Scanning the horizon for invasive plant threats using a data-driven approach

Early detection and eradication of invasive plants are more cost-effective than managing well-established plant populations their impacts. However, there is high uncertainty around which taxa likely to become in a given area. Horizon scanning that combines data-driven approach with rapid risk assessment consensus building among experts can help identify invasion threats. We performed horizon scan potential threats Florida, USA—a state influx introduced species, conditions generally favorable for establishment, history negative impacts from plants. began an initial list 2128 non-native known invaders or crop pests. built on previous species scans by developing data-based criteria prioritize 100 assessment. The semi-automated prioritization process included selecting “on the horizon” (i.e., not yet target location noxious weed list) climate matching, naturalization history, “weediness” record, global commonness. derived overall scores evaluating likelihood each arriving, establishing, having impact Florida. Then, following consensus-building discussion, we identified six as risk, ranging 75 out possible 125. globally distributed, easily transported new areas, found regions climates similar Florida’s, native communities, human health, agriculture. Finally, evaluated our final lists biases. Assessors tended assign higher had available information. In addition, biases towards four families certain geographical origin. Our conforming metrics used methodology refined be applied other locations.