The Approach for Wind Vector Inversion Using Multisource Collocated Radar Measurements

A growing number of spaceborne radar sensors provide increasingly abundant collocated ocean measurements, raising new opportunities for synergetic remote sensing techniques. However, interpreting this diverse observational dataset requires novel approaches to consistently process different data types and invert geophysical fields. Although radar scatterometers continuously monitor ocean winds, joint estimation of speeds from various platforms remains challenging. Additionally, new instruments lack intercallibration algorithms and exhibit varying sampling and resolution characteristics. Thus, joint utilization of such multimodal data is not straightforward.We introduce a framework for synergistic ocean surface wind vector retrieval from heterogeneous radar measurements on multiple satellite platforms. A key point is deriving probabilistic geophysical model functions (PGMFs) mapping observed backscatter to wind speed/direction probability distributions. PGMFs encapsulate inherent backscatter variability from hidden factors. The methodology enables the combined utilization of various radar data and includes recalibration techniques. The wind inversion algorithm combines probabilities based on PGMF of each sensor. The key elements of the approach is demonstrated using CFOSAT and ASCAT data. Initial calibration enables the derivation of the sample PGMF and refinement of ambiguities. Ongoing efforts are to improve the accuracy of the PGMF, take into account non-wind effects, and evaluate multi-sensor performance. The flexible probabilistic technique promises improved wind vector estimation by exploiting various radar missions.