Evaluation of TanDEM-X elevation data for geomorphological mapping and interpretation in high mountain environments - A case study from SE Tibet, China

Digital elevation models (DEMs) are a prerequisite for many different applications in the field of geomorphology. In this context, the two near-global medium resolution DEMs originating from the SRTM and ASTER missions are widely used. For detailed geomorphological studies, particularly in high mountain environments, these datasets are, however, known to have substantial disadvantages beyond their posting, i.e., data gaps and miscellaneous artifacts. The upcoming TanDEM-X DEM is a promising candidate to improve this situation by application of state-of-the-art radar technology, exhibiting a posting of 12 m and less proneness to errors. In this study, we present a DEM processed from a single TanDEM-X CoSSC scene, covering a study area in the extreme relief of the eastern Nyaingentanglha Range, southeastern Tibet. The potential of the resulting experimental TanDEM-X DEM for geomorphological applications was evaluated by geomorphometric analyses and an assessment of land-form cognoscibility and artifacts in comparison to the ASTER GDEM and the recently released SRTM 1 DEM. Detailed geomorphological mapping was conducted for four selected core study areas in a manual approach, based exclusively on the TanDEM-X DEM and its basic derivates. The results show that the self-processed TanDEM-X DEM yields a detailed and widely consistent landscape representation. It thus fosters geomorphological analysis by visual and quantitative means, allowing delineation of landforms down to footprints of similar to 30 m. Even in this premature state, the TanDEM-X elevation data are widely superior to the ASTER and SRTM datasets, primarily owing to its significantly higher resolution and its lower susceptibility to artifacts that hamper landform interpretation. Conversely, challenges toward interferometric DEM generation were identified, including (i) triangulation facets and missing topographic information resulting from radar layover on steep slopes facing toward the radar sensor, (ii) low coherence values on leeward slopes, (iii) decorrelation effects over water bodies, and (iv) challenges for phase unwrapping in settings of strong topographic contrasts. There is, however, a high probability that these drawbacks can be overcome by applying multiple interferograms exhibiting different perpendicular baselines as planned for the generation of the final TanDEM-X DEM product. (C) 2015 Elsevier B.V. All rights reserved.