Community invasion resistance is influenced by interactions between plant traits and site productivity.

Plant communities are predicted to be more resistant to invasion if they are highly productive, harbor species with similar functional traits to invaders, or support species with high competitive potential. However, the strength of competition may decrease with increasing abiotic stress if species more heavily invest in traits that confer stress tolerance over competitive ability, potentially influencing community trait-resistance relationships. Recent research examining how community traits influence invasion resistance has been predominantly focused on single vegetation types, and results between studies are often conflicting. Few studies have evaluated the extent to which abiotic factors and community traits interact to influence invasion along vegetation gradients. Here, we use an in situ seed addition experiment to examine how above- and below-ground plant traits and vegetation type interact to influence community resistance to invasion by a recently introduced annual grass, Ventenata dubia, along a productivity gradient in eastern Oregon, USA. To measure invasion resistance, we evaluated V. dubia biomass in seeded subplots with varying trait compositions across three vegetation types situated along a productivity gradient: scab-flats (sparsely vegetated dwarf-shrublands), low sage-steppe, and ephemeral wet meadows. Trait-resistance relationships were highly context dependent. In wet meadows (the most productive sites), resistance to invasion increased with increasing resident biomass and as community weighted mean trait values for specific leaf area, fine-to-total root volume, and height become more similar to V. dubia's trait values, although these relationships were relatively weak. We did not find evidence that neighboring species influenced invasion resistance in less productive vegetation types, in contrast to our expectations that facilitative interactions may increase with decreasing productivity as posited by the stress-gradient hypothesis. Unlike V. dubia, which heavily invaded all three vegetation types, introduced species with similar trait values, including Bromus tectorum, were not abundant throughout the study area demonstrating V. dubia's unique ability to take advantage of available resources. Our results illustrate how community traits and site productivity interact to influence community resistance to invasion and highlight that communities with lower overall biomass and few functionally similar species to V. dubia may be at the greatest risk for invasion.