Quantifying vulnerability to plant invasion across global ecosystems

The widely referenced "tens rule" in invasion ecology suggests that approximately 10% of established, non-native species will become invasive. However, the accuracy of this estimate has been questioned, as the original analysis focused on small groups of plant species in Great Britain and Australia. Using a novel database of 9501 established plants and 2924 invasive plants, we provide a comprehensive evaluation of the tens rule and the first empirical analysis of how invasion rates vary across spatial scales, islands/mainlands, and climate zones. We found that invasion rates (the percentage of established species with negative impacts) are highly variable across the globe. Well-sampled environments (those with at least 2000 total non-native species recorded) had invasion rates that ranged from 7.2% to 33.8%. Invasion rates were strongly scale-dependent, averaging 17% at the country scale and 25% at the continental scale. We found significantly higher invasion rates on islands when compared with mainlands, regardless of scale. Tropical ecosystems are often considered to be resistant to invasion; however, our results showed significantly higher invasion rates on both tropical islands and mainlands, suggesting unexpectedly high vulnerability of these species-rich ecosystems. We conclude that the tens rule is a poor general estimate of invasion rates for plants, as calculated invasion rates vary widely and are frequently much higher than 10%. Most locations would be better served by using invasion rates that vary based on the recipient environment. Our updated estimates of invasion rates should be highly relevant for invasive species management strategies, including weed risk assessments, which can be adjusted to identify more species as high-risk in areas where invasion rates are higher. Assuming that 10% of established species will become invasive is likely to substantially underestimate invasion rates in most geographies.