These are the results of my simulations. Prior to cleaning the results, there was a total of 50,000 home range samples. This was a representative sample of 2500 raster tiles (10.6 %) with 12,500 home range samples for each size category (5, 10, 100, 250 km2).

Once the data was collected, I then cleaned the data by removing any NAs and removing any samples that had an average TRI of 0. I did this because a TRI of zero represents that there was no elevation data within that home range. Once I did this, I had a total of 40,972 samples. Furthermore, I eliminated outlier values by discarding samples that had extremely high values. This resulted in a final data set with 39,513 samples.

I. Percent Difference vs TRI

II. Percent Difference vs TRI (Without Outliers)

III. Distribution of Percent Difference