 This video will introduce you to raster data. Raster data is composed of individual cells or pixels. Each one of these pixels will have a single raster value. Raster data is always numeric but a numeric data can either be discrete or thematic data in which the value represents a code, for example land cover classification, or it can be continuous data like an elevation model in which each pixel value represents the ground elevation. Let's dive into these different types of raster data. In this ArcGIS Pro project I have two raster datasets. One is a DEM or digital elevation model. The other is a land cover data set in which each cell has a value of 1 to 7 corresponding to a unique land cover code. The DEM is an example of a continuous raster data set and the DEM a value of 100 is greater than a value of 50. It represents twice the height in elevation. Going over to the source tab in the layer properties and expanding the raster information section gives us more information about this raster data set. We see that this raster data set is stored in a file geodatabase. It's got a cell size of 1.5 by 1.5. If we go to the projection information we know the units are feet. We also see that it's 32-bit. This means the raster data has 2 to the power 32 possible values. The statistical information gives us the mean max min and standard deviation of the DEM. Moving over to the properties for the land cover data set we see that although it's still a raster data set it has some specific differences compared to the continuous raster data that's the DEM. It's still the same cell size stored in file geodatabase but in this case the pixel depth is 8-bit and it's an unsigned integer. This means the land cover data set can only have 0 to 255 or 256 possible values and they can't have decimal places. The statistics while there are really meaningless because each pixel code from 1 to 7 represents an individual land cover clause. When we zoom in far enough to any raster data set we'll see the pixel outlines. When we click on those pixels it will return 2 values. The actual pixel value which is the top number, 6 right now, and the bottom value which is the stretch value that's the display value in your computer monitor. So the actual value of the cell is the top value, the bottom value is just the display value and that depends on the symbology. Moving over to the DEM we see that the pixel values have decimal places because it's a 32-bit floating point data and we see that the stretch value has a much narrower range because your computer monitor can't display that fidelity of data. In order to symbolize your raster data effectively you have to understand what your raster data represents and how you want it to appear to the end user. Not all symbology options are available for all raster data sets so for example the DEM because it's a continuous data set has too many unique values to symbolize. A stretch is a good option but I could also use a discrete number of colors as you see me doing here. In most cases continuous raster data like DEMs are going to be best represented using a stretch symbology type. There are a number of things you should pay attention to when you're using the stretch symbology. First is the color scheme. Here I'm going to choose a standard elevation gradient. The second is the display background value or no data value. Although in this case my DEM doesn't have any zero values it's a good idea to turn that off just in case we do. Finally the stretch type. The stretch type does not alter the pixel data but it will change the way your data are displayed. Different stretch types may be appropriate in differing situations. It all depends on what you're trying to show using your raster data. Keep in mind that changing the stretch type just changes the way the data are displayed. It never alters the actual pixel values. Those are always the same. When it comes to symbolizing discrete or thematic raster data like my land cover classification it really makes sense to use unique values especially in this case because I only have seven different land cover codes. When I click on unique values in symbology it prompts me to build a raster attribute table. Raster attribute tables are a prerequisite for symbolizing raster data by unique values and can only be built if you have thematic integer raster data. We can now open up the attribute table for land cover layer and see that for each cell value one through seven we have the count or the number of cells corresponding to each value. There's no attribute table for the DEM nor is it possible to build one due to the large number of unique values. The codes for my land cover data set aren't very meaningful by themselves so I'm going to go into the symbology tab and adjust the color and provide a description for each one of my seven land cover types. Finally I'm going to adjust the transparency so that the imagery underneath displays through. In this video we introduce you to raster data. Raster data is nothing more than a collection of cells. Each cell in a raster data set is the same size and contains a single digital value. In the case of discrete or thematic data sets those digital values may correspond to some coded value like a land cover class. Continuous raster data sets will display some sort of measured phenomenon. It could be elevation or air pollution. Symbolizing your raster data effectively requires that you understand the type of raster data that you're working with.