Module 6 - Georeferencing

This week's lab was a great deal of fun. Our task was to make two maps of the University of West Florida campus with polygon and line feature classes provided to be geographically adjusted to raster satellite images of the campus. In addition, we were given the coordinates for a protected eagles nest nearby which was to be protected with a buffer zone of 330 feet and 660 feet.

To adjust the feature classes to the raster data, we familiarized ourselves with the Georeferencing toolbar. With this we were able to not only drag the vector data to line up as best as possible with the raster but also set control points between the two layers. Once control points were set, the images will then do their best to distort the image we are attempting to line up (raster) with our geographically referenced layer (vector). When georeferencing, you are given a numerical error value, called an RMS Error, that, in essence, communicates the accuracy of how well your data is lining up. For most successful georeferencing, keeping this numerical value under 15 is ideal. For my first map, showing the northern section of campus, my RMS Error was 9.585924 and, for the southern section of campus, it was 11.989374.

Hypothetically, unless incorrectly placed or the images are incredibly disparate in terms of projected location, the more control points one has, the greater the accuracy of aligning ones images. However, incorrectly placed control points (matching one piece of image to as close to the same location as one can get it) can ruin a projection and greatly increase your RMS Error. 

To take our understanding of georeferencing one step further, we were given a parcel survey for a proposed new building on the campus and we had to georeference it. This was a bit more tricky as we had to hop on GoogleMaps to see where the building is located and then attempt to see through the transparency of a white .jpeg where to place our control points.

We then toyed with the snapping tool and connected a Campus Lane into our roads feature class. Lastly, we set a nested buffer, as explained several modules back, around the eagle's nest. 

Here is my final map:

In the second part of this lab, we took the provided vector and raster data and attempted to make a 3D map of campus. We used Lidar (laser-based) elevation information, which appeared on our map as an odd dataset, and converted it to a raster image which illustrated the elevation in color gradients. 
We then placed our satellite images and vector data on top. Due to the elevation data being connected We then used the extrusion tool to illustrate the maximum height of the buildings and now had a very, dare I say cool, 3D image of the campus buildings and the surrounding area. This was difficult, not in the actual process, but it kept slowing down my program, to the extent that it froze twice. Save often and persevere. 
We then chose an angle we liked and created a 3D map of our work.