A Scatterometer Stress Model Function Applicable to Varying Sea States

It has been speculated that scatterometer calibrations for wind speeds are biased for shallow water. A methodology to deal with this problem is discussed. Scatterometers have been calibrated to wind speed (which has been observed in great quantity): however, scatterometers respond to surface roughness, which is a function of wind stress. This stress is a function of wind velocity, surface current, wave characteristics, and atmospheric density. A complication is that wave characteristics change as waves enter shallow water. For current scatterometer algorithms, these changes should result in an overestimation of wind speed. We discuss in general how scatterometers (e.g., NSCAT, QuikSCAT, ASCAT) can be calibrated to measure surface turbulent stress (Bourassa 2006), which is ideal for forcing ocean models. Our previous stress model function for QuikSCAT is improved. The stress can also be combined with oceanographic information to determine the surface wind velocity. Traditional scatterometer model functions (Naderi et al. 1991; Stoffelen 1998; Wentz and Smith 1999) relate the observed radar backscatter to vector winds. A QuikSCAT model function for wind is converted to a model function for surface turbulent stress.