Monitoring high-mountain terrain deformation from repeated air- and spaceborne optical data:: examples using digital aerial imagery and ASTER data

High mountains represent one of the most dynamic environments on earth. Monitoring their terrain changes is necessary to understand mass-transport systems, to detect related environmental variability, and to assess natural hazards. Here, we apply standard software to automatically generate digital elevation models (DEM) from aerial photography and Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) satellite stereo imagery. By comparison to a photogrammetrically derived DEM, an accuracy of +/- 60 m RMS of the ASTER DEM was found for rough high-mountain topography, and +/- 18 m RMS for moderately mountainous terrain. Differences between multi-temporal DEMs are used to determine vertical terrain changes. Horizontal movements are computed from multi-temporal orthoimages. The techniques are applied for three case studies. (1) The flow-field of Tasman glacier, New Zealand, as measured from ASTER data, showed glacier speeds of up to 250 m per year and a surprising minimum speed in the middle of the glacier. (2) The velocity-field of creeping mountain permafrost in Val Muragl, Swiss Alps, with speeds of up to 0.5 m per year was determined with high resolution from aerial stereo imagery and provided new insights in the spatial coherence of permafrost creep. (3) Deformations of up to 0.1 m per year on a large landslide near Aletsch glacier, Swiss Alps, could be detected. As a rule of thumb, we estimate the achieved accuracy for elevation changes and horizontal displacements to approximate the size of one image pixel, i.e. 15 m for ASTER and 0.2-0.3 m for the here-used aerial photography. (C) 2002 Elsevier Science B.V. All rights reserved.