The role of international trade in non-avian terrestrial vertebrate invasions

Background/Question/Methods

The introduction of non-native species by humans is a defining feature of the Anthropocene, accelerating in rate and scale with globalized trade and transport networks. Given the ecological and economic harm that some alien species cause, understanding which species are most likely to become invasive is an urgent question. The best predictor of invasion success is introduction effort (or ‘propagule pressure’) – the rate and abundance at which individuals are introduced. However, estimates of trade intensity are typically used as proxies for introduction effort and invasion risk, while the relationship between trade and invasion success is rarely tested explicitly.

 We investigate the relationship between international trade and invasion success in a global-scale comparative analysis of terrestrial non-avian vertebrates. From multiple existing sources, we have assembled the largest dataset of historical introductions of mammals, reptiles and amphibians yet. We obtained data on the volume of international trade of live specimens in recent years from the US Fish and Wildlife Service (1999-2015) and the CITES Trade Database (1975-2015). Accounting for phylogenetic relatedness, we investigate the relationship between the number of shipments of live individuals and the probability of introduction, the number of introduction locations, the probability of establishment, and the probability of spread.

Results/Conclusions

We find that the probability that species are introduced strongly increases with both US and global trade volume, across mammals, reptiles and amphibians. The number of introduction locations does not increase with either measure of trade volume in mammals, but the number of introduction locations is positively associated with both measures of trade volume in reptiles. In amphibians, the number of introduction locations increases only with US trade. Controlling for introduction effort, the probability of establishment decreases with US trade volume, but is unrelated to global trade volume, while the probability of spread is unrelated to trade volume in both datasets, across the vertebrate classes.

Our results suggest that species traded live internationally are highly likely to be introduced in non-native areas. However, the number of introduction locations and the probability of establishment and spread do not consistently increase with trade volume across vertebrate classes. Therefore, while trade volume is a good proxy for the probability of historical, and possibly future, introduction, it seems not to be highly informative about the probability that introduced species become invaders. In future work, we will investigate whether current findings hold for recent introductions specifically, and investigate the role of specific trades in invasions.