Titel:
Uncertainties in Digital Elevation Models: Evaluation and Effects on Landform and Soil Type Classification
Beteiligte:
Kramm, Tanja
[Verfasser:in]
Erschienen:
Cologne University: KUPS, 2022
Sprache:
Deutsch;
Englisch
Entstehung:
Anmerkungen:
Diese Datenquelle enthält auch Bestandsnachweise, die nicht zu einem Volltext führen.
Beschreibung:
Digital elevation models (DEMs) are a widely used source for the digital representation of the Earth's surface in a wide range of scientific, industrial and military applications. Since many processes on Earth are influenced by the shape of the relief, a variety of different applications rely on accurate information about the topography. For instance, DEMs are used for the prediction of geohazards, climate modelling, or planning-relevant issues, such as the identification of suitable locations for renewable energies. Nowadays, DEMs can be acquired with a high geometric resolution and over large areas using various remote sensing techniques, such as photogrammetry, RADAR, or laser scanning (LiDAR). However, they are subject to uncertainties and may contain erroneous representations of the terrain. The quality and accuracy of the topographic representation in the DEM is crucial, as the use of an inaccurate dataset can negatively affect further results, such as the underestimation of landslide hazards due to a too flat representation of relief in the elevation model. Therefore, it is important for users to gain more knowledge about the accuracy of a terrain model to better assess the negative consequences of DEM uncertainties on further analysis results of a certain research application. A proper assessment of whether the purchase or acquisition of a highly accurate DEM is necessary or the use of an already existing and freely available DEM is sufficient to achieve accurate results is of great qualitative and economic importance. In this context, the first part of this thesis focuses on extending knowledge about the behaviour and presence of uncertainties in DEMs concerning terrain and land cover. Thus, the first two studies of this dissertation provide a comprehensive vertical accuracy analysis of twelve DEMs acquired from space with spatial resolutions ranging from 5 m to 90 m. The accuracy of these DEMs was investigated in two different regions of the world that are substantially different in terms of relief and ...