15/07/2025

Sea state dependency of scatterometer vertical and horizontal polarized surface returns

  • Wind scatterometers measure the Normalized Radar Cross Section (NRCS) of the ocean surface. This is done by emitting microwaves to the ocean at moderate incidence angles and receiving the returned power. Both vertical (VV) and horizontal polarization (HH) are used, while both microwave polarizations are dominated by different ocean scattering mechanisms. The VV NRCS has been closely related to the ocean surface wind vector and is mainly dominated by the so-called Bragg scattering. The HH NRCS, on the other hand, also depends on the distribution of slopes of the ocean surface, which may depend on non-local sea state conditions, i.e., swell. In this study we exploit a new and sensitive methodology to detect the swell dependency of the HH NRCS. It is found that a HH swell dependency is modest at best and further method refinements are suggested to more accurately characterize such dependency.

    • Graphic displaying the Anomalies of the Maximum Likelihood Estimator Anomalies of the Maximum Likelihood Estimator

  • Figure: Maximum Likelihood Estimator anomalies, given as the ratio between the HH term and the VV term of the normalized MLE, as a function of wind speed for different sea states.

  • Objectives and framework of the study

    In the context of the OSI SAF Visiting Scientist Program, Ana Trindade and Federico Cossu from the Institut de Ciències del Mar (ICM–CSIC) in Barcelona, conducted a study on Vertical and Horizontal components of the Normalized Radar Cross Section of a Ku-band scatterometer. The work took place in 2023 and 2024 and was supervised by Marcos Portabella and Ad Stoffelen with support from Anton Verhoef.

    Many Ku-band scatterometers employ dual polarized measurements (HH/VV) of the NRCS (σ0). In addition, the EUMETSAT Polar System Second Generation (EPS-SG) SCA mission, currently in preparation, will employ a C-band scatterometer with dual polarization capabilities. The HH-polarized radiation scattered from the ocean surface contains both the Bragg contribution (from locally wind-generated sea state) and the wave breaking contribution (from both locally and non-locally generated sea state). The geophysical model function (GMF) used in the wind retrieval processing does not include, at the moment, the non-local wave breaking contribution to the total σ0, which may lead to an increased uncertainty in the wind vector estimation. This study aims at investigating the HH non-local σ0 contributions for different sea state conditions. The search is performed by analyzing the behavior of the wind inversion residuals (the so-called Maximum Likelihood Estimator or MLE) as a function of the measured surface wind speed (U10S) and of different sea state parameters representing, for example, wave height, wave period and wave direction.

  • Conclusions

    We identified three sea state conditions: growing sea, fully developed sea and decaying sea conditions in order to diagnose sea state dependency. The analysis of the wind inversion residuals indicates, in general, no significant MLE anomalies as a function of wind speed. The MLE anomalies are slightly larger for growing seas, suggesting a possible GMF difference for wind-dominated conditions compared to swell-dominated conditions. When compared against wave height and wave period, the MLE anomalies show a more complex behavior that depends both on sea state and on the values of the wave parameters. We found no MLE anomalies instead for any specific value of wave direction (relative to the wind direction).

    Some of the MLE anomalies can be due to ill-classified rain-contaminated measurements that the current quality control (i.e., the KNMI_QC filter) could not identify. A better KNMI_QC filter has been developed in the meantime and a new analysis with the new filter is suggested as a follow on of this study.

  • Benefits for the SAF

    • This work helps to better understand the non-local wave breaking contribution to the HH σ0 measured by dual-polarization scatterometers and its correlation to different sea state conditions. This is relevant for both MetOp-SG SCA and Ku-band scatterometers.
    • These findings may be used to modify/correct the HH GMF and to ultimately reduce the wind vector error.

  • About this study

    Find here the complete repport for this Visiting Scientist Activity: Analysis of the sea state dependency of VV and HH surface returns

     

  • Authors

    Ana Trindade - Institut de Ciències del Mar (Barcelona)

    Federico Cossu - Institut de Ciències del Mar (Barcelona)

    Marco Portabella - Institut de Ciències del Mar (Barcelona)

    Ad Stoffelen - Koninklijk Nederlands Meteorologisch Instituut (De Bilt)

New ticket helpdesk
CAPTCHA