High-resolution Land/ice Imaging Using Seasat Scatterometer Measurements

In this paper we introduce a new method for obtaining high resolution images (to 4 km) of the land backscatter from low-resolution Seasat-A scatterometer (SASS) measurements. The method utilizes the measurement cell overlap in multiple spacecraft passes over the region of interest and signal processing techniques to generate high resolution images of the radar backscatter. The overlap in the 0 ' resolution cells is exploited to estimate the underlying high resolution surface radar backscatter characteristics using a robust new multivariate image reconstruction algorithm. The new algorithm has been designed to operate in the high noise environment typical of scatterometer measurements. The ultimate resolution obtainable is a function of the number of measurements and the measurement overlap. We describe the method and provide sample results based on SASS data. We describe modifications to future scatterometers which would permit the method to produce 1-2 km resolution over land and ice. The high-resolution scatterometer imaging technique, coupled with the frequent, multiple-incidence angle coverage afforded by the scatterometer, is expected to significantly enhance the utility of sca.tteroineter data in various land, ice, and vegetation studies.