|A digital terrain model (DTM) is different than a digital elevation model (DEM), the first representating a continuous surface as if all the objects are removed – a bare surface. Alternatively, a DEM represents a surface where objects are included in the surface measurement. It is possible to process DEM, so they are DTM. For all intents and purposes a DTM represents the earth’s land surface. The processing involved to turn a DEM into a DTM means added cost to a project. The obvious question is, why use a DTM instead in the first place? So let’s start there.
Houses come and go, bridges come and go and trees just grow and die. Skyscraper’s are built, and even false covering can obscure the land beneath. For these reasons, a DTM represents the earth’s true surface, regardless of all of these objects and their changing natures.
Surfaces are usually interpolated. This recognizes the principle that not all points on the earth’s surface are actively measured and determined, thus they have to be calculated (interpolation). Suffice it to say there are many interpolation methods, but we won’t go into that here. Similarly, there are many processing methods for DTM which are derived from airborne radar, laser (lidar) or photographically derived surfaces. There are reasons for considering each of these methods for building a DTM, none should be counted out.
Since DTM files are large – very large – in file size, processing them and handling them is a critical task. Beginning on this point, today’s DTM needs to be available to user’s quickly, this includes Internet downloadable files usually, which should be geo-referenced for immediate use. Some people compress DTM data and many people deliver it using different approaches – all designed to speed up user download time.
Traditionally DTM have been used to build elevation models of the earth’s surface. They are 3D in nature and thus lend themselves to several important uses. These days, few people solely build a DTM for that use alone, though some do, particularly those who like to see a 3D model of the earth’s surface, either digitally or in physical form using clay, cement or other material.
It is the 3D nature of a DTM that really begins to unleash its true value
- DTM can more closely be linked to the earth’s physical processes, like water infiltration, overland flow, floods, vegetation distributions and os on. This is because the models that drive these processes are each impacted by 3D space – elevation. Water runs downhill quicker on steeper land etc. These days floods are a huge issue, thus DTM are in high demand for insurance reasons, as example.
- DTM can provide an accurate surface upon which to drape imagery. Since much of today’s imagery is high resolution as well, then more photo-realistic draping and visualisation is possible. The two go hand-in-hand.
- DTM have a role in military application. It is probably a brighter idea to use a DTM if you have cruise missiles rather than using a DEM, for example. Can you explain why?
- DTM can also be useful for man-made processes and events including more accurately siting telephone towers, forest observation points and and other viewshed determinations.
These are the basic uses for a DTM, but today, the uses are far more varied, covering a wider territory of value-added products. They are used to tie in vehicle navigation (Intermap) for example has been working on linking car-road navigation more closely, thereby reducing the number of vehicle accidents. The folks at ITT (Visual Information Solutions), the makers of ENVI image processing software have been working with CREASO in Germany to processing SAR data, extracting feature objects (change) from it.
ESRI ArcGIS, AutoCAD Map 3D 2008, Bentley Microstation XM, ERDAS Imagine – all handle DTM data, which is critical for integrating with other feature derived data or building models upon.
The idea today is to build valuable products upon DTM, extending their usefulness, but also addressing real issues that incorporate other spatial information. The link of DTM Surface – Processes – Other Spatial Data is critical to understand.
The issues are about change, accordingly, the link back to DEM should also not be ruled out, since DEM show changes upon the surface.
Today visualisation is an important role for DTM, it enables real world fly-through and walk-through of landscapes.
But the link to CAD software is not yet fully embrace. Bentley and Autodesk are two companies that could be linking civil engineering processes, at the design stage, directly to DTM. Some DTM will be raster based (grids) while others will be vector based (point derived-interpolated). Issues like cut-fill are impacted by these DTM model types, as are processes.
We should not rule out gaming and the use of DTM. A DTM can be wholly fictional in nature. We could build an imaginary world with an imaginary surface and populate it with unusual buildings and other objects. Thus a real verus virtual aspect impacts 3D visualisation, and the data involved.
DTM are integral pieces to the building infrastructure, managing it and communicating about our world. Without a surface, there are only objects in space – little to reference anything.
It is the DTM that actually makes 3D happen – in the real world.
Digital Terrain Models
Digital Elevation Models
Pubs about DTM
||A digital elevation model (DEM) represents the elevation of Earth’s surface, including features (vegetation, buildings, etc.). A digital terrain model (DTM) provides a bare earth representation of terrain or surface topography. Both are highly useful data sets for visualizing our planet for scientific and commercial landscape study.
