Recommended Free Tools
A digital elevation model (DEM) is a digital representation of elevation across a geographic area, usually stored as a georeferenced grid of values. In U.S. Geological Survey (USGS) usage, a DEM normally shows the bare-ground surface, with trees, buildings, and other surface objects excluded. Outside that convention, “DEM” can be a general umbrella term, so the surface a particular dataset represents has to be confirmed in its documentation.
What the term means
The USGS FAQ defines the term this way: “A Digital Elevation Model (DEM) is a representation of the bare ground (bare earth) topographic surface of the Earth excluding trees, buildings, and any other surface objects.” The FAQ page, as checked for this article, does not show a publication date, so the definition should be attributed to the USGS without a year.
As an Amazon Associate I earn from qualifying purchases.
The USGS glossary adds a structural detail. Its convention describes elevation values at regularly spaced x/y positions, with elevations referenced to a common vertical datum. A DEM is therefore digital data rather than one fixed file format. The same elevation model can be delivered in different containers, and the file format tells you little about what the elevations mean.
The abbreviation alone does not guarantee identical surface treatment across producers. A peer-reviewed terminology review describes DEM as a general term and explains that the related terms DSM and DTM specify how features above the ground are handled. The same review notes that datasets differ in vertical datum, in the ground area each pixel represents, and in spacing. Some grids use spacing that varies with latitude.
#1 Best Overall
DEM, DSM, and DTM: the terms that get confused
DEM: usually bare earth in USGS and common U.S. usage
In USGS and much U.S. practice, a DEM is a raster of bare-earth elevation. Its values describe the ground under vegetation and structures, not the top of the canopy or roof. Elsewhere, producers may use “DEM” for any elevation grid, including one that keeps surface objects. That is why the metadata, rather than the acronym, should settle the question.
DSM: the surface including objects
A digital surface model (DSM) records elevations of the uppermost surface. Tree crowns, building roofs, towers, and similar features remain in the values. A DSM is the better choice when those features are the object of study, for example when estimating canopy height above ground, or when checking whether a building or transmission structure is in a line-of-sight problem. For a flood or drainage analysis that depends on the ground surface, a DSM can misrepresent where water flows, because the model records objects that water moves around.
Rank #2
DTM: a term whose meaning depends on the community
The digital terrain model (DTM) is the least uniform of the three. In some countries, DTM is used as a synonym for a bare-earth DEM. In USGS lidar terminology, however, a DTM can mean a vector dataset of terrain mass points and breaklines. A continuous TIN or DEM surface can then be derived from those points and lines. A file labelled DTM may therefore be a set of points, a raster, or a derived surface, and the producer’s definition determines which.
Reading a DEM grid: resolution is not accuracy
A raster DEM assigns an elevation value to each cell, or to each regularly spaced sample location. The linear size of a cell, also called cell size or grid spacing, is commonly what people mean by resolution. Any feature smaller than one cell cannot be represented explicitly in that raster, so a 10-metre grid will smooth away a narrow ditch that a 1-metre grid could capture.
Cell size describes horizontal sampling scale only. It does not establish how close the stored elevations are to the true ground. A USGS 1-Meter Digital Elevation Model specification describes cells of 1 metre by 1 metre. That is a property of one product specification, not a standard resolution that every DEM meets. A finer grid with weak vertical control can be less reliable than a coarser grid with documented, tested vertical accuracy.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What to check before comparing or using a DEM
When two or more elevation datasets are available for the same area, compare them along the following axes. These are the dimensions that USGS DEM specifications and the terminology review both treat as essential.
Rank #4
- New
- Mint Condition
- Dispatch same day for order received before 12 noon
- Guaranteed packaging
- No quibbles returns
- Represented surface: bare earth, surface including objects, or another treatment stated by the producer.
- Coordinate system and vertical datum: the horizontal reference and the datum to which elevations are referenced. Two DEMs on different vertical datums can differ by a systematic offset even where the terrain is identical.
- Cell size, grid spacing, and pixel-area convention: the horizontal spacing and the ground area each pixel is said to represent.
- Vertical accuracy and quality control: the stated accuracy and the checks used to verify it.
- Source and processing: the sensor, acquisition or tile dates, coverage, any data voids, and the processing applied.
If the metadata does not state one of these items, treat that gap as a limitation of the dataset rather than an assumption about it.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
