Delineating Reliable Ground Control Points in SBAS-Insar Analysis with Phase Derivative Variance

Abstract

In the SBAS-InSAR process, the conventional approach of determining ground control points (GCPs) can be error-prone. Thus, a new GCP selection approach using the phase derivative variance is proposed. Analyzing 45 descending-orbit Sentinel-1 datasets between October 2014 and June 2017 over the 2017 Xinmo landslide site, we separately identified ~10,600 and 100 GCP candidates with the conventional and proposed approaches. With four GCPs from the conventional method, the time-series displacement curves do not capture the accelerated deformation related to the landslide event. Corresponding to two sets of GCPs from the proposed approach, the displacement curves change gradually and capture the accelerated deformation. Thus, the approach can efficiently identify a small number of reliable and accurate GCP candidates for the SBAS-InSAR analysis.