Monitoring the mining deformations by Time Series InSAR Integrating DT and SDFPT

High-intensity mining leads to severe ground deformation and secondary geological disasters in coal mines. Persistent Scatters Synthetic Aperture Radar Interferometry (PSInSAR) has strong deformation monitoring capability, but cannot detect enough target points in the mining core and surrounding low-coherence areas. This paper attempts to combine Distributed Target (DT) and Slowly-decorrelating Filtered Phase Target (SDFPT) to improve the density and coverage of deformation monitoring points in mining areas. The Fast Statistically Homogenous Pixel Selection (FaSHPS) and the amplitude dispersion index method were used to select DT and SDFPT candidate points, respectively. Then phase optimization and stability analysis were carried out for the two types of points, and the qualified DT and SDFPT were screened out.  Both kinds of points were then fused, and three-dimensional phase unwrapping was performed. The phase time series were recovered. The spatiotemporal filtering was performed, and the deformation time series and the annual average deformation rate of the fused point set were finally obtained. The 60-scenario Sentinel-1 images covering the Buertai Coal Mine acquired from April 2018 to April 2020 were selected for deformation monitoring. The results show that the density and coverage of deformation points are significantly improved after the fusion of DT and SDFPT, and the maximum deformation level that can be monitored is also increased. There are 5 deformation funnels in the experimental area, and the maximum cumulative deformation reaches -309.76 mm. The influence range of deformation and the difference between time series deformation in different years are highly correlated to mining activities in mining areas.