Validation of Quikscat-Derived Coastal Winds

This study presents a validation analysis of the coastal winds derived from the pencil-beam SeaWinds scatterometer, which flew onboard the US satellite platform QuikSCAT, by means of the winds derived from the European Envisat Advanced Synthetic Aperture Radar (ASAR). QuikSCAT Normalized Radar Cross Sections (NRCSs) are first cor-rected to mitigate land contamination with a method called “noise regularization”. ASAR -derived wind directions are obtained with a Residual Neural Network (ResNet) scheme, which are then used to invert wind speeds using the C-band SAR Geophysical Model Function (GMF) CSAR_Mod2. Preliminary results on nine ASARlQuikSCAT collocations offshore Norway show that when ASAR winds are upscaled to a spatial resolution comparable to that of QuikSCAT-derived winds, the vector Root Mean Square Difference (vRMSD) amounts to 2.9 ms-<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> within 40 km to the coast-line, better than 3.2 ms-<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> for offshore winds.