Impact of Polarization Basis on Wind and Wave Parameters Estimation Using the Azimuth Cutoff From GF-3 SAR Imagery

Optimized Estimation of Azimuth Cutoff for Retrieval of Significant Wave Height and Wind Speed From Polarimetric Gaofen-3 SAR Wave Mode Data

“…Then several studies showed that cross-polarization (HV and VH) information can help to improve SWH retrieval performance [30]. Recent studies have shifted their focus towards exploring the enhancement of combination of multiple polarizations [17], [22]- [23], [25]- [26], [29]. The SAR SWH retrieval entails a multitude of features, and the inclusion of multi-polarization information leads to a multiplication in their number.…”

“…In the field of SAR WS retrieval, NRCS is a very important and classic parameter. In addition to NRCS, λ c is found to be another SAR parameter that exhibits a strong correlation with wind speed (e.g., [23], [42]- [43]). It has been used to retrieve wind speed through linear correlations for wind speeds less than 25 m/s [42].…”

“…Furthermore, this technique requires neither calibration of the SAR data nor any a priori knowledge on wind direction [43]. Therefore, the λ c -based technique is expected to serve as an effective supplement to the NRCS-based technique for wind retrieval, and it has recently gained more attention [23], [43].…”

“…The effectiveness of the λ c -based algorithms in retrieving SWH and WS from SAR is heavily reliant on the accuracy of λ c estimation, which is affected by many factors. The λ c estimation primarily relies on the velocity bunching modulation, intrinsic scene coherence time, and ocean wave spectrum [23]. Furthermore, several studies have demonstrated that it is also affected by SAR pixel spacing, polarization, and incidence angle [23], [43]- [44].…”

“…The λ c estimation primarily relies on the velocity bunching modulation, intrinsic scene coherence time, and ocean wave spectrum [23]. Furthermore, several studies have demonstrated that it is also affected by SAR pixel spacing, polarization, and incidence angle [23], [43]- [44]. Corcione et al [43] analyzed and optimized the SAR pixel spacing, the size of SAR image boxes, and the texture of SAR ocean scenes in terms of λ c estimation.…”