Abstract's details
Understanding of Ku-band backscattered signal properties using combined multi-instrumental CFOSAT measurements
CoAuthors
Event: 2022 CFOSAT Science Team Meeting
Session: SCAT product assessment
Presentation type: Type Oral
Contribution: PDF file
Abstract:
The Chinese French Ocean Satellite (CFOSAT) is a space mission dedicated to the simultaneous observation and monitoring of a global ocean sea state and the sea surface vector winds. CFOSAT operates two Ku-band rotating radars: the nadir/near-nadir Ku-band wave scatterometer (SWIM) and the dual-polarization, moderate incidence angle, Ku-band wind scatterometer (SCAT). This instrumental configuration provides regular collocated measurements of radar backscatter in a wide range of incidence angles and the built antenna azimuthal diversity. Two sensors can also be used to improve the quality of the retrieved parameters by combining both data sources. Particularly, this approach can be demonstrated on the improvement of wind vector retrievals using SCAT and SWIM observations.
IFREMER Wave and Wind Operational Center (IWWOC) works on the development of SWISCA and SCA Level 2 CFOSAT products, which are aimed to enable and facilitate the use of dual-instrument collocated data. Both CFOSAT radar data sources represented in a common 25 km resolution spatial grid were measured backscatter (σ0) adjusted to conform the NSCAT-4 Geophysical Model Function (GMF). During correction, different distortion sources like radar antenna swath pattern, satellite platform depointing and noise signal modulation were taken into account.
Present collocated product allows building a multi-parametric GMF to establish the relation between collocated σ0 measurements and various environmental parameters. Such alternative Ku-band GMF was developed using a neural network (NN) approach. The traditional set of GMF variables (wind vector, incidence angle, polarization, ….) could be extended with various additional geophysical parameters: significant wave height, sea surface current vector, sea surface temperature, ice concentration, precipitation rate. To avoid model biasing, special attention had to be addressed to the normalization and uniformization of input values during the learning process. As well, the numerical learning strategy was adapted to reduce the negative impact of numerical weather prediction models (NWP) mismatch on the backscatter measurements regression task.
The resulting NN GMF could be considered as the approximation of the Ku-band radar cross-section as a function of a multi-parameter environment. This function allows separating the impact of different geophysical variables on the backscattering coefficient value. Additionally, it provides a platform for rapid signal calibration and re-adjustment during mission exploitation. We anticipate the implementation of the demonstrated model to extend the existing SCAT data processing with collocated SWIM nadir/near-nadir observations and additional NWP variables.
IFREMER Wave and Wind Operational Center (IWWOC) works on the development of SWISCA and SCA Level 2 CFOSAT products, which are aimed to enable and facilitate the use of dual-instrument collocated data. Both CFOSAT radar data sources represented in a common 25 km resolution spatial grid were measured backscatter (σ0) adjusted to conform the NSCAT-4 Geophysical Model Function (GMF). During correction, different distortion sources like radar antenna swath pattern, satellite platform depointing and noise signal modulation were taken into account.
Present collocated product allows building a multi-parametric GMF to establish the relation between collocated σ0 measurements and various environmental parameters. Such alternative Ku-band GMF was developed using a neural network (NN) approach. The traditional set of GMF variables (wind vector, incidence angle, polarization, ….) could be extended with various additional geophysical parameters: significant wave height, sea surface current vector, sea surface temperature, ice concentration, precipitation rate. To avoid model biasing, special attention had to be addressed to the normalization and uniformization of input values during the learning process. As well, the numerical learning strategy was adapted to reduce the negative impact of numerical weather prediction models (NWP) mismatch on the backscatter measurements regression task.
The resulting NN GMF could be considered as the approximation of the Ku-band radar cross-section as a function of a multi-parameter environment. This function allows separating the impact of different geophysical variables on the backscattering coefficient value. Additionally, it provides a platform for rapid signal calibration and re-adjustment during mission exploitation. We anticipate the implementation of the demonstrated model to extend the existing SCAT data processing with collocated SWIM nadir/near-nadir observations and additional NWP variables.