Surface-deformation monitoring in the permafrost regions over the Tibetan Plateau, using Sentinel-1 data

Abstract

Differential Interferometric Synthetic Aperture Radar (D-InSAR) has been widely used to measure surface deformation over the Tibetan Plateau. However, the accuracy and applicability of the D-InSAR method are not well estimated due to the lack of in-situ validation. In this paper, we mapped the seasonal and long-term displacement of Tanggula (TGL) and Liangdaohe (LDH) permafrost regions with a stack of Sentinel-1 acquisitions using the Small Baseline Subset InSAR (SBAS-InSAR) method. In the TGL region, with its dry soils and sparse vegetation, the InSAR-derived surface-deformation trend was consistent with ground-based leveling results; long-term changes of the active layer showed a settlement rate of around 1 to 3 mm/a due to the melting of ground ice, indicating a degrading permafrost in this area. Around half of the deformation was picked up on monitoring, in contrast with in-situ measurements in LDH, implying that the D-InSAR method remarkably underestimated the surface-deformation. This phenomenon may be induced by the large soil-water content, high vegetation coverage, or a combination of these two factors in this region. This study demonstrates that surface deformation could be mapped accurately for a specific region with Sentinel-1 C-band data, such as in the TGL region. Moreover, although the D-InSAR technology provides an efficient solution for broad surface-deformation monitoring in permafrost regions, it shows a poor performance in the region with high soil-water content and dense vegetation coverage.