Do associations between native and invasive plants provide signals of invasive impacts?

Do invasive plant species act more as passengers or drivers of ecological change in native plant communities? Snapshot studies based on correlations at the site scale ignore longer-term dynamics and variation in how particular invaders affect particular native species. We analyzed patterns of co-occurrence between three invading species (Alliaria petiolata, Lonicera x bella, and Rhamnus cathartica) and 70 native plant species in 94 southern Wisconsin forests at two scales to test four hypotheses. Surveys at these sites in the 1950s and again in the 2000s allowed us to assess how initial plant diversity and site conditions affected subsequent patterns of invasion. Sites with more native species in the 1950s experienced fewer invasions of Lonicera and Rhamnus. However, this result may reflect the fact that more fragmented habitat patches supported both fewer species in the 1950s and more invasions. At the site-level, few negative correlations exist between invasive and native species' abundances. Sites with higher Alliaria densities in the 2000s, however, support fewer native species and lower populations of several declining natives. Rhamnus-invaded sites support lower populations of two increasing species. Association (C-score) analyses detect more associations and more negative associations at the 1 m(2) scale than at the site scale. Most strong associations between invasive and increasing native species are positive while those with declining natives are often negative. Species restricted to specialized habitats rarely co-occur with invaders. Alliaria has more negative associations at fragmented sites where it is more abundant and invasions may be older. Fine-scale invasive-native associations were stronger, easier to detect, and less consistent than those detectable at the site-level. Thus, screening large numbers of local associations using observational data may allow us to identify particular invasive-native interactions worth further investigation. Although invading plants sometimes act as passive passengers, increasing in tandem with certain native plants in response to disturbed fragmented habitats, they may also contribute to the declines we observe in many native species. Monitoring invasions would allow us to assess whether local associations serve to predict subsequent invasive species impacts.