Plant species pattern plays a role in community structure effects on productivity

Background/Question/Methods

Plant diversity experiments often overlook effects of species pattern when assessing effects of species richness and evenness gradients on primary productivity.  Because changing richness and evenness levels alters the frequency of intra- and interspecific interactions, plot-scale responses in diversity studies may be driven by fine-scale (<< plot scale) species pattern.  To address the effect of pattern on grassland productivity, the plant Species Pattern and Community Ecology (SPaCE) field experiment was established in 2012 with transplants (64 individuals planted in 1 x 1 m plots) from a pool of 16 tallgrass prairie species.  The experiment is designed to test the effects of plant species richness (2, 4, or 8 species and monocultures), evenness (0.64, 0.8, or 1.0), and pattern (planted randomly or in groups of four individuals) on grassland productivity.  Treatment diversity was maintained by weeding throughout the growing season, and soil moisture levels and percent photosynthetically active radiation were measured monthly.  At the end of the first growing season, aboveground plant biomass was clipped, dried, and weighed. Structural equation models were used to assess the direct and indirect effects of richness and evenness on biomass production. 

Results/Conclusions

In the first growing season, biomass production depended on planted species richness, evenness, and pattern.  In two and eight species plots, intermediate evenness plots were the most productive.  In four species plots, the most even plots were the most productive.  Clumping increased biomass production in low and intermediate evenness plots, but negatively affected biomass production in the most even plots.  Differences in plant biomass production among treatments were consistent with selection, but not complementarity effects.  Richness positively affected selection, and selection was greatest in the most even plots.  In contrast to expectations, selection was lower in clumped plots, indicating that some species mixture and monoculture yields were more inconsistent when species were aggregated.  Although there were not indirect effects of richness and evenness via species self-association on biomass production or selection, total effects associated with richness and evenness depended on initial planting pattern for both responses.  Inconsistencies between random and clumped model effects in conjunction with treatment effects suggest that pattern is contributing to community structure effects on grassland productivity and selection.