Error checks and quality control
Luke Zachmann made the following changes to the original (digitized) GIS dataset (stored shapefiles) between 2007 and 2008:
1) Shapefiles were rotated to have a consistent North-South vertical orientation;
2) Species names for large unlabeled or obviously mislabeled polygons were assigned based on species names assigned to the same features in previous or later years;
3) Shapefiles were processed using R and Python scripts to cut polygons and point features at the map boarders and remove any small polygon “slivers” generated accidently while digitizing;
4) Other miscellaneous corrections based on visual inspection of the shapefiles;
5) All species were then classified as either density- or cover-type features. All forbs, with the exception of two species, show up as points features. All grasses as polygon features. All shrubs show up as both point and polygon features. Small, young shrubs without significant canopy cover were mapped in the field as point features, so they remain point-features in the shapefiles. The canopies of larger shrubs were mapped in the field, so these show up as polygon features in the shapefiles. Many of these larger shrubs have stems that are mapped as points features, but linked to the canopy using a common identifier in the shapefile attribute table;
6) Plant names were corrected for synonyms based on the USDA PLANTS Database (http://plants.usda.gov/). Some questionable, infrequent taxa lumped into "spp." categories; and
7) x,y coordinates of each polygon centroid were added to shapefile attribute tables using an R function.
