The estimation of sea surface winds near and within hurricanes, with the spatial coverage of a satellite radar (scatterometer), is an important objective for public safety. It is also a significant technical challenge when intense rain is present in the scatterometer Field-Of-View (FOV). The presence of rain affects the measured Ku-band normalized radar cross section (NRCS or SIGMA_0) in three ways: rain, cloud and vapor in the atmosphere attenuate the scatterometer signal; rain backscatter augments the signal that comes from the ocean surface; finally, rain hitting the ocean surface induces surface roughening ("splash") that also augments the wind-related signal from the ocean surface. Scatterometer wind retrievals assume that variations in the the rneasured. SIGMA_0 are solely caused by variations in the wind- induced ocean surface roughening. Hence, any rain-related effects have to be accounted for before the scatterometer measurements hi rain can be used to estimate the near-surface wind velocity. The MIDORI-II mission, during 2003, carried five earth-observing sensors including the SeaWinds scatterometer and the Advanced Microwave Scanning Radiometer (AMSR). The latter's six frequency brightness temperatures are collected to derive atmospheric water-related parameters and to measure the sea surface temperature. Since its coverage was closely coincident and collocated with the scatterometer, it provided the opportunity to obtain the precipitation measurements necessary to estimate the attenuation, volume backscatter and surface roughening by the raindrops within the scatterometer beam. Corrections to the scatterometer measurements of ocean surface winds can be pursued with either empirical or physical modeling. While both methods rely on. the AMSR-based geophysical retrievals, they differ in how the information is used. The empirical method compares the observed sigma0 to the NCEP-model-wind-inferred SIGMA_0 to estimate the rain corrections (attenuation and backscatter that combines the rain backscatter and the "splash") as function of the AMSR-derived geophysical parameters. The physical method estimates the three rain effects separately using parametrized relationships between total liquid water, rain rate, surface roughening, volume attenuation and rain backscatter. As such, the physical method does not take into account the NCEP model winds and the produced corrections are more directly related to the AMSR-derived geophysical parameters. The AMSR was designed to measure atmospheric water-related parameters on a spatial scale comparable to the SeaWinds scatterometer (similar to 25km). Optimal estimates of the volume backscatter and attenuation require a knowledge of the three dimensional distribution of reflectivity on a smaller scale comparable to that of the precipitation. Studies selected near the US coastline enable the much higher resolution NEXRAD reflectivity measurements to help evaluate, understand, and improve the AMSR estimates and to conduct research into the effects of different beam geometries and nonuniform beamfilling of precipitation within the field-of-view of the AMSR and the scatterometer.
