Meeting the Challenge:
Invasive Plants in PNW Ecosystems
Abstracts
 

 
printable pdf version of all abstracts

Challenges in Predicting the Potential Distribution of Exotics using Habitat       Distribution Models.

Chad C. Jones, University of Washington
Box 352100, College of Forest Resources
University of Washington, Seattle, WA 98195-2100
ccjones@u.washington.edu
Steven A. Acker, Olympic National Park
Charles B. Halpern, University of Washington

Habitat distribution models (HDM) are increasingly used in conservation biology and have the potential to inform efforts to monitor and control exotic species by predicting the potential distributions of exotic species. However, several challenges complicate the use of HDM for exotic species.  First, these models assume that the species is in equilibrium with the environment, which is generally not the case with exotic species.  This has several important ramifications, including increased sensitivity to autocorrelation in the habitat variables and difficulty in validating the models. Second, data on current distributions of exotic species usually come from a variety of sources, often leading to a clumped distribution of sample points and uncertainty in the recorded absences.  Since many modeling techniques are sensitive to the recorded prevalence of the exotic species, the distribution of the sample points can greatly impact results. Third, modeling techniques vary in their assumptions and data requirements, but it is unclear which techniques work best for exotic species.  We illustrate these challenges by modeling the potential distribution of three exotic species on the Olympic Peninsula, Washington.  We tested several modeling techniques and methods to overcome these challenges. We found that there is considerable variation in the predicted distributions using different modeling techniques.  We also found that results were sensitive to the level of clumping in the current species distributions. Results were generally improved by setting minimum distances between plots.  However, significant problems remain and further consideration of these challenges is greatly needed.

    


 

 

top of page