Kotrupi landslide deformation study in non-urban area using DInSAR and MTInSAR techniques on Sentinel-1 SAR data

Advances in Space Research

Niraj Khatri Chhetri., Sharad Kumar Gupta., & Dericks Praise Shukla.

2022-12-15

The DInSAR (Differential Interferometric SAR) and MTInSAR (Multi-Temporal Interferometric SAR) techniques are used to examine a surface displacement in the Kotrupi Region in this paper. DInSAR and MTInSAR are capable of determining the surface deformation of a rapidly moving landslide. The Vertical Line of Sight (LOS) displacements were estimated with millimeter accuracy using a DInSAR technique followed by phase unwrapping. MTInSAR approaches using APS (Atmospheric Phase Screen) and sparse point processing, on the other hand, measure earth surface displacements and track their trends over time. DInSAR is limited by several difficulties, including phase decorrelation errors, phase unwrapping errors, and atmospheric phase disturbances, all of which reduce the quality of deformation effects. MTInSAR approaches, on the other hand, use data from several different periods within the same temporal baseline. As a result, the effect atmospheric phase disturbances and unwrapping error is greatly decreased when compared to classic DInSAR approaches. After the Kotrupi landslide of August 13, 2017, conventional DInSAR with a single pair of SLC (Single Look Complex) Sentinel-1 SAR images showed a maximum displacement value of about −60 mm to 60 mm in the Kotrupi landslide runoff region. While the advanced MTInSAR results indicate a maximum rate of velocity of about −45 mm/year for the continuous deformation happening in the Kotrupi landslide area. The coherence values along Pathankot National Highway (NH) −154 before and after the landslide were 0.857 and 0.019, respectively. Deformation studies before and after the landslide occurrence are also conducted using a DEM (Digital Elevation Model). Based on the DEM obtained, erosional and depositional deformations of −3.5 m and +3.0 m have been recorded in the Kotrupi landslide runoff area, resulting in 6–7 m change and material loss. The Root Mean Square Error (RMSE) for MTInSAR DEM 0.787 m and DInSAR DEM 1.026 m was estimated after the DEMs produced from DInSAR and MTInSAR processing were compared with TanDEM-X. MTInSAR produces much better results than DInSAR in non-urban areas.