Biotic constraints on the invasion of diVuse knapweed (Centaurea diVusa) in North American grasslands

Knapweeds (Centaurea spp.) are among the most invasive of non-indigenous plant species that have colonized western North America over the last century. We conducted a 4-year experiment in a reconstructed grassland to test hypotheses related to the ability of grasslands to resist the invasion of diVuse knapweed (C. diVusa). We experimentally invaded C. diVusa and three native species into areas where we manipulated soil nitrogen (N) and phosphorus (P) availability and removed extant grasses to reduce competition. We evaluated the growth response of these species to these resources and competitive manipulations. Of the native species that were experimentally added, only one species, Ratibida pinnata (prairie coneXower), established in any numbers. Establishment values in intact vegetation were low for both species, but establishment by C. diVusa (0.02%) clearly outperformed that of R. pinnata (0.001%). Under reduced grass competition, establishment was enhanced, but the values for C. diVusa (0.68%) were not statistically diVerent from those of R. pinnata (0.57%). Neither species performed better under higher soil nutrients in the presence of competing grasses. In plots with both species, biomass of the two planted species was positively correlated, but the biomass of both species was negatively correlated with non-added weedy species. Subsequent harvests of C. diVusa indicated that establishment was enhanced in treatments with higher soil nutrients but that the biomass of these plants could only be enhanced when plant competition was also reduced. These results indicate that C. diVusa can establish in intact grasslands at rates higher than natives, but opportunism rather than competitive ability best describes the invasiveness of C. diVusa. Thus, the mechanisms contributing to the establishment of this knapweed species are diVerent from factors identiWed as contributing to the dominance of this invader.