In permafrost active layer cycles of excess ice formation in winter and loss in summer result in seasonal vertical movements of the ground in both directions. Additionally, relatively uniform thawing of the ice-rich layer at the permafrost table, contributing to irreversible lowering of the surface, was reported for a number of Arctic locations. We use a simple method to quantify surface lowering (subsidence) and uplift in the Lena River Delta, Siberian Arctic, using more than 30 reference rods (fiberglass and metal) installed deeply in permafrost. We repeatedly measured the length of a rod part, which is emerged above the ground, in 2013-2017. Measurements show seasonal subsidence in a range from 0 to 4.6 cm (median: 1.6 cm; 8 measurements) in the cold summer of 2013 and from 0.8 to 8.6 cm (median: 4.8 cm; 31 measurements) in the warm summer of 2014. A pronounced multi-year subsidence of 9.3±5.7 cm was measured in the end of summer 2017 relative to the initial measurements in spring 2013. Additionally, we observed high spatial variability of subsidence even at the sub-meter scale. Differential Synthetic Aperture Radar Interferometry (DInSAR), most often used to measure ground displacement caused by tectonic or volcanic processes, is adapted now for the detection of subsidence in permafrost. Our study tests the viability of repeat pass (11 days) TerraSAR-X (TSX) data for the detection of thaw subsidence over the same study area. Due to TSX short wavelength and, therefore, shallow penetration depth, interferometry is strongly hampered by poor phase coherence. We built a stack of 11-day interferograms for the summer of 2013 where coherence of some single interferograms was on the edge of the acceptable. The stack showed only a minor subsidence with a mean of 0.3±0.3 cm over the studied area. Given the discrepancy between the DInSAR and field data we discuss the limitations of TSX data for an accurate representation of permafrost thaw subsidence.