 This video will walk you through the steps of taking multiple image tiles and separate raster datasets and generating a new output mosaic raster from these image tiles. In this example, I'm working with over a dozen individual image tiles. Each one of these image raster datasets has four bands and it's an 8-bit raster. As you can see here, they're assembled in individual small rectangular tiles. To generate the mosaic, we're going to use a geoprocessing tool called mosaic to new raster. So from the analysis menu, clicking on tools, and then in geoprocessing, searching for the mosaic to new raster tool. The mosaic to new raster tool is fairly straightforward to use. Starting off, I'm going to specify the input rasters. These are all of the individual image tiles that I want to participate in the output mosaic. Next, I'm going to set the location for my output mosaic. This is the folder that the output mosaic is going to be written to. And then finally, I'm going to specify the output name for my mosaic dataset. You can set the output file format of your mosaic by specifying the appropriate file extension. In this example, I'm entering .img for ErdassImagine, but you could also enter .tif for geotip format. Scrolling down, we see the red asterisk next to number of bands indicating this parameter also needs to be filled out. I know that my input image tiles have four bands, so I'm going to enter four into the parameter setting. Running the geoprocessing tool will generate a new raster dataset. Keep in mind that this can take a long time, perhaps hours or even days, for large collections of raster data. Once the geoprocessing tool has finished running, the mosaic will appear within the table of contents. Here it is above the original grouped collection of tiles. It's always a good idea to explore the properties of your output mosaic just to confirm everything is in good order. So I'll right click on the mosaic dataset, go into properties, to open up the layer properties window. The most useful information is contained within the source tab, and then we'll expand the raster information section. We expect this to be a four band raster, and we can see that it still is. We also have information on the number of columns, number of rows, and we can confirm the pixel size just to confirm that there were no modifications from the original raster datasets. When we compare the new mosaic to the original tiles, we see there are some advantages, particularly as it relates to symbology. Rather than the stretch being applied individually to each image tile, it's applied consistently to the entirety of the mosaic, giving it a more uniformed appearance. Mosaicing your data to a new raster file is very useful in those situations where you need to move the mosaic between software packages. If you don't need to do that, consider using a virtual mosaic, otherwise known as a mosaic dataset.