Instead of trying to fit the waveform data to an a priori reference set of waveforms, from which one can deduce the “real” satellite-earth surface range, we try to model the observed waveform in terms of the configuration of the topography and water and their backscattering properties beneath the satellite. For this, we simultaneously fit a set of waveforms over the study region, instead of using the usual approach of optimizing each waveform individually.

The current version of the inversion algorithm uses a set of parametric generators from which several parameters can be chosen to be inverted (topography, sigma-naught, radar penetration). An iterative optimization algorithm based on evolutionary computation is applied on a realistic altimetry waveform model, obtaining suitable parameters of the generators to describe the waveforms acquired by the satellite(s) over the study area.

An application is shown for the NW Mediterranean region, and the range results compared to those of standard retracking and the PISTACH coastal data processing chain.