In traditional time-series InSAR analysis, a single empirically defined stable point is often required to stabilize the inherently ill-posed problem of surface displacement estimation. Due to the absence of redundant observations, InSAR errors tend to accumulate in time-series displacement estimates. This study presents a novel time-series InSAR method, termed MPC-InSAR, which integrates multiple prior constraints (MPC) in time-series InSAR analysis for mitigating error accumulation. Firstly, a new two-step algorithm was proposed to automatically identify stable surface points in an area of interest, as most terrestrial regions exhibit relative stability. These stable points and in situ displacement observations (if available) served as multiple prior constraints to correct interferometric unwrapping phase errors. Finally, time-series displacements were solved based on the corrected unwrapping phases using an iterative least-square solver. The MPC-InSAR method was tested using simulation datasets and then applied on Hawaiʻi Island in the USA. The results suggest that the mean absolute error (MAE) and standard deviation (STD) of time-series displacements estimated by the MPC-InSAR can be improved by 59 and 71% compared with the MAEs and STDs of the classical small baseline subset InSAR algorithm, even though the spatiotemporal filtering was not performed in the MPC-InSAR for noise mitigation. The MPC-InSAR method offers a new framework for incorporating multisource prior constraints into InSAR time-series displacements, thereby improving the robustness of time-series displacement detection.
Yang et al. (Wed,) studied this question.