Delineating generalized species boundaries from species distribution data and a species distribution model |
| |
| Authors: | Matthew P Peters Stephen N Matthews Louis R Iverson Anantha M Prasad |
| |
| Institution: | 1. Northern Research Station, USDA Forest Service, Delaware, OH, USAmatthewpeters@fs.fed.us;3. Northern Research Station, USDA Forest Service, Delaware, OH, USA;4. Terrestrial Wildlife Ecology Lab, School of Environment and Natural Resources, The Ohio State University, Columbus, OH, USA |
| |
| Abstract: | Species distribution models (SDM) are commonly used to provide information about species ranges or extents, and often are intended to represent the entire area of potential occupancy or suitable habitat in which individuals occur. While SDMs can provide results over various geographic extents, they normally operate within a grid and cannot delimit distinct, smooth boundaries. Additionally, there are instances where a zone of primary occupancy (i.e., a mostly continuous region where species exists, excluding outliers) is better suited for particular analyses, such as when examining source/sink population dynamics or modeling movement into new habitats. We present a semi-automated method to delineate a generalized species boundary (GSB) from SDM output, which provides a practical alternative to digitizing. This preliminary boundary is then manually updated based on inventory data and historical ranges. We used the method to generate contemporary boundaries for 132 tree species of the eastern United States, which are complementary to the ranges generated by Elbert Little for North America during the 1970s, but are not replacements. The methods we present can broadly be applied to other grid-based SDM to generate GSBs. |
| |
| Keywords: | eastern United States forest inventory and analysis geoprocessing R statistical language tree ranges |
|
|
