Digital elevation model

from Wikipedia, the free encyclopedia

There is no generally applicable definition of the terms digital elevation model ( DHM ) and digital terrain model ( DGM ) in the scientific literature . The digital terrain model ( DTM ) is also used internationally .

Represented surface of a DOM (red) and a DTM (turquoise)

The digital surface model ( DOM , English digital surface model ) represents the earth's surface (boundary layer pedosphere - atmosphere) together with all objects located on it ( buildings , streets, vegetation , water, etc.). By contrast, represents the digital terrain model ( DTM , Eng. Digital terrain model ) the earth's surface without vegetation and structures (see. Li u. A. 2005, AdV in 2004, see figure).

The definitions of the term digital elevation model differ greatly with regard to the surface represented. Some definitions are discussed in more detail below.


Digital terrain model

3D visualization of a digital terrain model using the example of a canyon on Mars

The fact that the term DGM refers to the earth's surface as a reference surface is quite clear in the specialist literature. In addition, some definitions require additional information about the terrain surface such as the terrain height in addition to the terrain heights. B. terrain edges , ridges , prominent elevation points or the floor plan elements with floor plan information (cf. ISO 18709-1, Bill 1999 and Li et al. 2005).

The working group of the surveying administrations of the federal states of the Federal Republic of Germany (ADV 2004) defines the DGM as follows:

"Database for describing the elevation of the site. It consists of regularly or irregularly distributed terrain points that adequately represent the height structure of the terrain - optionally supplemented by morphological structural elements (e.g. terrain edges, distinctive height points). "

The discontinuous or non-monotonous areas of the surface can be described in the model by means of structural information (in particular break lines, ridges, perimeter and recess areas or fault line (see fault line (geology) ) of the surface; see structure line ). Breaklines represent discontinuities in the gradient, i.e. kinks in the terrain, between the support points. The ridges are the ridge and valley lines.

Digital elevation model

3D visualization of a digital surface model

Much contradictory the definition of digital elevation model (Engl. Digital elevation model ). In addition to regional and subject-specific differences, there are also differences within the specialist disciplines. The term is often used as a generic term for digital terrain models and digital surface models (Peckham & Gyozo 2007, Hofmann 1986).

So z. B. from Geobasis NRW:

"DHM are the generic term for digital terrain models (DGM) and digital surface models (DOM)."

This definition is also used by most data providers ( USGS , ASTER -ERSDAC, CGIAR -CSI). The well-known, almost global elevation data sets SRTM DEM and ASTER GDEM are de facto digital surface models.

During the Shuttle Radar Topography Mission (SRTM) in February 2000, an almost global elevation model was created using the Synthetic Aperture Radar (SAR) sensor or radar interferometry. The SRTM data are in the public domain. The ASTER Global DEM based on optical remote sensing is available free of charge for many purposes ( ASTER ).

Other definitions equate the DEM with the DTM (Podobnikar 2008), or define the DTM as an extended DEM that also describes the terrain (break lines, ridges, etc.) (Graham et al. 2007). So z. B. ISO 18709-1 (terms, abbreviations and symbols in surveying - Part 1: General) (see web links DIN):

"It (the DGM, author's note) consists of elevation points of the DHM and additional information (terrain edges, distinctive elevation points)."

An overview, even if not complete, of the extremely numerous and different definitions can be found in Li et al. a. (2005). It must be checked very carefully in each individual case which surface the model refers to.


German Aerospace Center (DLR): Elevation model of Iceland based on satellite data (2012)

In the following, the term DEM is used as a generic term for DGM and DOM. Digital elevation models have been used in many areas of geosciences and technology since around 1980 - including in geodesy and photogrammetry for terrain mapping, in construction for the alignment of traffic routes, for military tasks (e.g. control of cruise missiles along the surface of the earth) up to planning the sewage system. Recently, elevation models of other planets have even been created, as made possible by radar probes around Mars and Venus .

Data formats

Definitions can be found on the Internet according to which the DHM is defined as a raster database and the DGM as a real three-dimensional model (e.g. an irregular triangular network of the original measuring points; see web links Landslide Glossary USGS). More often, however, the data format is defined using the terms primary DEM and secondary DEM (Toppe 1987). In order to obtain a digital terrain model, the objects on the earth's surface (houses, trees, etc.) must first be filtered out using complex algorithms in recording processes that use flight objects or satellites as platforms (Li et al. 2005).

