The spatial extent of thermoerosional processes and related landforms (e.g. gullies, valleys) and their impact on the widespread degradation of permafrost is still not well quantified. Remote sensing data and digital elevation models (DEMs) are widely used to derive and analyse geomorphometric relief characteristics in order to detect periglacial landscape dynamics such as thermal erosion. However, geometric correction of remote sensing data and generation of DEMs in polar lowlands is challenging due to the low relief of arctic tundra landscapes and often scarce reference data. Therefore, high-resolution DEMs of high quality are hardly available, especially for Siberian study regions. Addressing these difficulties, this study is using a multi-sensor and multi-temporal data approach for a detailed inventory and spatial analysis of thermo-erosional landforms on Kurungnakh Island in the central Lena Delta, which is composed of ice-rich permafrost. Two DEMs were generated from ALOS PRISM stereo-datasets acquired in 2006 and 2009 for detection of short-time volumetric change. Mapping of thermo-erosional landforms was then performed using a time-series of orthorectified GeoEye-1 and RapidEye datasets. We present a validation of the generated DEMs based on extensive ground measurements on Kurungnakh Island, performed during an expedition in July 2013. Longitudinal and transversal profiles of thermo-erosional valleys were measured at three key sites, each representing different valley types. Profiles were extracted from the DEMs and compared to profiles measured in the field in order to provide error ranges for DEM-based valley profile analyses. Our datasets provide a new level of accuracy and a basis for further studies on Kurungnakh Island.