Published online by Cambridge University Press: 20 January 2017
Concern raised against using highly competitive, exotic, large-statured, perennial grasses with fast growth rates as bioenergy crops has led to calls for risk assessment before widespread cultivation. Weed risk assessments (WRAs) are decision support tools commonly used throughout the world to determine the invasion risk of new plant taxa—primarily used as a pre-entry screen. Here, we compare the common Australian (A-WRA) and newer U.S. (US-WRA) models to evaluate the invasion risk of 16 candidate bioenergy crops and to compare their WRA scores to 14 important agronomic crops and 10 invasive species with an agronomic origin. Of the 40 species assessed, the A-WRA and US-WRA ranked 34 and 28 species, respectively, as high risk, including the major crops alfalfa, rice, canola, and barley. Surprisingly, in several cases, both models failed to effectively parse weeds from crops. For example, cereal rye received scores above (US-WRA) or comparable to (A-WRA) kudzu, a widespread damaging invader of the Southeastern United States introduced as forage. Our results indicate that these models are unable to accurately address broad, intraspecific variation and that species introduced for agronomic purposes pose special limitations to WRAs. This further supports other calls for postborder evaluation (e.g., field testing) following WRA screening. We should be cautious of the role of WRAs in setting policy, as illustrated by this relative evaluation of novel crops.