Week 4: Vector Analysis

In Part 1 of this week's lab, we explored geodatabases, queries, and spatial analysis using a variety of spatial data about infrastructure, environmental features, and parcels in Escambia County.

I forgot to check the metadata and had to go back at the end of the lab once I looked back over the process summary. I couldn't find the metadata at first--it came up with no information when I clicked on the layers in the Contents pane. Eventually I figured out that I had to click on the geodatabase files in the Catalog pane to view the metadata. I then got a message saying the metadata needed to be updated to ArcGIS metadata format to be viewed for the Pensacola_streets_08 and escambia_cellular feature classes, so I had to Google how to upgrade it. It's ironic that I completely forgot about metadata after reading a chapter about how important it is to evaluate metadata early in a project. I've gotten so used to having every step of a lab spelled out in exact detail that I just followed the lab instructions and didn't even think about whether there was anything else to do.

For Part 2 of this lab, we explored buffers, overlays, and spatial queries using a map of roads and water in part of De Soto National Forest, Mississippi. We then created a map of suitable campsite areas that were within a certain distance from both roads and water and not within conservation areas. The buffer distance was 150 meters for lakes, 500 meters for rivers, and 300 meters for roads. To identify these areas, I found the intersection of the areas within the buffers for the roads and water, then used the Erase tool to exclude anywhere within a conservation area. 

After calculating the area of each of these discrete sections that met our qualifications for suitable campsites, I color-coded them by their area. I considered using either Equal Interval or Natural Breaks to divide the areas into classes, but I decided to use Natural Breaks. When I tried Equal Interval, the two largest classes only got one area each, and the upper value for the smallest class was 194 hectares, which is pretty large. Natural Breaks puts a lot more areas in each class except the top class, which still has one area, so the colors look more balanced. Also, Natural Breaks assigns the lowest class an upper value of slightly over 50 hectares, which I think better represents the size of a "small" area. Four classes seemed to be the simplest visually while still showing the differences in sizes. 

I chose a color scheme with hues that were distinct but had a clear ordinal relationship (the colors progress in rainbow order and from lightest to darkest as the value increases). I thought that these four colors with Natural Breaks was aesthetically pleasing because all the areas of the same color are of similar sizes. I didn't like the long decimals in the area values, so I edited the labels in the legend to be simpler without changing which areas were in which classes. To do this, I looked at the attribute table to see what values were the highest and lowest for each class and rounded the largest area in each class up to a multiple of 5 to be the upper value for that class.

Because ArcGIS Pro identified areas that were within the buffer for a river separately from those within the buffer for a lake and those within buffer distance of both types of water, some of the smaller campsite areas on the map are actually connected to a larger area. It seems more intuitive to count them as all part of the same connected area, and I wonder if there is a way to get this result with the tools in ArcGIS. On the other hand, some people might have a preference for lakes or rivers, so this map might be better for helping someone find a campsite in an area with their desired size and water feature.

One difficulty I had was in creating the inset map. I wasn't able to figure out how to create a border around the extent of the main map's area in my inset, so I just made the roads red and zoomed out enough that it looked like a little red square. In the future I'd like to learn better methods for creating inset maps.