 This video will introduce you to virtual mosaics, otherwise known as mosaic data sets within ArcGIS. Virtual mosaics are great in these situations where we have multiple individual raster tiles and we want to manage them as a single consistent mosaic. And we can do this without actually having to generate a new raster. As you can see in this particular example, I have a number of individual image tiles and to apply the same symbology to each one of these tiles is cumbersome and difficult. In a mosaic data set, all the input rasters are treated as a virtual mosaic that I can apply consistent symbology and analysis functions. A mosaic data set is stored within a geodatabase. So the first step is to right click on your geodatabase, go to new, and create a new mosaic data set. When you create the mosaic data set, all you're going to do is specify the mosaic data set name along with some parameters. Setting these parameters correctly requires that you know something about your input rasters. For example, you want your coordinate system to generally match your rasters and you want your product definition to be appropriate for the particular data sets you're loading into the mosaic data set. I have four-band aerial imagery and thus the visible infrared product definition is appropriate. Under the product properties, I do have the option to adjust some of the parameters if I see a need to. And then finally, under the pixel properties, I want to set the appropriate pixel type. My data are on side 8-bit and recall at any time if you're in doubt, you can right click on the properties of your input rasters and go into the properties section and have a look at the source information there to confirm the raster information properties. When you click run, the geoprocessing tool generates a new mosaic data set within your geodatabase. This is empty at this time and contains no data. The three components of the mosaic data set, the boundary, the footprint and the image data appear in the table of contents. Now it's time to populate our mosaic data set with our source rasters. We'll go to our mosaic data set, right click on it and choose add rasters. This will launch the add rasters to mosaic data set dialog. I'm not working with a particular sensor so I don't need to adjust the raster type in this particular case, nor do I need to adjust the processing template. However, I am going to change the input data from file to a folder as all my image tiles are located within a single folder. I'm going to navigate to the folder containing my image tiles, select that folder and then click OK to populate that parameter. As you scroll down you'll notice that there are a number of other parameters that can be set. These are optional and please note that some of them such as calculating statistics and building overviews can be done after you've added rasters to your mosaic data set. Running the geoprocessing tool populates the mosaic data set with the rasters. Important to keep in mind, no data are actually being copied to a new location, these are just virtual pointers back to the original tiles. Now that our mosaic data set has been populated with rasters, we can display both the boundary which is the full extent of the rasters and the individual footprint which are the locations of each and every single one of those image tiles. Clicking on a footprint feature will display the attribute information which includes the tile name and other associated attributes. Turning on the boundary allows us to view the full extent of all the tiles within our mosaic data set. As we zoom in, the actual image data becomes available. It's important to note that without overviews which we'll build later, the image data is not displayed until we zoom into the appropriate resolution. You'll notice that as we zoom in pan throughout this mosaic data set, even as we cross tile boundaries, the collection of rasters are treated as a single raster mosaic. In order to see our raster data at all scales, we'll have to build overviews. We'll go back into the mosaic data set, right click, go to optimize, and then choose build overviews. This launches the build overviews geo-processing tool. There are no parameters that need adjusting for the build overviews geo-processing tool, so we'll simply click run, and this will generate the overviews for our mosaic data set. Now you can see that as we zoom in and out, the imagery are available at all scales. Because our mosaic data set treats all our rasters as a single virtual mosaic, any modifications we make to the appearance affect all the rasters within that data set consistently. Here I am adjusting the stretch type, and as you can see it's applied across all the individual tiles, but they're treated as a single consistent mosaic. I can even do this with band combinations, changing the band combinations from natural color to color infrared changes the band combinations from my entire mosaic data set, not just each individual raster. Raster analysis functions work the same way as symbology. I can apply a raster analysis function, and it'll work across the entire mosaic data set, treating it as a single composite raster. So here I am going into the analysis tab, and I'm going to launch a raster function, and I'm going to compute NDVI, the Normalized Difference Vegetation Index, for my entire mosaic data set. I'm choosing the NDVI colorized raster function, I'm choosing my mosaic data set as the input, I'm adjusting the parameters specifically, selecting the appropriate visible and infrared band, and then I'm creating a new layer. Benefits of using a raster data set in combination with a raster function in this case means that I can compute NDVI over a large collection of rasters much faster than if I was generating a brand new mosaic individual data set and then using an NDVI geo-processing tool. I haven't created any new data in this case. My mosaic and my NDVI are entirely virtual. If you need to work with your mosaic data set outside of ArcGIS in another software package, you can simply right click on your mosaic data set, choose data, export raster. This will open the export raster geo-processing tool, where you can select the appropriate format and location for your output raster data set. Mosaic data sets are a great way for working with large collections of image tiles when you want to apply symbology or raster functions seamlessly across all of those raster data sets.