The accuracy of high resolution winds from QuikSCAT

The accuracy of the high resolution QuikSCAT wind product was quantified using spatially and temporally collocated 10 m equivalent neutral stability winds, calculated from selected NOAA NDBC buoy measurements. Only buoys that had sample correlation higher than 1.5 were used in this validation, a total of 5704 records. The validation followed the Freilich nonlinear statistical analysis approach. This analysis of the collocated buoy-scatterometer data set yielded the following statistics for wind speed: deterministic offset -0.25 m/s; linear gain 1.02; standard deviation of component errors 1.9 m/s; and RMS error 2.2 m/s. Wind direction errors were more pronounced for lighter winds, typically for winds up to 5 m/s. The RMS directional error for buoy-QuikSCAT pairs for which Δθ<90° is 20.6°.