be transparent, open to review and evaluation by experts; it must have a logical framework that includes independent factors identified through critical observation or experimentation, or both; and use of the framework must be repeatable and lead to the same outcome, regardless of who makes the predictions. Many of the systems of prediction, including risk assessments, do not always meet those criteria.
After considering the history of invasions of plants and plant pests in the United States, reviewing scientific knowledge about the factors associated with invasive species, and examining efforts to predict the potential of species to invade, the committee reached the following four conclusions:
Conclusion 1. The record of a plant’s invasiveness in other geographic areas is currently the most reliable predictor of its ability to establish and invade in the United States. The same is true for arthropods and pathogens if plants that they can use elsewhere occur in the United States.
Conclusion 2. There are currently no known broad scientific principles or reliable procedures for identifying the invasive potential of plants, plant pests, or biological control agents in new geographic ranges, but a conceptual basis exists for understanding invasions that could be developed into predictive principles.
Conclusion 3. The inability to predict accurately which nonindigenous species will become invasive stems from a lack of comprehensive knowledge of the events that dictate species’ immigration (arrival), persistence (survival), and invasion (proliferation and spread) in new environments. The requisite knowledge would be based on critical observation of the natural history of nonindigenous species and experiments designed specifically to evaluate nonindigenous species in the stochastic environments they encounter in new ranges.
Conclusion 4. Some data on the natural history of plant pests exist, but they often reside in grey literature and in datasets that are not easily accessible. Data on events that potentially lead to invasions are frequently collected by federal or state agencies in the course of pest surveys and inspections and after releases of biological control agents. Such collections need to be more comprehensive and need to be implemented on a quantitative, statistically sound basis. Datasets need to be organized in a way that allows them to be analyzed and evaluated from the perspective of understanding invasiveness.
In the following recommendations, the committee points to ways of strengthening the scientific basis of predicting the invasive potential of plant pests. The