On the Mitigation of Phase Bias in SAR Interferometry Applications: A New Model Based on NDWI

Abstract

This work presents a study of the relationship between decorrelation phase in SAR interferogram, soil moisture and water content in vegetation with the aim of mitigating the contribution of decorrelation phase in SAR interferometry (InSAR) estimates of terrain displacements. A methodology for the mitigation of the phase bias based on the temporal variation of the vegetation water content is presented. Decorrelation phases are computed using timeseries of Sentinel-1 images and compared to in-situ measurements of soil moisture. It is shown that soil moisture can partially explain the observed values of decorrelation phases pointing out the role of vegetation water content. A new model is proposed to compute the contribution of vegetation to the decorrelation phase based on the NDWI (Normalized Difference Water Index) index. The methodology is applied to all short temporal baseline interferograms obtained from the timeseries of Sentinel-1 SAR images, using the NDWI maps generated from Sentinel-2 multispectral images. The cumulative displacement is computed by integrating the short temporal baseline interferograms, corrected for the land cover and soil moisture changes. It is shown that the proposed methodology can reduce the variance of estimated cumulative displacement in areas covered by vegetation.