Abstract's details
Open-Sea CryoSat-2 data in SAR and PLRM mode in South East Pacific and North East Atlantic
CoAuthors
Event: 2015 Ocean Surface Topography Science Team Meeting
Session: Instrument Processing: Measurement and retracking (SAR and LRM)
Presentation type: Type Poster
Contribution: PDF file
Abstract:
This work aims to generate, validate and analyse the altimetric geophysical parameters measured by the CryoSat-2 in SAR and PLRM Mode during one full year in the interval 10/2012-10/2013 in the South East Pacific box and North East Atlantic box at distance to coast larger than 10 km (open-sea).
The selection of two aforementionned regions provides us with a very calm region (the Pacific Ocean, with wave height between 1 and 4 meters) and a dynamic region with high seasonal variability (the Atlantic Ocean, with wave height between 0 and 10 meters).
The CryoSat-2 Data have been Delay-Doppler processed as from the FBR (i.e. L1a) to Level 1B and subsequently re-tracked using the SAMOSA's SAR Echo Model (full solution) and a curve-fitting scheme based on Levenberg-Marquard Least Square Minimization Algorithm. The Delay-Doppler processing (L1B) and the re-tracking processing (L2) has been carried out by the EOP-SER Altimetry Team at ESA/ESRIN.
In the open ocean analysis, Hamming function is not applied to the collected burst data. The zero-padding prior range-FFT is applied and an extended vertical swath window is used to mitigate the on board tracker shift errors. Along with multilooked return waveform, also the RIP (Range Integrated Power) is built from the stack and fitted in order to retrieve significant geo-parameters (slope, mean square slope, skewness).
The altimeter wind speed is derived using the wind model used for the Envisat mission and correcting for a sigma nought bias to align CryoSat absolute backscattering to Envisat absolute backscattering. Finally, a sea state bias solution is built for the two areas of interest gridding the residuals elevation from mean sea level in the space SWH-Sigma Nought.
Instantaneous sea surface height (SSH), sea level anomalies (SLA), significant wave height (SWH) and wind speed at 10 meter from sea surface (U10) at 20 Hz and 1 Hz are derived and cross-compared against PLRM L2 data, as generated by TU Darmstadt (TUDa). TUDa L1b PLRM data are retracked using both a conventional Brown model with a Look Up Table (LUT) and a convolutional numerical model, without the approximation of the radar Point Target Response (PTR) as Gaussian function. Numerical retracking in PLRM allows to estimate biases and systematic trends between SAR and PLRM mode and to highlight imprecisions in the SAR re-tracking scheme. Wind speed and sea state bias are derived like in the SAR processing.
SAR and PLRM derived geophysical parameters are cross-validated and validated against numerical models outputs. Performance metrics and plots are produced to assess the quality of the results (as scatter plots, cross-correlations, standard deviations, temporal monthly differences, regression slopes, performance curves, histograms and color-coded density plot).
Given the one year long analysis and the long ground-tracks, averaged wavenumber spectra of each retrieved geophysical parameters are estimated in the Pacific Box. In the North East Atlantic we assess the SAR capability in retrieving wave heights at the low end of the sea state spectrum and the impact of seasonal phenomena (as swell fields) in the SAR data.
The selection of two aforementionned regions provides us with a very calm region (the Pacific Ocean, with wave height between 1 and 4 meters) and a dynamic region with high seasonal variability (the Atlantic Ocean, with wave height between 0 and 10 meters).
The CryoSat-2 Data have been Delay-Doppler processed as from the FBR (i.e. L1a) to Level 1B and subsequently re-tracked using the SAMOSA's SAR Echo Model (full solution) and a curve-fitting scheme based on Levenberg-Marquard Least Square Minimization Algorithm. The Delay-Doppler processing (L1B) and the re-tracking processing (L2) has been carried out by the EOP-SER Altimetry Team at ESA/ESRIN.
In the open ocean analysis, Hamming function is not applied to the collected burst data. The zero-padding prior range-FFT is applied and an extended vertical swath window is used to mitigate the on board tracker shift errors. Along with multilooked return waveform, also the RIP (Range Integrated Power) is built from the stack and fitted in order to retrieve significant geo-parameters (slope, mean square slope, skewness).
The altimeter wind speed is derived using the wind model used for the Envisat mission and correcting for a sigma nought bias to align CryoSat absolute backscattering to Envisat absolute backscattering. Finally, a sea state bias solution is built for the two areas of interest gridding the residuals elevation from mean sea level in the space SWH-Sigma Nought.
Instantaneous sea surface height (SSH), sea level anomalies (SLA), significant wave height (SWH) and wind speed at 10 meter from sea surface (U10) at 20 Hz and 1 Hz are derived and cross-compared against PLRM L2 data, as generated by TU Darmstadt (TUDa). TUDa L1b PLRM data are retracked using both a conventional Brown model with a Look Up Table (LUT) and a convolutional numerical model, without the approximation of the radar Point Target Response (PTR) as Gaussian function. Numerical retracking in PLRM allows to estimate biases and systematic trends between SAR and PLRM mode and to highlight imprecisions in the SAR re-tracking scheme. Wind speed and sea state bias are derived like in the SAR processing.
SAR and PLRM derived geophysical parameters are cross-validated and validated against numerical models outputs. Performance metrics and plots are produced to assess the quality of the results (as scatter plots, cross-correlations, standard deviations, temporal monthly differences, regression slopes, performance curves, histograms and color-coded density plot).
Given the one year long analysis and the long ground-tracks, averaged wavenumber spectra of each retrieved geophysical parameters are estimated in the Pacific Box. In the North East Atlantic we assess the SAR capability in retrieving wave heights at the low end of the sea state spectrum and the impact of seasonal phenomena (as swell fields) in the SAR data.