Construction of geophysical model function for a platform-borne C-band microwave scatterometer

• In-situ experiment is conducted to validate a newly developed C band scatterometer.
• Platform based sigma0 exhibits difference from space-borne ASCAT and CMOD5.n.
• A new GMF for the platform-borne C-band microwave scatterometer was developed.

An in situ field experiment has been conducted during 2011–2012 to test the capability of a newly developed C-band microwave scatterometer which was deployed on an oil platform in South China Sea. Compared to the wide application of the satellite-based microwave scatterometer, a platform-borne microwave scatterometer has been less studied. This paper aims to present our first attempt for construction of a Geophysical Model Function (GMF) for the platform-borne C-band microwave scatterometer. We compared the platform-borne scatterometer data with the space-borne advanced scatterometer (ASCAT) data and C-band GMF-CMOD5.n, and found that obvious differences exist. To quantify such differences, three approaches including a linear regression method, physical model method, and Neural Network (NN) were proposed, and the linear regression method and NN were found to be better solutions with smaller root mean square (RMS) errors, which contributed to a new GMF by superposing on CMOD5.n. The proposed GMF was validated by the backscatter data of April, 2012 and in situ meteorological instrument wind measurements from field experiment with an accuracy of 1.40 m/s for the RMS error. An inversion method based on multi-observations from the platform-borne scatterometer at different incident angles was developed to retrieve wind speed. The work presented in this paper provides a reference method to evaluate the quality of platform-borne scatterometer data.