High native species diversity generally suppresses non-native invasions, but many ecosystems are now characterized by non-native assemblages that vary in species diversity. How this non-native species diversity affects subsequent invaders and its environmental dependence remains unclear. Here, we conducted a plant-soil feedback experiment to investigate how non-native plant species diversity affects the growth of subsequent non-native plants, the role of soil microbes in this process, and the dependence of these patterns on drought. We found that under well-watered conditions, the biomass of subsequent invaders was higher with soil inocula generated by high non-native diversity, which was associated with higher arbuscular mycorrhizal fungal richness. However, under drought conditions, the biomass of subsequent invaders did not depend on soil inocula generated by non-native diversity. Our study reveals soil microbial legacies likely contribute to the commonly observed co-occurrence of multiple non-native species in nature and the importance of environmental conditions for these effects.

Fluctuating resource availability plays a critical role in determining non-native plant invasions by mediating the competitive balance between non-native and native species. However, the impact of fluctuating resource availability on interactions among non-native species remains largely unknown. This represents a barrier to understanding invasion mechanisms, particularly in habitats that harbor multiple non-native species with different responses to fluctuating resource availability. To examine the responses of non-native plant species to nutrient fluctuations, we compared the growth of each of 12 non-native species found to be common in local natural areas to nutrients supplied at a constant rate or supplied as a single large pulse in a pot experiment. We found that seven species produced more biomass with pulsed nutrients compared to constant nutrients (hereafter ‘benefitting species’), while the other five species did not differ between nutrient enrichment treatments (hereafter ‘non-benefitting species’). To investigate how nutrient fluctuations influence the interactions among non-native plant species, we established experimental non-native communities in the field with two benefitting and two non-benefitting non-native species. Compared with constant nutrient supply, the single large pulse of nutrients did not influence community biomass, but strongly increased the biomass and cover of the benefitting species and decreased those of the non-benefitting species. Furthermore, the benefitting species had higher leaf N content and greater plant height when nutrients were supplied as a single large pulse than at a constant rate, whereas the non-benefitting species showed no differences in leaf N content and were shorter when nutrients were supplied as a single large pulse than at a constant rate. Our results add to the growing evidence that the individual responses of non-native species to nutrient fluctuation are species-specific. More importantly, benefitting species were favored by nutrients coming in a pulse, while non-benefitting ones were favored by nutrients coming constantly when they grew together. This suggests that nutrient fluctuations can mediate the competitive balance among non-native plants and may thus determine their invasion success in a community harbouring multiple non-native plant species.