Digital Elevation Models are digital representations of the elevation of the Earth's surface or terrain, usually stored as a raster grid or another structured elevation surface. Each spatial position is associated with an elevation value relative to a specified vertical reference. The term can be used in different ways by different organisations, and it is important to determine whether a particular dataset represents bare earth terrain, the elevation of visible surfaces or another elevation concept.
A Digital Terrain Model is often used specifically for terrain or bare earth, while a Digital Surface Model commonly represents the elevation of the uppermost surface including buildings and vegetation. Usage varies by discipline and data provider, so the metadata definition should be checked rather than assuming universal terminology. Digital Elevation Models are fundamental to terrain analysis, hydrology, engineering, cartography, remote sensing, environmental modelling and three dimensional geographic visualisation because elevation influences drainage, visibility, slope, aspect, accessibility and many physical processes.
Digital Elevation Models can be created from many sources, including surveying, photogrammetry, airborne laser scanning, satellite remote sensing, radar measurements and interpolation from elevation points or contours. Traditional topographic mapping used contour lines and surveyed spot heights from which elevation surfaces could later be derived. Photogrammetric methods can estimate three dimensional terrain from overlapping Aerial Photography, while airborne laser scanning can measure large numbers of surface points using laser pulses.
Satellite radar techniques can produce elevation information over very large areas. Different acquisition methods produce different types of elevation model with different resolutions, accuracies, vertical references and characteristics. A model derived from dense airborne observations may represent small terrain features that a coarse global model cannot resolve. Consequently, the suitability of a Digital Elevation Model depends strongly on the geographic scale, terrain complexity, vertical accuracy and purpose of the analysis.
Raster Digital Elevation Models commonly represent elevation as regularly spaced cells. The cell size determines the horizontal sampling resolution, while the elevation value represents a location or area according to the dataset's definition. Smaller cells can capture finer terrain variation but require greater storage and processing resources. Interpolation may be necessary when the original observations are irregularly distributed, and the interpolation method influences the resulting surface.
Common approaches include nearest neighbour, bilinear interpolation, spline methods and various forms of statistical or terrain aware interpolation. Data voids may require filling, but filling can introduce artificial terrain if the missing information is extensive. Vertical units, horizontal coordinate systems and vertical datums must also be documented. A numerical elevation value has no complete meaning without knowing the reference surface and coordinate framework to which it relates.
Digital Elevation Models support numerous terrain derivatives. Slope describes the rate of elevation change and is commonly expressed as an angle or percentage. Aspect describes the direction toward which a slope faces. Hillshading creates a simulated illumination effect that makes terrain structure easier to visualise. Contour lines can be generated from elevation surfaces, while profiles can show elevation along selected paths. Watershed and drainage analysis can use terrain surfaces to identify flow directions, accumulation patterns and potential catchment boundaries.
Visibility analysis can determine whether one location can be seen from another under specified assumptions. Curvature measures can help characterise terrain shape. These derivatives are mathematical interpretations of the elevation model rather than independent observations, meaning errors in the underlying elevation data can propagate into derived products. The choice of cell size, processing method and terrain correction can also influence the results.
Hydrological applications are among the most important uses of Digital Elevation Models. Water tends to move downslope under gravity, so terrain structure can provide information about potential drainage pathways and catchment organisation. Hydrological preprocessing may involve correcting sinks or depressions that are interpreted as artefacts, although real closed depressions can exist and should not automatically be removed. Flow direction algorithms estimate how water could move between cells, while flow accumulation identifies areas receiving contributions from upstream cells.
Derived drainage networks and watershed boundaries can support flood modelling, catchment studies and environmental analysis. However, a Digital Elevation Model alone does not constitute a complete hydrological model. Soil properties, rainfall, land cover, drainage infrastructure, vegetation and subsurface conditions can significantly affect actual water movement. Elevation provides an important geometric foundation, but hydrological conclusions require appropriate physical and environmental assumptions.
Digital Elevation Models are also widely used in engineering, planning and environmental management. Infrastructure designers can examine terrain constraints for roads, railways, pipelines and construction sites. Planners can analyse slope suitability and potential visual exposure. Telecommunications engineers can model line of sight for communication infrastructure. Coastal studies can use elevation data to analyse low lying areas and model potential inundation under specified water level scenarios. Forestry and ecology can derive terrain variables that influence habitats and species distributions.
Geological analysis can use terrain morphology to investigate landforms and structural patterns. In mapping, elevation models can support three dimensional visualisation and orthorectification of imagery. The quality required varies considerably. A regional climate or watershed study may use a relatively coarse elevation model, while detailed engineering or construction work may require highly accurate local survey data.
Accuracy and uncertainty are critical when using Digital Elevation Models. Horizontal resolution and vertical accuracy are separate properties, so a dataset with small cells is not automatically highly accurate. Terrain complexity can cause greater error in steep or obstructed areas. Vegetation and buildings can affect surface measurements, especially when the model represents the upper surface rather than bare earth. Radar and optical methods can have their own geometric limitations. Vertical datum differences can create systematic offsets when datasets are combined.
Resampling can smooth or alter terrain features, while interpolation can create values that were never directly measured. Users should examine metadata describing acquisition method, spatial resolution, vertical accuracy, datum, voids and processing. Combining elevation models without harmonising these properties can introduce apparent terrain differences that are actually caused by measurement or reference inconsistencies rather than genuine geographic change.
Digital Elevation Models have strong relationships with Aerial Photography, Satellite Mapping, Geographic Databases and Digital Mapping. Photogrammetric processing of overlapping aerial images can produce dense three dimensional point clouds from which elevation surfaces are generated. Satellite systems can provide regional or global elevation datasets. Geographic Databases can store elevation related features and derived terrain products, while Digital Mapping and Web Mapping systems can display terrain through contours, shaded relief, three dimensional scenes or elevation based analysis.
Digital Elevation Models can also support georeferencing and orthorectification of imagery because terrain height affects the apparent position of features in perspective imagery. Their broader significance comes from representing one of the fundamental dimensions of geographic space. By converting terrain elevation into structured digital information, they allow geographic systems to analyse and visualise the vertical structure of landscapes alongside their horizontal location and other spatial attributes.
