Sea Surface Wind Speed Inversion Using the Low Incident NRCS Measured by TRMM Precipitation Radar

• Published in: IEEE journal of selected topics in applied earth observations and remote sensing 2016-11, Vol.9 (11), p.5262-5271
• Main Authors: Bao, Qingliu, Zhang, Youguang, Lang, Shuyan, Lin, Mingsen, Gong, Peng
• Format: Article
• Language: English
• Subjects: Accuracy; Atmospheric precipitations; Buoys; Correlation coefficient; Correlation coefficients; Empirical analysis; Empirical GMF; Inversion; low incident angle; Maximum likelihood estimates; Meteorological satellites; Ocean surface; Oceanography; Precipitation; Radar; Radar cross sections; Radar measurements; Rain; Rainfall; Satellites; Scatterometers; Sea measurements; Sea surface; Spaceborne radar; Standard deviation; Surface water waves; Surface waves; Surface wind; TRMM satellite; Tropical climate; tropical rainfall measuring mission precipitation radar (TRMM PR); Water pollution effects; Wind; Wind speed; wind speed inversion
• ISSN: 1939-1404; 2151-1535
• DOI: 10.1109/JSTARS.2016.2581215

As the launch of radars that observe the ocean at low incident angles, such as the Precipitation Radar (PR) on Tropical Rainfall Measuring Mission (TRMM) satellite and the Surface Wave Investigation and Monitoring (SWIM) on China France Oceanography SATellite, more and more normalized radar cross sections (NRCS) σ 0 of ocean surface in low incident angles are obtained. In this paper, the ocean surface NRCS of PR, when there are no rain, are used for wind speed retrieve. The sea surface wind speeds are retrieved by the Maximum Likelihood Estimate. First, the data preprocessing and wind speed inversion method of PR 2A21 data are introduced. In order to improve the inversion accuracy, the empirical GMF at low incident angle is established, using the PR NRCS and QuikScat wind speed. The accuracy of retrieved wind speed is analyzed by comparing it with the buoy, Advanced Scatterometer (ASCAT), and QuikScat. The retrieved wind speeds have a standard deviation of about 1.5 m/s when compared with the buoy, and a standard deviation of 1.1 m/s when compared with ASCAT. The year averaged global wind speed maps of PR as also calculated and compared with that of QuikScat, ASCAT, and TRMM Microwave Imager. Their spatial structures of wind speed are very consistent and correlation coefficient is greater than 0.99. At last, the standard deviation and bias of retrieved wind speeds versus incident angles and wind speeds are also analyzed.