Abstract's details
Validation of Open-Sea CRYOSAT-2 20 Hz Data in SAR Mode in the German Bight Area from 2010 to 2014
CoAuthors
Event: 2014 Ocean Surface Topography Science Team Meeting
Session: Instrument Processing: Measurement and retracking (SAR and LRM)
Presentation type: Type Poster
Contribution: PDF file
Abstract:
Altimetry Data acquired by the CryoSat-2 in SAR Mode in the time span 2010-2014 are processed and validated in the area of the German Bight at distance to coast larger than 10 kilometers (open sea points) against in-situ observations and models. Sea surface height above the reference ellipsoid (SSH), significant waveheight (SWH) and wind speed at 10 meter above the sea surface (U10) are the three validated quantities. In-situ data are sea level, GPS, wind and wave measurements from a network of platforms, buoys and tide gauges maintained by the Waterway and Shipping Administration (WSV). Output from two model simulations are used.
The CryoSat-2 Data in SAR mode have been Delay-Doppler processed from the FBR (Full Bit Rate) Level 1A to Level 1B and subsequently re-tracked using the SAMOSA's SAR Echo Model and a fitting scheme based on Levenberg-Marquard Least Square Minimization Algorithm. Sea surface height, significant wave height and wind speed at 20 Hz and 1 Hz have been derived. The Delay-Doppler processing (L1B) and the re-tracking processing (L2) has been carried out by the EOP-SER Altimetry Team at ESA/ESRIN. Pseudo pulse-limited (PLRM) data derived from CryoSat-2 in SAR mode and provided via the RADS database are compared with geophysical parameters derived from the CryoSat-2 SAR Data to estimate possible biases and trends between SAR mode and PLRM mode and tune up the SAR re-tracking scheme.
The ESRIN SAR data and the RADS PLRM data are now both provided at 20 Hz posting rate.
Efforts are here made to homogenize PLRM/RADS and SAR/ESRIN procedures: (1) a zero-padding operation prior to the range FFT is performed to override waveform range aliasing, (2) the wind speed is derived by using the same wind model used in Envisat mission and by correcting for a small sigma nought bias (-3.04 dbW) to align CryoSat-2 absolute backscattering to Envisat absolute backscattering, (3) the backscatter coefficient sigma nought is corrected for atmospheric attenuation. In SAR a look-up table (LUT) is used to overcome the SAMOSA Model limitation due to the approximation of the instrument Point Target Response with a Gaussian Curve, while any LUT is currently used in RADS..
Performance metrics to measure the quality of the results, scatter plots, cross-correlations, standard deviations, regression slopes and biases between the in-situ and the CryoSat-derived measurements (SSH, SWH, U10) will be presented.
The wavenumber Spectrum of sea surface height, significant wave height and wind speed is computed either for the RADS PLRM dataset either for ESRIN SAR dataset.
Cross-and in-situ validation over two years (2011-2012) have shown that the SAR are more precise, less noisy and more accurate than PLRM data. The overall most significant improvement is made for SWH. For a radius between 10 and 20 kilometers the standard deviation of the difference between altimeter and in-situ data are 8 cm, 12 cm and 1.3 m/s respectively. Comparison with ocean, wave and wind models shows that the ECMWF wind forecast agrees better to the altimeter data than the COSMO-EU windfields do. The altimeter-derived SWH is less consistent with model WW3-ATNE, which does not assimilate altimeter-derived SWH, than with the LSM model, which is nested with a model assimilating altimetric SWH. Also the BSHcmod ocean model is in good agreement with the altimeter sea level, with standard deviation 18 cm, correlation 0.91 and slope 0.87. The above described improvements are expected to improve further precision and accuracy of the SAR data.
The CryoSat-2 Data in SAR mode have been Delay-Doppler processed from the FBR (Full Bit Rate) Level 1A to Level 1B and subsequently re-tracked using the SAMOSA's SAR Echo Model and a fitting scheme based on Levenberg-Marquard Least Square Minimization Algorithm. Sea surface height, significant wave height and wind speed at 20 Hz and 1 Hz have been derived. The Delay-Doppler processing (L1B) and the re-tracking processing (L2) has been carried out by the EOP-SER Altimetry Team at ESA/ESRIN. Pseudo pulse-limited (PLRM) data derived from CryoSat-2 in SAR mode and provided via the RADS database are compared with geophysical parameters derived from the CryoSat-2 SAR Data to estimate possible biases and trends between SAR mode and PLRM mode and tune up the SAR re-tracking scheme.
The ESRIN SAR data and the RADS PLRM data are now both provided at 20 Hz posting rate.
Efforts are here made to homogenize PLRM/RADS and SAR/ESRIN procedures: (1) a zero-padding operation prior to the range FFT is performed to override waveform range aliasing, (2) the wind speed is derived by using the same wind model used in Envisat mission and by correcting for a small sigma nought bias (-3.04 dbW) to align CryoSat-2 absolute backscattering to Envisat absolute backscattering, (3) the backscatter coefficient sigma nought is corrected for atmospheric attenuation. In SAR a look-up table (LUT) is used to overcome the SAMOSA Model limitation due to the approximation of the instrument Point Target Response with a Gaussian Curve, while any LUT is currently used in RADS..
Performance metrics to measure the quality of the results, scatter plots, cross-correlations, standard deviations, regression slopes and biases between the in-situ and the CryoSat-derived measurements (SSH, SWH, U10) will be presented.
The wavenumber Spectrum of sea surface height, significant wave height and wind speed is computed either for the RADS PLRM dataset either for ESRIN SAR dataset.
Cross-and in-situ validation over two years (2011-2012) have shown that the SAR are more precise, less noisy and more accurate than PLRM data. The overall most significant improvement is made for SWH. For a radius between 10 and 20 kilometers the standard deviation of the difference between altimeter and in-situ data are 8 cm, 12 cm and 1.3 m/s respectively. Comparison with ocean, wave and wind models shows that the ECMWF wind forecast agrees better to the altimeter data than the COSMO-EU windfields do. The altimeter-derived SWH is less consistent with model WW3-ATNE, which does not assimilate altimeter-derived SWH, than with the LSM model, which is nested with a model assimilating altimetric SWH. Also the BSHcmod ocean model is in good agreement with the altimeter sea level, with standard deviation 18 cm, correlation 0.91 and slope 0.87. The above described improvements are expected to improve further precision and accuracy of the SAR data.