The role of soil heterogeneity in the recruitment of new species in restored prairie

Background/Question/Methods

Soil resources influence plant diversity in tallgrass prairie through their effects on dominant grasses. High soil nitrogen availability often results in high grass productivity and a reduction in the cover and richness of subordinate forb species. Soil depth can also influence species composition by reducing cover of the dominant grasses and allowing higher richness of forbs in shallow relative to deep soil. Native tallgrass prairie communities are heterogeneous in space and time owing to variation in topography, drivers (fire and grazing), and soil resources. Conversion of prairie to row-crop agriculture homogenizes many aspects of this ecosystem. We hypothesized that increasing heterogeneity in a formerly cultivated field would increase niche availability and dimensionality to result in higher species coexistence in restored prairie. We compared responses in cover, species richness, and diversity to four whole-plot soil heterogeneity treatments (control, depth heterogeneity, nutrient heterogeneity, and depth + nutrient heterogeneity) created by manipulating nitrogen availability (3 levels) and soil depth (2 levels). We used plant functional traits (PFTs) of twelve newly seeded species to identify a trait space for each species. We aimed to use recruitment of these species to assess whether soil heterogeneity promoted niche availability and dimensionality.

Results/Conclusions

Consistent with long-term responses in this experiment, the whole-plot heterogeneity treatments did not support different levels of plant diversity or richness at the whole plot level in the year of study. However, subplot factors of soil depth and N interacted to affect species diversity (P = 0.034). The interaction resulted from higher diversity in the shallow soil relative to deep soil under ambient N availability, but not under reduced or enriched N conditions. Diversity was strongly and negatively correlated with the cover of the dominant grass, Andropogon gerardii (P < 0.001). Variation in total richness was due to richness of the newly recruited species, which represented 10 distinct trait space groupings.  There was similar niche availability across all treatments based on similar recruitment richness that was significantly greater than zero for each treatment (P < 0.001), but dimensionality could not be ascertained yet because newly germinated seedlings were difficult to confidently identify. These results suggest that the interaction of soil depth and nutrients have some influence on the plant community, propagule supply influences richness, but variation in cover of dominant species may be the strongest driver of niche availability and dimensionality.