Abstract's details
Fully focused SAR altimetry studies for optimal data processing and new applications on different surfaces using S6-MF
Event: 2023 Ocean Surface Topography Science Team Meeting
Session: Instrument Processing: Measurement and Retracking
Presentation type: Oral
Sentinel-6MF (S6-MF) embarks a SAR altimeter which evolves significantly from instruments of previous generation (Cryosat-2 and Sentinel-3), offering enhanced performance over its predecessors. It notably features a new operating mode, that allows the quasi-continuous acquisition of echo data, maximizing the fully focused SAR (FF-SAR) processing capabilities [Egido & Smith, 2017]. The S6-MF FF-SAR processing has since been used by several groups, all highlighting the ability of this technique to image reflective water surfaces as never reached before in altimetry, and potentially provide more accurate estimates of the surface height compared to the unfocused-SAR (USAR) approach, thanks to its gain in resolution and improved noise reduction. Despite its very promising results, this processing method is still relatively new and may not be fully exploited. It is therefore important to continue research and studies in order to obtain a more comprehensive and in-depth assessment of the S6-MF FF-SAR performances and to see to what extent it can be optimized and better exploited to gain as much advantage or benefit from it as possible, in particular with a view to implement this processing solution in the operational ground segment of current and future altimeter missions.
To this end, a multi-partner collaboration with a range of expertise in data processing and analysis has been set up seeking to fully exploit the new altimeter imagery capability to enhance applications and potentially define new products over different targets. This work had a two-fold objective:
(a) to determine and provide an optimal configuration for the FF-SAR, using the fast-computing omega-kappa (WK) processor [Guccione et al., 2018], and for different ground target cases (transponder, open-water leads, inland waters, coastal regions and long ocean waves). The tuning of this WK algorithm is discussed, addressing some aspects not discussed before in the literature such as the replica mitigation, the mean square slope (MSS) and its relation with the Doppler bandwidth, and the trade-off between resolution and noise level.
(b) to assess the capability of this optimally configured FF-SAR processing to image surfaces overflown by the S6-MF satellite (represented in two-dimensional FF-SAR images aka radargrams), from which useful and valuable information can be derived for different applications. Several case studies under investigation are listed below :
- In hydrology, with 2D radar scenes, it is now possible to identify the river’s features emerging as bright curves from the darker surroundings and local ponds and locate its position to ease the 1D retracking step and improve the water level accuracy. On-going study aimed to develop an innovative processor able to determine accurate water levels, combining a cutting-edge image processing techniques (for the detection of inland water bodies in S6-6MF FF-SAR data) and a new physically based retracker algorithm, inspired from [Boy et al., 2021]. The results obtained so far will be presented.
- In open-ocean, swell waves signature can appear as amplitude modulation in the FF-SAR radargram that can be used to derive 2D spectra featuring characteristics such as period, amplitude and direction of the swall waves [Altiparmaki et al., 2022]. An interesting case study in the vicinity of coastal areas will be presented illustrating and perfectly demonstrating the FF-SAR’sability to image ocean waves and their evolution along the satellite track, by successfully capturing the various spectral peak periods over time.
- Regarding sea ice leads, we took advantage of existing collocations between S1 SAR images, and S6-MF altimetry data to assess the capability of the S6-MF FF-SAR to detect narrow leads. To this end, various techniques, as the one described in [Longépé et al., 2019], were performed to detect lead signatures in S1 and S6-MF data and their consistency was evaluated.
- Finally, to expand this portfolio of applications, we present a newly developed retracking scheme that aims to improve the determination of sea surface height and other parameter estimates in complex coastal areas (especially in the presence of specular returns). First, the method entails analyzing the radar intensity variations in the S6-MF FF-SAR images to classify point clouds of similar backscattered properties. The second step is the ocean FF-SAR retracking which makes use of the ocean targets located in the previous step to retrack only the useful part of the FF-SAR altimeter waveform. This approach is compared with a shapefile-based method that uses a static coastline and thus does not take into account natural phenomena such as tides or storms.
Back to the list of abstractTo this end, a multi-partner collaboration with a range of expertise in data processing and analysis has been set up seeking to fully exploit the new altimeter imagery capability to enhance applications and potentially define new products over different targets. This work had a two-fold objective:
(a) to determine and provide an optimal configuration for the FF-SAR, using the fast-computing omega-kappa (WK) processor [Guccione et al., 2018], and for different ground target cases (transponder, open-water leads, inland waters, coastal regions and long ocean waves). The tuning of this WK algorithm is discussed, addressing some aspects not discussed before in the literature such as the replica mitigation, the mean square slope (MSS) and its relation with the Doppler bandwidth, and the trade-off between resolution and noise level.
(b) to assess the capability of this optimally configured FF-SAR processing to image surfaces overflown by the S6-MF satellite (represented in two-dimensional FF-SAR images aka radargrams), from which useful and valuable information can be derived for different applications. Several case studies under investigation are listed below :
- In hydrology, with 2D radar scenes, it is now possible to identify the river’s features emerging as bright curves from the darker surroundings and local ponds and locate its position to ease the 1D retracking step and improve the water level accuracy. On-going study aimed to develop an innovative processor able to determine accurate water levels, combining a cutting-edge image processing techniques (for the detection of inland water bodies in S6-6MF FF-SAR data) and a new physically based retracker algorithm, inspired from [Boy et al., 2021]. The results obtained so far will be presented.
- In open-ocean, swell waves signature can appear as amplitude modulation in the FF-SAR radargram that can be used to derive 2D spectra featuring characteristics such as period, amplitude and direction of the swall waves [Altiparmaki et al., 2022]. An interesting case study in the vicinity of coastal areas will be presented illustrating and perfectly demonstrating the FF-SAR’sability to image ocean waves and their evolution along the satellite track, by successfully capturing the various spectral peak periods over time.
- Regarding sea ice leads, we took advantage of existing collocations between S1 SAR images, and S6-MF altimetry data to assess the capability of the S6-MF FF-SAR to detect narrow leads. To this end, various techniques, as the one described in [Longépé et al., 2019], were performed to detect lead signatures in S1 and S6-MF data and their consistency was evaluated.
- Finally, to expand this portfolio of applications, we present a newly developed retracking scheme that aims to improve the determination of sea surface height and other parameter estimates in complex coastal areas (especially in the presence of specular returns). First, the method entails analyzing the radar intensity variations in the S6-MF FF-SAR images to classify point clouds of similar backscattered properties. The second step is the ocean FF-SAR retracking which makes use of the ocean targets located in the previous step to retrack only the useful part of the FF-SAR altimeter waveform. This approach is compared with a shapefile-based method that uses a static coastline and thus does not take into account natural phenomena such as tides or storms.