Invaders Main | Noxious Weeds List | Links Database | Queries by: Name, List, Area, and Map

Project Map

Data Sets
    Training Set
    Prediction Set
    Initial Experiments
    Neural Network

        Over the last century we have seen outstanding technological advancements and economic expansion. These improvements have enabled humans to travel faster and more freely, conduct global trade, create possibilities for new recreational activities, etc. Along with these advantages we have experienced negative effects, including the spread of detrimental plants outside of their native habitats to other parts of the world. Initial introduction of plants to other countries could be the result of intentional action such as to benefit the timber industry or for agricultural purposes, or non-intentional such as dispersal as a result of travel and commercial activities. If used for horticultural purposes, some of these plants could escape their cultivated habitats, and become highly abundant. In the case of accidental introduction as a result of travel or trade, it could be quite some time before detecting the new invaders. In the US these plants of foreign origin are generally referred to as exotic species. Once they are widespread in their newly found habitat, they have the potential to cause harm to the native flora, thus becoming invasive species. As a result, invasive species can have negative economic impact and can threaten native biodiversity and ecosystem function (Kolar & Lodge, 2001). Invasives that have high impacts on the economy or the environment may be declared noxious by state or provincial governments and subject to regulation.

        Current mechanisms to prevent species from entry into the United States based on their invasive potential are largely ineffective (Westbrooks, 1991). The Federal Noxious Weed Act provides the main authority for the restriction of “weeds” from entry into the U.S. (Reichard & Hamilton, 1997). Reichard points out that the Act has been largely ineffective for a number of reasons, primarily because it targets only a small list of species known as agricultural pest plants and because of inadequate funding. She outlines four possible strategies for screening the invasive potential of new species, including calling for “an informed estimate…” In particular, the informed estimate strategy would rely on developing a predictive model for evaluation of invasive potential based on what is already known about other invasive species.

        Currently there are 554 known exotic plants in the Idaho plus Montana (Rice, P.M., 2002). Already 89 of them have been declared as “noxious” by the regulatory state agencies in the five Pacific Northwest States. “Noxious weed” is a legal term used by agencies in their respective states, with every state having a somewhat different process for designating plants as noxious weeds. However, these noxious weeds all share the same invasive behavior, including causing significant environmental and economic damage.


        The goal of this project is to develop a predictive risk assessment model that uses currently available data to recognize potential noxious weeds among the new exotic plants reported in Idaho or Montana. More specifically, this project focuses on using and evaluating the effectiveness of selected machine learning models for assessing the noxious potential of the exotic species introduced into the Idaho plus Montana region during the last 50 years (1951-2000). Attribute data for 73 invasive exotics that were found in Idaho and/or Montana before 1951, and have already been declared noxious by at least one of the five northwest states will be used to develop the models.

<< Summary Background >>