Fine-scale species pattern affects grassland diversity-invasion relationships

Background/Question/Methods

Altering plant species richness and evenness changes the extent to which individuals are adjoined by conspecific neighbors (self-association), an effect that is often overlooked in studies assessing plant species diversity – function relationships.  Such species self-association may be an important determinant of community-scale responses if the effects of heterospecifics and conspecifics on individual performance are unequal and if responses are driven by interactions among close neighbors. In 2012 the plant Species Pattern and Community Ecology (SPaCE) field experiment was established with transplants (64 individuals m^-2; 1 x 1 m plots; n = 5) of 16 tallgrass prairie species.  The SPaCE experiment simultaneously tests effects of plant species richness (2, 4 or 8; functional diversity controlled), evenness (0.64, 0.8 or 1.0), and pattern on grassland communities.  To separate confounded effects of richness and evenness on species pattern, species were assigned to planting positions within a grid at random or in groups of four individuals (clumped).  All community structure treatments were replicated in weeded and non-weeded plots.  Plant aboveground biomass was collected, dried, and weighed at the end of the first growing season and structural equation modeling was used to assess direct and indirect effects of richness and evenness on invasion.

Results/Conclusions

During the first growing season, plant species richness affected invader relative abundance in non-weeded plots, but this effect was not consistent across evenness and pattern treatments.  In the lowest and highest evenness plots, increasing richness marginally reduced invasion.  In intermediate evenness plots, negative effects of richness on invasion only occurred in plots were species were randomly positioned.  These results indicate that interactions that determine richness effects can occur over short (<< plot-scale) distances and may be sensitive to species pattern.  When accounting for effects of richness and evenness on species self-association through a structural equation model, total effects of richness on invasion only occurred in random plots and arose from a negative direct effect and a marginally positive indirect effect via self-association.  These results indicate that richness likely inhibits invasion due to the addition of more resistant species into any given neighborhood, but this effect may be off-set by greater invasion into areas where species are more interspersed.  The absence of these relationships in clumped plots indicates that effects of richness on invasion are dependent on local species interspersion and we should not expect richness to contribute to invasion resistance in areas that are regionally species-rich, but locally species-poor.