Open AccessComparison of high‐resolution wind fields extracted from TerraSAR‐X SAR imagery with predictions from the WRF mesoscale model
Author(s) -
Thompson Donald R.,
Horstmann Jochen,
Mouche Alexis,
Winstead Nathaniel S.,
Sterner Raymond,
Monaldo Francis M.
Publication year - 2012
Publication title -
journal of geophysical research: oceans
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.67
H-Index - 298
eISSN - 2156-2202
pISSN - 0148-0227
DOI - 10.1029/2011jc007526
Subject(s) - weather research and forecasting model , synthetic aperture radar , mesoscale meteorology , remote sensing , wind speed , satellite , meteorology , radar , inversion (geology) , environmental science , geology , computer science , geography , physics , astronomy , telecommunications , paleontology , structural basin
We discuss the status of our efforts to determine a suitable Geophysical Model Function (GMF) that relates X‐band normalized radar cross section (NRCS) to the near surface wind vector over the ocean. Development of an X‐band GMF has become particularly relevant due to the recent launches of several X‐band satellite synthetic aperture radar (SAR) systems. We concentrate here on SAR data collected by the TerraSAR‐X satellite launched by the German Space Agency in 2007. Inversion of TerraSAR‐X NRCS imagery to wind speed is accomplished using both a simple physics‐based GMF as well as an empirical GMF derived by interpolating more accurately tested C‐ and Ku‐band GMFs to X‐band. We compare the retrieved wind speeds from three TerraSAR‐X scenes with in situ data when available and also with predictions from the Weather Research and Forecast (WRF) model. At vertical polarization, these comparisons show reasonable agreement with in situ data for both the physics‐based and empirical GMFs. At horizontal polarization however, the NRCS from the physics‐based GMF is generally too low, resulting in an under prediction results of the corresponding retrieved wind speeds. An unexpected result from the WRF comparisons is the similarity between the small‐scale structure (on scales ≈ 5–10 km) observed in the SAR imagery and the corresponding WRF output. We believe that this similarity may allow TerraSAR‐X imagery to provide a quantitative measure of the quality of the WRF boundary‐layer parameterization schemes.