NASA scatterometer beam balance using homogenous land targets

Satellite scatterometers are spaceborne radars designed to measure normalized radar cross section ($so$) of the target illuminated by sensor antennas. $so$ measurements are used to retrieve various geophysical parameters, primarily wind speed and direction over sea surface. To remove direction ambiguity, which is inherent in the relation between wind vector and $so$, multi-azimuth observations are necessary. Antennas providing multiple looks must be well calibrated to eliminate beam biases. A simple method is proposed for post-launch scatterometer beam-bias removal. It relies on homogenous land targets with azimuth-isotropic radar response. The method is applied to calculate beam balance corrections for the NASA Scatterometer (NSCAT) using Amazon rainforest as a calibration target. NSCAT flew on board Japan's Advanced Earth Observing Satellite (ADEOS) between Aug. 96 and June 97. Calculated corrections agree well with results from other methods (ice and open ocean measurements). Difference is noticed in corrections calculated separately from ascending and descending passes over Amazon. It is shown that imperfect ADEOS attitude can cause this inconsistency. The beam balance is used to estimate the attitude bias. Mean yaw angle bias of 0.25° between ascending and descending passes is suggested as the main cause of different ascending/descending corrections. Applying this attitude adjustment decreases standard deviation of $so$ measurements and makes ascending and descending corrections consistent.