Community level processes such as competition, facilitation, and community assembly interact at the phenotypic trait level. Dominant species, due to their ubiquity in the ecosystem, have been shown to contribute disproportionately to community and ecosystem processes, including above-ground productivity and invasion resistance. Genotypic variation within the dominant species may account for the overall phenotypic trait variation found in a given community. Studies have shown that genetic variation can influence community species diversity, and conversely, similar mechanisms may allow species diversity to affect the genetic variation of a given population. Thus, understanding the relationship between all three levels of diversity - genotypic variation, phenotypic variation, and community diversity - is critical to determining what governs community and ecosystem level processes. For our study, in 54 1-m² intact tallgrass prairie plots, we measured the genotypic and corresponding phenotypic variation of individuals of the dominant C₄ grass species (Andropogon gerardii), plant community richness and composition, and aboveground productivity (a measure of ecosystem functioning in this system).

Results/Conclusions Using multiple regression analysis, we found that the four most important drivers of productivity at the plot level were relative abundance, relative growth rate, and variation in end of season weight of A. gerardii, and community family richness. When holding species richness constant (~11-13), family richness - an indication of phylogenetic and trait diversity - was positively correlated with productivity (r² = .088, P = 0.016). We found support for previous findings that the relative abundance of the dominant species was positively correlated to productivity (r² = 0.126, P = 0.003). We also found that relative growth rate (r² = .054, P = 0.015) and variation in end-of-season weight of A. gerardii individuals (r² = 0.044, P = 0.069) had a significant impact on productivity. This suggests that phenotypic variation, and possibly genotypic variation, within a dominant species may be an important determinant of ecosystem functioning in this system. We did not find significant relationships between genotypic variation and our tested parameters, suggesting that either phenotypic plasticity plays a stronger role in determining trait variation in the population or that more detailed field and greenhouse studies are needed to tease apart possible genotype by environmental interactions.