New stacking method for removing the SAR sensitivity to swell

François Boy (CNES, France)

CoAuthors

Thomas Moreau (CLS, FRANCE); Pierre Thibaut (CLS, FRANCE); Pierre Rieu (CLS, FRANCE); Aublanc Jérémie (CLS, FRANCE); Nicolas Picot (CNES, FRANCE); Pierre Femenias (ESA, Italy); Constantin Mavrocordatos (ESA, Netherlands)

Event: 2017 Ocean Surface Topography Science Team Meeting

Session: Instrument Processing: Measurement and Retracking

Presentation type: Type Oral

First evidence of swell impact on SAR-mode altimetry has been shown with real data by [Aouf and Phalippou, 2015]. In this study, a comparison of Cryosat-2 SAR-mode derived wave heights with model (MFWAM) has revealed that altimeter measurements are likely biased by ocean swell, and most noticeably for the longest waves and swell fields that are propagating in a direction parallel to the satellite track. During last OSTST held in La Rochelle, France in 2016, several expert groups also emphasized that Sentinel-3 SAR observations are sensitive to swell conditions (in terms of large increase of 20-Hz range noise).

As explained by Moreau et al. [RD 2], the higher along-track resolution provided by SAR altimetry (typically of ~300m for Cryosat-2 and Sentinel-3A SAR mode data) makes it close to the wavelength scales of ocean swell with the consequence that the Gaussian statistics of heights and slopes in the range/Doppler resolution cells are no longer satisfied. This may cause distorted echoes shape and subsequent impacts on parameter retrievals. In addition, the multi-looking process stacks the data within 2,5 seconds which may also disturb the homogeneity of looks within the stack in such swell waves conditions where the ocean surface moves quickly.

ESA and CNES jointly run studies with the support of CLS to test different approaches with the objective to reduce the swell impact on SAR mode performances. In this presentation, we focus on a new stacking method, originally developed and used by Thales Alenia Space [Phalippou and Demeester, 2011] to assess the Cryosat-2 SAR mode performances. After the beam sharpening, using an along track FFT as it is currently performed), all the beams contained in a radar cycle (4 bursts of 64 beams) are, first, range migrated with respect to the nadir beam, and then averaged. This approach brings the benefit to stack the data within a much shorter time laps (~50ms) than in the current multi-looking process.

This new stacking method has been evaluated on one cycle of Sentinel-3 SAR mode and performances have been assessed and compared to operational products. We demonstrate that the swell sensitivity is totally removed and that the range/SWH range noises are even improved by 15 to 40% depending on wave heights. Spectral analyses on SLA and SWH also confirm that the SAR mode finer spatial resolution is maintained in Open Ocean. Given these very promising results, the relevance of using such method for Sentinel-3 operational processing is discussed too. Moreover, this stacking approach is already part of the incoming Sentinel-6 science processors (LR-RMC).

RD 1 L. Aouf (Meteo France), L. Phalippou (TAS-F), “On the signature of swell for the Cryosat-2 SAR-mode wave data”, Oral presentation, Ocean Surface Topography Science Team 2015, October 19-23
RD 2 T. Moreau et al., “Effect of swell and wind-waves on the altimeter-derived estimates: Analyzing real and simulated data”, Oral presentation, Ocean Surface Topography Science Team 2016, November 1-4
 

Oral presentation show times:

RoomStart DateEnd Date
Symphony Ballroom IV Tue, Oct 24 2017,09:45 Tue, Oct 24 2017,10:00
François Boy
CNES
France
francois.boy@cnes.fr