Wednesday, October 19, 2016

Creating non-spatial and "not quite true" spatial figures in GIS

I recently was tasked with creating figures to show some variables collected along a trapping grid. The problem with these data were that the trapping locations were very close to one another yet the distance among the trapping grids was not.  As a result, it was very nearly impossible to show meaningful differences at the scale of the entire study. In the figure on the right you can see the location of the trapping grids.  Each grid had 64 stations arrayed in 16 columns and 4 rows. I've removed any geographic information that may suggest the location of this study for the sake of privacy.

The solution?  What I call a "not quite true" spatial map.  In the figure below we see that each bold box shows a trapping grid  and each smaller box is a trapping station. The colors represent the intensity of some value.  It may be the number of animals caught in the traps at the locations or some habitat variable having to do with plant cover or soil type. The reason why it is "not quite true" spatial is that distances among trapping grids is much larger than they are in real life and distances between individual traps isn't always exactly even. Nonetheless it shows spatial patterns in a succinct and compact form.

To make this map all I had to do was create relative row and column X and Y coordinates. That table was then imported into ArcMap as X and Y data. In this example there were a total of 36 rows and 48 columns. Although this workflow could have taken place using R or python I found that it was quite easy to accomplish in ArcMap.

This is a nice reminder of how GIS can be a powerful tool for all sorts of visualizations, not just for maps. Using GIS we can very easily change color schemes using different kinds of classification s (i.e. natural breaks, quantiles, equal intervals, etc.), edit individual lines and polygons, convert from raster to vector formats.  Some of this stuff is tedious to do in other types of software.

We're probably all familiar with some examples of "not quite true" spatial maps.  Subway maps are a  prime example.  They are designed to show relative space, but distance isn't always accurate in the true geographic sense.  However, we can also take raster GIS outside of the realm of normal geographic space and actually use GIS for displaying things like time series or even data space.  More on that later.

No comments:

Post a Comment