Bentgrasses (Agrostis) are a group of native and non-native grasses found in golf courses, agricultural fields, disturbed sites, roadsides, grasslands and other ecosystems. Gene flow between these widely distributed species complicates the potential commercialization of genetically-modified, herbicide resistant (HR) creeping bentgrass. The escape of HR creeping bentgrass could lead to risks that are difficult to predict such as more destructive management practices, genetic pollution, increased ‘weediness’, and loss of community and ecological diversity.  This project aims to predict ecological risk by characterizing: 1) bentgrass distribution in and around the deciduous forests of Connecticut, and 2) the probability that gene flow and seed dispersal will move the HR trait into feral and cultivated bentgrass populations.  A landscape-scale ‘Habitat Suitability Model’ (HSM) is being developed using geospatial information, ecological data, and mathematical modeling (multivariate logistical regression).  Researchers visited 290 random plots (100 m2) within the 8.5 square km HSM study site in Bloomfield, Connecticut, and recorded the presence/absence of bentgrass species and other ecological variables.
Results/Conclusions

The data is being used to create a map indicating the likelihood that bentgrasses occur in any location.  Survey data showed that the presence of bentgrasses was positively correlated with sun exposure.  Fifty four percent of the plots with bentgrasses also contained at least one invasive plant species. Sixty one percent of the plots with bentgrasses occurred in public parks, utility right-of-ways, and other managed areas. The study also showed that bentgrass populations co-occurred with nine state-listed species in the CT Natural Diversity Database. These results indicate that the future commercialization of HR creeping bentgrass could negatively impact management practices for endangered species and invasive plants.  If validated, the HSM could become an important tool to support predictive ecological risk assessments and stewardship for genetically-engineered perennial grasses.  Current projects are working to characterize ecological risk from HR creeping bentgrass by identifying plant communities at risk and quantifying the increased fitness of HR creeping bentgrass under herbicide selection pressure. Supported by grants from USDA Biotechnology Risk Assessment Program and the University of Connecticut.