Insar Phase Reduction Using the Remove-Compute-Restore Method

Abstract

Satellite InSAR time series are used to estimate the displacements of radar scatterers. This estimation problem includes the estimation of integer phase ambiguities, which is an ill-posed problem. Consequently, InSAR displacement estimation cannot yield unique solutions and may therefore be significantly biased. Here we show that phase reduction, using a priori information and the remove-compute-restore (RCR) methodology, is a viable way to solve this problem, as it reduces the likelihood of ambiguity errors. We found that application of this methodology to pastures on peat soils leads to a significant improvement in the estimated displacements. We assert that InSAR displacement estimation should always include an explicit statement on the first-order approximations and included assumptions on expected signal smoothness. We anticipate that a more systematic inclusion of the RCR method in standard processing algorithms will lead to more reliable and repeatable results of InSAR analyses.