 Hello, my name is Hans van de Quast, lecturer at IHE Delft Institute for Water Education. In this lecture I'm going to show you what digital elevation models are that we use in GIS. After this lecture you'll be able to define what is a DEM, a DTM or a DSM, to describe different methods of DEM acquisition, to give examples how DEMs can be used and to describe what data can be derived from DEMs. Digital elevation models can be defined as DTMs, digital terrain models, where we only consider the terrain surface. This is often used in catchment hydrology, where we want to route water over the terrain in catchment. The digital surface model, however, is a DTM, plus all natural and human-made features on top of it, such as trees and buildings, as we can see on the picture. This is often useful for hydraulic modeling. Here we see a digital surface model of the city center of Rotterdam, where we used a 50 centimeter elevation model, and aerial photographs draped over it. You can also generate a digital surface model based on vector data. In that case, the building shapes are much clearer than when we use raster, but we don't have the information of the texture of the buildings. This is data from Rotterdam based on city GML data that can freely be downloaded from the internet. So how do we acquire DEMs? Well, the traditional way is using ground surveying, where we extrapolate from a known XYZ point to other locations where we accurately measure the position using a teodolite. This requires very skilled people, and it's also very laborious to cover a large area, especially in mountains, as you can see on the picture. And in the end, all these points need to be interpolated if you want to use it as a continuous raster in, for example, hydrological modeling. A more digital way is using a differential GPS, where we use satellite information to determine our position very accurately, in combination with a base station. And in this way, we can get centimeter accuracy. But still, the points need to be surveyed, and it's a lot of work to interpolate these points to a continuous raster to be used in modeling. Another traditional way is to use stereofotogrammetry. On the picture, you see a stereoscope, and you see a stereo pair of aerial photographs. They have an overlap. In the overlapping area, we can see through this device in 3D, and there are special rulers that we can use to measure the elevation of points we see on the photographs. Of course, that's also very laborious, and you need good skills to do that. And then you, in some way, need to digitize those points to get a raster out of it. Also very laborious is digitizing contour lines. So many maps have contour lines printed on it, and we can use a device to digitize these contour lines, a lot of manual work, and then we can interpolate those contour lines to a raster that we can use in the models. More modern way is to use LIDAR, or laser automotry. With this method, the laser scanner is mounted to an aircraft, and it sends pulses to the earth, which are reflected back. And the reflection, the time difference of the reflection, is recorded. And from that, together with GPS information and the rotation of the plane, we can derive the elevation of the surface. This generates a lot of very accurate points, which need to be post-processed and also interpolated to get a raster. If we don't have access to all these other acquisition methods, we can always use the radar interferometry method. There has been the shuttle radar topography mission. I'll talk more about it later, which acquired a lot of data, almost full coverage of the earth at an acceptable resolution for catchment modeling, for example. So how do we use DEMs? We can use DEMs to determine the catchment area and to delineate drainage networks, which is covered in a separate video. We can calculate slopes. We can calculate the aspect, which is the orientation of slopes, according to the compass direction. That's very useful for applications where we want to know the amount of solar energy that is received by a slope, so north slopes versus south slopes, for example. We can use DEMs to identify geological structures, because when there are abrupt changes in the elevation, it's an indication of a change in geology. We can use it for viewshed analysis. Viewsheds are the areas that can be seen from a point or to determine which point you can see from a certain area. Very useful for military purposes or for spatial planning. Another use of DEMs is orthorectification. Orthorectification is the geo-referencing of aerial photographs or satellite images, where we take into account the relief displacement that is an effect of these kind of images. We can also use, obviously, DEMs for 3D simulations, like landslides, mass movements. We can use it for change analysis, and we can use it to create contour maps. I will illustrate the use of DEMs using an example from the French Alps. This is an area near Digny-les-Bains, where our students from IHC Delft in the specialization hydrology do their fieldwork. During this fieldwork, the students have to study the hydrological processes in their own study areas, their catchments. Prior to going to the fieldwork, it can be useful to study first the digital elevation models of those catchments in order to understand the relation between hydrology and elevation differences. It can help them in understanding better their study area and to prepare their sampling strategy while they are in the field. Also, when coming back from the field, a digital elevation model can help them in further interpretation of the study area. Here we look at the digital elevation model in the form of a ruster. Each cell represents elevation value, and we use a color ramp to give different elevations different colors. Note that this is a continuous ruster without sharp boundaries, and therefore, ramps are used. Also, make sure that you use intuitive colors. So, blue is normally associated with low areas as well as green, while the darker colors or even white for snow are associated with higher areas. We can also calculate the hill shade in a GIS. In this case, an artificial light source illuminates the scene. Normally, the light source is put in the northwest, which doesn't exist in reality. However, if we put it in the southeast, for example, this will give an inverted relief, as you can see in the animation. A nice visualization trick is to combine this ruster DEM with a hill shade. And in that case, we get this more dramatic view. We can either use transparency, but more impressive is it to use the blending if your software supports this. This is made in QGIS. Elevation data can also be visualized using contour lines. Contour lines can also be derived from DEM rusters. The contour lines are formed by connecting locations with the same elevation. This is done with a certain equidistance, the elevation difference between two lines. Can you see what is used in this case? Here, an equidistance of 50 meters is used. What we can also see in a contour line map is when the lines are close together, that the area is very steep, and when they widen further apart, that it gets less steep. So we can interpret also the shapes of the landscape from a contour line map. We can also visualize elevation in so-called 2.5D. It's not real 3D, because then we need special devices. But the effect of 3D is created by perspective and by shading. In a GIS, we can also visualize this in animations, which gives us further insight in the study area. Besides looking at the color of a DEM, we can also drape an auto-photo or a satellite image over the DEM, like in this case. Here we see the animation of the area around Dignia Le Ben, where we can further interpret the shape of the hills, the geomorphology for hydrological applications. In a GIS, we can also calculate the slope from DEMs. We can choose between slope in degrees or in percentage. Here we have calculated it in degrees and we have blended the result with the hill shade to better interpret the results and the color scale from blue to red. The more red it is, the steeper the slope. Another layer that we can derive from digital elevation models is the aspect. Aspect is the orientation of the slope and usually reflected in compass direction. Note that in this case we need a circular or directional legend, where north and south and east and west are opposite. The aspect is a useful map because the orientation of the slope determines the amount of solar radiation that is received by the slope and that affects the hydrology and growth of vegetation, for example, which also affects erosion and weathering. In this example we see a viewshad. We have calculated the areas that are visible from the center of Dignia Le Ben. These are indicated in red. We can also derive more complex things from DEMs by applying equation. One example is the topographic wetness index, which is a function of the upslope contributing area divided by the tangent of the slope and then the natural logarithm of it. It indicates, as you can see in the layer, which areas are dry and which are expected to be wetter. A GIS also comes with a lot of analysis tools, such as creating transects. In this example we see a transect through an open pit mine, and the red line gives the elevation before using the so-called Phil Sings algorithm that we use in hydrology, which is indicated by the green line. In another video I'll explain how to do the catchment delineation using a GIS. I hope this video was useful for you and you learned a lot about using digital elevation models. If you want to learn more, please subscribe to my YouTube channel or go to GISOpenCourseWare.org to see more free materials and links to the courses that we give at IHC Delft.