Quality Assessment and Validation of Digital Surface Models Derived from the Shuttle Radar Topography Mission (srtm)

The first mission using space-borne single-pass-interferometry was launched in February this year – the Shuttle Radar Topography Mission (SRTM). The goal of this mission has been to survey the Earth’s surface and generating a homogeneous and dense elevation data set of the entire world. Antennas with two different wavelength were used. One of the hardware components was the German / Italian Synthetic Aperture Radar X-band antenna (X-SAR). In this paper we report on an attempt to check the quality and accuracy of the elevation data derived from the X-band instrument of SRTM. First of all possible error sources influencing the data quality and accuracy will be described and the effects of these errors will be demonstrated. Reference data of a well known test site will be used to assess the data and to derive quality measures. The area with a size of 50x50 km2 is situated in Germany a few kilometers south of Hanover. Reference data are being made available by the Surveying Authority of Lower Saxony, Germany (Landesvermessung und Geobasisinformation Niedersachsen, LGN Hannover). The Digital Terrain Model of LGN (ATKIS DGM5) is said to have an accuracy of about ±0,5 meters. Also Trigonometric Points, which are the base of the fundamental geodetic network of Germany, are being used as reference data. The tool used for assessing the data is a spatial transformation. As a result 7 parameters are being obtained, which describe the position, orientation and a scale of the SRTM elevation data with respect to the reference data. To obtain influences of terrain slope, the orientation of the terrain with respect to the sensor position, vegetation, land use and land cover the test side is divided into several subareas. The accuracy and quality of the data as a function of these parameters will be calculated.