Analysis of Srtm Dtm Methodology and Practical Results

In February 2000 the first mission using space-borne single-pass-interferometry was launched – the Shuttle Radar Topography Mission (SRTM). The goal of the mission was to survey the Earth surface and to generate a homogeneous elevation data set of the world with a grid spacing of 1 arcsec. Antennas with two different wavelengths were used: Beside the American SIR-C the German / Italian X-SAR system was on board. This paper deals with the assessment of the Interferometric Terrain Elevation Data derived from the X-SAR system. These so called ITED-2 data were compared to reference data of higher quality of a well known test site in the South of Hannover (Trigonometric Points and Digital Terrain Model). The approach used is based on a spatial similarity transformation without using any kind of control point information. The algorithm matches the SRTM data onto the reference data in order to derive seven unknown parameters which describe horizontal and vertical shifts, rotations and a scale difference with respect to the reference data. These values describe potentially existing systematic errors. The standard deviation of the SRTM ITED-2 was found to be ±3,3 m in open landscape, after applying the spatial similarity transformation. Maximum horizontal systematic shifts of 4-6 m were detected, representing only 20-25 % of the ITED-2 grid size. In summary, it can be stated that the results are much better than predicted before the start of the mission. Thus, the quality of the SRTM ITED-2 is indeed remarkable.