Various reconstruction methods have been used to enhance the spatial resolution of scatterometer data. Most of the image reconstructions are two-dimensional problems, which combine multiple passes of overlapping data over the temporally homogeneous surface, and thus are only suitable for land and ice applications. This paper attempts to address the one-dimensional reconstruction to enhance the azimuth resolution of scatterometer data using a single pass of observations. Since the range resolution determined by the on-board dechirping technique is generally up to several hundred meters, the one-dimensional reconstruction is adequate for certain near real-time ocean applications, such as the development of coastal scatterometer winds. Three well-known reconstruction algorithms, including additive algebraic reconstruction technique (AART), multiplicative algebraic reconstruction technique (MART), and scatterometer image reconstruction (SIR), are evaluated. The spatial resolution and the reconstruction precision resolved by each algorithm are separately analyzed using the local impulse response and Monte Carlo methods. The dependence of the spatial resolution and the reconstruction precision on a variety of parameters, such as the mean backscatter coefficient and its variance, the beamwidth of spatial response function (SRF), and the SRF function type, is evaluated using a simulation framework. In particular, the tradeoff between the spatial resolution and the reconstruction precision is examined for three algorithms. The results show that SIR offers the quickest convergence and lowest noise

Peer Reviewed

11 pages