Primary DEM

TIN (blue) with superimposed contour lines

With regard to the position irregularly arranged support points are typical for measured or primary DEM, in which the support points represent the original measurement data. The points are saved as vector data together with the structural information.

The most common form is the irregular triangular mesh (Engl. Triangulated irregular network , TIN ). With the TIN, the support points are connected to form a triangular network. The surface is modeled as a polyhedron . The height can be interpolated linearly within a triangle.

Secondary DEM

A regular grid-like arrangement of the support points can be found especially in calculated or secondary DEM. Here we speak of grid-DHM . A uniform grid is placed over the area. A height value is assigned to each grid point. The raster data format is suitable for grid DEMs . No structural information can be saved. Heights between the support points can be calculated using the interpolation method of digital image processing (see e.g. bilinear interpolation ). In order to be able to reproduce the surface exactly, the grid width must be selected so narrow that striking structures do not fall through the grid. The grid widths are locally or regionally 2 to 500 meters, for global models 1 to 5 km.

Hybrid DEM

A hybrid DEM is a grid DEM, to which structural information in the form of additional points, lines and areas is added.


The image-based digital surface model b DOM forms the earth's surface and the objects on it, such as B. Vegetation and buildings in grid form.


DEM of the Vomper Loch based on ASTER data
DEM of the Vomper Loch , based on openDEM data

The accuracy of DEM is mainly dependent on the recording method, the grid width and the surface roughness. The accuracy is made up of a position and a height accuracy component, the position component being dependent on the surface inclination (tan α). . The average errors - depending on the purpose and price of the models - range from a few centimeters (e.g. for determining flood areas in the course of flood protection concepts) to a few 100 meters.

The two graphics compare the digital terrain model of the Vomper Loch in the Alps. The graphic above is based on data from ASTER . The basis for the picture below is provided by SRTM data, which are freely available in processing stage 2.1 and which have been further processed by the openDEM project . Height artifacts as steep mountain formations emerge clearly in the ASTER image. They are caused by shadowing during the survey of the terrain. Missing height values ​​were estimated due to a lack of plausibility analyzes using heuristic assumptions. The SRTM measured values ​​are less resolved than those from ASTER. Since they were subsequently compared with other information, for example using data from OpenStreetMap , the elevation profile is consistent. The water surface of lakes should be flat. Deviations directly reflect the spread of the measured values.

See also


Web links


  • Working group of the surveying administrations of the federal states of the Federal Republic of Germany (AdV) (2005). Glossary for defining the GeoInfoDok - sub-area DGM . LINK.
  • R. Bill: Fundamentals of Geographic Information Systems . Volume 2: Analyzes, Applications and New Developments. Herbert Wichmann Verlag, Heidelberg 1999, ISBN 3-87907-326-0 .
  • AW Graham, NC Kirkman, PM Paul: Mobile radio network design in the VHF and UHF bands: a practical approach . West Sussex 2007, ISBN 978-0-470-02980-0 .
  • W. Hofmann: Once again: The digital terrain / elevation model. In: image measurement u. Luftbildwesen , 54 (1986), H. 1, pp. 31-31; Karlsruhe.
  • Z. Li, Q. Zhu, C. Gold: Digital Terrain Modeling: principles and methodology . CRC Press, Boca Raton 2005, ISBN 0-415-32462-9 , pp. 7-9.
  • Robert Joseph Peckham, Gyozo Jordan (Ed.): Development and Applications in a Policy Support Environment . (Lecture Notes in Geoinformation and Cartography). Springer, Heidelberg 2007, ISBN 978-3-540-36730-7 .
  • Tomaz Podobnikar: Methods for visual quality assessment of a digital terrain model. In: Sapien.s. 1 (2008), no. 2. (online)
  • R. Toppe: Terrain models - a tool for natural hazard mapping. In: Bruno Salm (Ed.): Avalanche Formation, Movement and Effects . (Proceedings of the Davos Symposium, September 1986). Boarding school Assoc. of Hydrol. Sciences, Wallingford 1987, ISBN 0-947571-96-5 . (IAHS Publ. No.162)

Individual evidence

  1. , January 3, 2012: The earth in 3D a lot closer (December 23, 2016)
  2. surface model | LGB_Startseite. Retrieved March 30, 2020 .