Abstract's details
Assessment of the SARM processing sensitivity to swell
CoAuthors
Event: 2018 Ocean Surface Topography Science Team Meeting
Session: Quantifying Errors and Uncertainties in Altimetry data
Presentation type: Type Oral
Contribution: PDF file
Abstract:
The Sentinel-3A satellite was launched on February 16th 2016 as part of the EU Copernicus Programme. The on-board altimeter, SRAL, is based on a principle proposed by Raney (1998): the Synthetic Aperture Radar mode (SARM) also referred to as Delay Doppler Mode. Inherited from Cryosat-2 altimeter, it has been demonstrated that compared to conventional altimetry (or Low Resolution Mode), this technique allows to significantly improve the signal to noise ratio and provide a better spatial resolution in the along-track direction (Boy et al, 2017).
Given its high spatial resolution, the SARM radar altimeter is more likely to be sensitive to sub-mesoscales structures such as swell waves. This outstanding issue is one of the priorities raised by the expert team of the CP4O project (Cotton, 2015). It has also been guided by the recommendations of the OSTST altimetry community and strongly supported by the agencies for Sentinel-3 topography mission and for the further Sentinel-6 satellite, that will also embark a SAR-mode radar altimeter.
The first evidence of impact of swell on SARM altimetry real data was shown by Aouf and Phalippou in 2015. Trough the comparison between the MFWAM model and the Cryosat-2 SARM SWH, they demonstrated that the altimeter measurements are likely biased by ocean swell, and most noticeably for longest waves and swell fields propagating in a direction parallel to the satellite track. More recently, Moreau et al. (2018) were able to go a bit further in our understanding and explanation of the phenomenon, showing a high correlation between the 20-Hz noise of the altimeter-derived estimates and the swell parameters (namely the period, the swell height and the direction of swell propagation), without however clearly determined errors in SARM estimates due to the restricted number of ocean areas operated in this radar mode.
Unlike Cryosat-2, Sentinel-3A operates in full SARM orbit. This global coverage gives the opportunity to better understand and characterize the specificities of the SARM technique using real data. Thus, the objective of this study is to characterize, with a similar approach, the swell effects on SARM measurements. The impact on biases and noise of the altimeter parameters (range, SWH and sigma0) are assessed by using collocation between Sentinel-3A data and WaveWatch3 wave model. Moreover, a specific section is dedicated to the study of the SARM spectral slope observed at sub-mesoscales to demonstrate whether it is correlated to swell occurrences.
Given its high spatial resolution, the SARM radar altimeter is more likely to be sensitive to sub-mesoscales structures such as swell waves. This outstanding issue is one of the priorities raised by the expert team of the CP4O project (Cotton, 2015). It has also been guided by the recommendations of the OSTST altimetry community and strongly supported by the agencies for Sentinel-3 topography mission and for the further Sentinel-6 satellite, that will also embark a SAR-mode radar altimeter.
The first evidence of impact of swell on SARM altimetry real data was shown by Aouf and Phalippou in 2015. Trough the comparison between the MFWAM model and the Cryosat-2 SARM SWH, they demonstrated that the altimeter measurements are likely biased by ocean swell, and most noticeably for longest waves and swell fields propagating in a direction parallel to the satellite track. More recently, Moreau et al. (2018) were able to go a bit further in our understanding and explanation of the phenomenon, showing a high correlation between the 20-Hz noise of the altimeter-derived estimates and the swell parameters (namely the period, the swell height and the direction of swell propagation), without however clearly determined errors in SARM estimates due to the restricted number of ocean areas operated in this radar mode.
Unlike Cryosat-2, Sentinel-3A operates in full SARM orbit. This global coverage gives the opportunity to better understand and characterize the specificities of the SARM technique using real data. Thus, the objective of this study is to characterize, with a similar approach, the swell effects on SARM measurements. The impact on biases and noise of the altimeter parameters (range, SWH and sigma0) are assessed by using collocation between Sentinel-3A data and WaveWatch3 wave model. Moreover, a specific section is dedicated to the study of the SARM spectral slope observed at sub-mesoscales to demonstrate whether it is correlated to swell occurrences.