Investigating the InSAR Phase Bias in the SBAS Algorithm and Its Effect on Different Landcovers

The Interferometric Synthetic Aperture Radar (InSAR) technique is capable of evaluating and calculating the rate of earth surface displacement over time and across large spatial dimensions with millimeter accuracy. The most common InSAR algorithms for monitoring displacement are SBAS and PS time series. Recent investigations indicate that Multi-Look (ML) Synthetic Aperture Radar (SAR) interferograms with very short baselines are significantly impacted by unwanted short-lived phase artifacts, ultimately impairing the dependability of the InSAR products. In this study, we first calculated the average displacement velocity using a full connection between the interferograms processed with the SBAS algorithm. The results show that the SBAS (full-connection) algorithm is almost free of phase bias and has an 88% correlation with the results obtained from PS-InSAR. By using networks with fewer connections, the processing speed can be increased and the processing volume can be reduced compared to using a full connection SBAS network. Additionally, various networks were processed to assess the amount of phase bias in the SBAS algorithm and to propose an optimal network for this algorithm. The results indicated that the network with eight interferogram connections has the highest correlation with the full SBAS network and the lowest phase bias. Conversely, networks with one to three interferogram connections exhibited the lowest correlation and the highest phase bias. Incorporating interferograms with long time intervals (2, 3, and 4-month) into short-term interferograms (6, 12, and 18-day) has reduced the amount of phase bias. Furthermore, the phase bias varies in soil moisture and different types of vegetation. The results demonstrate a direct correlation between the amount of soil moisture and vegetation and the amount of phase bias; as their levels increase, so does the phase bias.