A Fully Analytical SAR Altimetry Retracker for the Estimation of Geophysical Parameters

Alejandro Egido (Starlab Barcelona S.L., Spain)


Chris Ray (Starlab Barcelona S.L., Spain); Salvatore Dinardo (SERCO-ESA, Italy); Marco Caparrini (Starlab Barcelona S.L., Sapin); Bruno Manuel Lucas (DEIMOS-ESA, Italy); David Cotton (Satoc, UK); Jérôme Benveniste (ESA, Italy)

Event: 2014 Ocean Surface Topography Science Team Meeting

Session: Instrument Processing: Measurement and retracking (SAR and LRM)

Presentation type: Type Poster

With the advent of current and upcoming altimetry missions based on SAR processing techniques, it is becoming more significant to use of a flexible processing tool that can provide Level-2 ocean geophysical parameters in a fast and reliable way.

Within the frame of the CryoSat Plus for Ocean (CP4O) ESA project, the SAMOSA-3 fully analytical model, i.e. the current baseline implementation for the Sentinel-3 Level-2 processing, has been updated in order to be able to account for different Level-1 processing approaches. The model updates were focused on the appropriate handling of the energy distribution over the different echoes of the delay-Doppler stack, an application of a Look-Up Table (LUT) for the selection of a variable width Point Target Response (PTR) as a function of SWH, the complete implementation of the SAMOSA-2 model, and an appropriate estimation of the thermal noise from the SAR waveform.

The updated SAMOSA model was integrated within a full waveform retracker, which performs the joint estimation of Sea Surface Height (SSH), Significant Wave Height (SWH), and Sigma_0, by means of an iterative Levenberg-Marquardt minimization algorithm. Within the frame of CP4O, the SAMOSA model was adjusted to the Level-1 processing of the Cryosat Product Prototype (CPP) provided by CNES. The retracker was applied to these data in order to estimate the Level-2 geophysical parameters, which were cross-validated with the SSH, SWH, and Sigma0 provided within the CPP Level-1b product and calculated by means of a numerical retracker.

In order to perform a statistically representative comparison of the retracker outputs, two full sub-cycles of CryoSat data from the South Pacific SAR patch were analysed. It was determined that the geophysical parameter estimation of the CNES numerical and the Updated SAMOSA analytical retrackers are fully consistent and can be considered equivalent: for the 20 Hz product, the bias in the estimation of SSH from both retracker solutions is around 3 mm, with a standard deviation below 1 cm; for SWH, the bias is around 5 mm, with 12 cm standard deviation, for a SWH range between 0 and 8 meters; and finally, the estimation Sigma0 shows an error between both estimations below 0.1 dB in mean and standard deviation. The 1Hz products reduced even further the errors standard deviation by a factor of 4 in all of the geophysical parameters.

Further work entails the validation of this fully analytical retracker with an extended dataset, including also other independent data sources. However, these results already demonstrate the potentiality of the updated SAMOSA retracker for the generation of Level-2 products.

Alejandro Egido
Starlab Barcelona S.L.