The composition and diversity of ecological communities of native species in the Anthropocene have been affected by many interacting components of global environmental change such as changes in climate, land-use, ecosystem trophic hierarchies, and invasions of non-native species. Yet, studies of the composite effects of these interacting drivers on plant communities are rare at large scales (e.g., a state) because experimental or direct observational field approaches are not suitable at the scale of large geographical regions with a large spatial variation in both plant community composition and its ecological drivers. We used large-scale databases available for New York State (including USDA FIA, NLCD, PRISM, and NYS DEC deer harvest) within a GLM analytical framework to tease apart the effects of climate (temperature and moisture), land-use (type and fragmentation), and deer browsing pressure on the diversity and composition of native and non-native plant species in forest understories of New York State. We specifically tested if non-native species diversity and composition responded to the environmental drivers in a similar way as native species did, and we quantified what environmental conditions were most likely to lead to large diversity and cover of invading non-native species and declines in native species.
Results/Conclusions
Many plant communities in New York forests contain large proportions of non-native species; maximum non-native species richness was up to 18% of the maximum native species richness (and greater locally). Non-native species were most numerous in areas of intermediate native-species richness, but they declined in areas with high native species richness. The most common non-native species were either invaders of broad concern (Rosa multiflora, Lonicera morowii, Alliaria petiolata, Rhamnus cathartica) or species of less concern but still very common (e.g., Veronica officinalis and Epipactis helleborine, the first and third most common non-native species). While native plant species richness was only weakly linked to climate, deer browse, or surrounding land-use, non-native species richness was significantly related to mean annual temperature (positively) and precipitation (negatively). Thus, warm and dry areas of New York State were characterized by richer and more abundant communities of non-native plants, while cool and moist areas, as well as the areas with low deer browsing pressure and contiguous forest cover, were characterized by low richness and low abundance of non-native species. Considering the IPCC climate change scenarios, these climate refugia from non-native species will likely vanish or dramatically decline in New York State by 2100.