Date of Award:

12-2011

Document Type:

Thesis

Degree Name:

Master of Science (MS)

Department:

Ecology

Committee Chair(s)

Karin M. Kettenring

Committee

Karin M. Kettenring

Committee

Charles P. Hawkins

Committee

D. Richard Cutler

Abstract

Globalization has created opportunities for plant and animal species to be transported to novel ecosystems. A subset of these species become invasive species, which are able to persist and spread rapidly to become influential components in their new ecosystems. Invasive species are often able to out-compete co-occurring native species and can alter fundamental ecosystem properties, such as soil nutrient availability or time between wildfires in a region. The United States is estimated to lose $34 billion per year to invasive plant species through costs to control unwanted species and lowered crop and forage production.

We used statistical models to better understand invasive plant species in our study region, the eastern Columbia and western Missouri river basins. These models, referred to as species distribution models, relate site attributes, such as proximity to roads or climatic conditions, to the presence or absence of the species of interest. One assumption of these models is that species are absent from sites for reasons that can be captured by predictor variables, such as cold temperatures or too little rain, rather than by the species' inability to disperse to the sites. Our first objective was to determine whether models performed better at predicting the presence of native species than the presence of invasive species since the latter have had limited time to spread in the area where they have been introduced. We also wanted to determine whether including a measure of the distance to nearby sites occupied by the focal invasive species increased model performance by capturing some of the regional dispersal potential. In our second study, we modeled the distributions of 11 plant species designated as noxious weeds by state regulatory agencies within our study area. Our objective was to determine the site attributes most frequently associated with invasion so we could make recommendations for management action to lower invasion risk in our study area.

We found that model performance was lower for the invasive species than for the native species and that including a measure of nearby occupied sites increased performance more for the invasive species than for the native species. We recommend that researchers consider dispersal limitations as part of their species distribution modeling process. Our results may improve the ability of distribution models to accurately predict invaded areas. In our second study, we found a particular site type that was more likely to be invaded by the majority of study species: hot, dry sites with high grass or shrub cover near roads with high soil nutrient levels and fluctuating stream water levels. Monitoring efforts should focus on sites that fit this description. Management actions that may limit invasion include limiting species' spread on roads, adjusting soil nutrient levels to favor native species over invasive species, and preparing for increased invasion with climate change. Research results will be shared with the Forest Service Rocky Mountain Research Station to be disseminated to land managers in our study region. Project results should help lower the rate of spread of invasive species and make species' control efforts more cost effective by focusing on the most vulnerable sites.

Checksum

08a5e484c26d791e85f7facece81ecec

Comments

Publication made available electronically December 21, 2011.

Download

Share

COinS

Copyright for this work is retained by the student. If you have any questions regarding the inclusion of this work in the Digital Commons, please email us at .