Overlay Querying
This approach differs from predictive modeling in that we examine existing data in the “results” layer(s) to see what properties of other layers are associated with them. This “hindcast” or “inverse” form of modeling is based on common spatial and attribute data in the “cause” and “effect” layers.
Characterizing based on coincidence
- “What the average slope of a pine forest compared to a corn field?”
- “Are south or north-facing slopes wetter? steeper?”
- “Is the range of preferred aspect greater for skunks or for squirrels?”
For this example in the “inverse” demo project, “at what elevation and/or slope are the different species likely to be found?”
In Arc-Speak, a zone is a group of cells or features that all share the same attribute, and a region is a group of cells that are contiguous (or a single feature). Thus a zone can comprise several regions. To examine the characteristics of a zone, we use “Zonal” functions in ArcMap.
You can now plot any of the zonal statistics from the table (I think….), or “join” it to the species map.
Cross Tabulating the attributes of intersecting themes
This allows you to see the range of values for a continuous or discrete theme with the range of values for another continuous (integer-only) or discrete themes. This is called “crossing tables” in some queries. Tools: Zonal Histogram or Tabulate Area in Spatial Analyst/Zonal toolbox
can repeat for slope too
I made plots from these output tables, which are dropped on two different axes below. Although the deer and hawks overlap, we can separate out the frogs based on slope and elevation. Adding more variables might allow you to separate them more effectively