Traditionally elevation was derived by direct survey of the land, then photogrammetry techniques allowed for terrain extraction with a pair of images with stereoscopic techniques. Today, highly accurate digital elevation can be achieved with light detection and ranging (LIDAR) and interferometric synthetic aperture radar (IfSAR). LIDAR transmits laser pulses that bounce off the Earth’s surface and are measured. IFSAR is a radar technique that uses stereo pairs of radar images and photogrammetric processing to derive the Earth’s surface.
With each technological advancement, the digital elevation models have improved in accuracy, resulting in a much more useful model of the Earth.
Data Sources and Resolution
To date, global data is available in approximately 1 km resolution from GTOPO30 and 90 m accuracy from the Shuttle Radar Topography Mission (SRTM).
A number of government IfSAR satellites are in orbit, the European Space Agency has ERS-2 (30 m) and more recently ASAR aboard Envisat (30 m), and the Canadian Space Agency has Radarsat-1 (10 m) and will soon launch Radarsat-2 (3 m). The new public/private partnership between the German government and Infoterra launched the TerraSAR-X satellite and is already capable of delivery elevation data in 1-2 meter accuracy. These space-based platforms are crucial for wide-area coverage and their frequent revisit rates means that a valuable archive will be built for scientific observation.
Commercial airborne-based elevation collection is ongoing as well. Intermap Technologies began an aggressive collection of IFSAR elevation data back in the late 90’s, and has delivered entire country data sets. Their NEXTMap program is active in Europe (with projected completion by the end of 2007) and the United States (with completion slated for the end of 2008). Intermap delivers elevation accuracy of .5 meter.
LIDAR service companies collect data on a project basis at an elevation resolution of roughly 15 cm. There are a number of LIDAR Services companies with headquarters in the United States that are conducting large-scale projects — EarthData has a project underway to collect the entire state of North Carolina, Merrick & Company has completed a number of large-scale projects including a project in Colombia, South America, 3001 and MJ Harden (now a part of GeoEye) are also very active.
While LIDAR delivers a more precise measurement, the advantage of radar technologies is that they cut through clouds and can be flown at night. Therefore, IFSAR is a much more efficient process for capturing large geographies.
Detailed elevation models are used for a wide variety of scientific and commercial purposes. Hydrology or water modeling is one of the premier applications as water flows downward and the greater the elevation accuracy the better the water model. Water models also help determine flood extent and assess flood damage.
Marine observations of wave heights and ocean characteristics can be determined by satellite elevation sensors, and highly accurate elevation in coastal zones greatly help the study of coastal change and storm impacts. Observing forests with elevation sensors returns forest characteristics such as crown height and canopy cover.
Geological land observations are also dependent on accurate elevation to determine such things as land subsidence, landslides or avalanches, and extracting geomorphological information.
Elevation data is critical for creating 2D relief maps, 3D flythroughs, and physical raised-relief models. Elevation in a city provides lie-of-sight information and an understanding of light and shadow that are important for energy consumption studies. A growing interest in city model collection will depend on accurate DEM and DTM data.
Elevation Projects of Note
In the United States there’s a large multi-year project underway to modernize the Federal Emergency Management Association (FEMA) flood maps. The National Flood Insurance Program relies heavily on accurate flood maps to reduce risk and limit payouts when flood disasters occur. Much of the current map products are paper based. The effort underway aims for modern Web-based mapping products with much greater accuracy by 2010.
The National Geodetic Survey has also been testing a variety of different elevation technologies in order to map the 95,000 miles of U.S. coastline that they’re responsible for.
The Federal Aviation Administration is also in the process of modernizing with a GPS-assisted Next Generation Air Transportation System (NGATS). Billions of dollars will be spent on this upgrade, and elevation data will likely be a key component.
Intermap Technologies has been one of the most active companies in exploring and exploiting new opportunities in the elevation data space. They have a transportation group that has tapped into intelligent transportation and automated driver assistance initiatives.
With accurate elevation data and know position, vehicles can respond to their surroundings. The use of data and systems in the long-haul trucking industry could allow the engine to respond in advance to terrain changes, saving both on maintenance and on fuel consumption. There’s also talk about adaptive lighting with beams from headlights following road changes. Driver safety becomes a huge benefit of this added awareness.
Infrastructure projects have just begun to exploit large-scale and accurate elevation data. Most design projects don’t take the earth surface into account. As this mindset changes, we’ll see an increasing opportunity for high-level accuracy for the design of more efficient systems.
Elevation is a critical component of any earth observation, whether you’re using a bare-earth digital terrain model or a detailed digital elevation model with ground features. As the accuracy of elevation observations have increased, on an ever-broadening scale, this data has grown in importance and utility. I’m certain we’ll see a growing number of interesting applications for elevation data in the next few years as more and better data becomes available.