Climate change has the potential to alter the genetic diversity of plant populations with consequences for ecosystem function. In this study we addressed whether a long-term climate change manipulation has altered the genetic diversity of a dominant C4 grass, Andropogon gerardii, which contributes disproportionately to ecosystem productivity in the tallgrass prairie, using the Rainfall Manipulation Plots (RaMPs). The RaMPs experiment, located at the Konza Prairie Biological Station (Manhattan, KS), is an ongoing rainfall manipulation experiment in which six sheltered plots receive ambient rainfall and six sheltered plots receive the same quantity of rainfall in an altered pattern, consisting of a 50% increase in the interval between rainfall events. The altered treatment significantly reduces mean soil water content and increases variability in soil water content. For comparison to the RaMPs experiment, 12 plots were established in the same spatial array in intact prairie adjacent to the RaMPs (hereafter ‘Out’ plots). In 2007, we collected leaf tissue from 40 tillers within in each plot. Amplified Fragment Length Polymorphism was used to differentiate between genotype families. We examined genetic structure and genetic diversity, at two spatial scales, plot (4 m2) and treatment (cumulative 24 m2). Here, genotypic structure refers to the number and abundance of genotype families, while genetic diversity refers to relatedness between individuals, calculated as distances between individuals on a cladogram using Phylocom.
We found a decade of altered rainfall patterns in the RaMPs significantly altered both genotypic structure and genetic diversity of A. gerardii. At the plot scale no differences in genotypic structure or genetic diversity between RaMPs and Out plots were observed. At the treatment scale, however, genotypic richness, evenness, and diversity were reduced in the altered RaMPs as compared with the Out plots, whereas the ambient RaMPs treatment had only reduced genotypic evenness. In addition, we found individuals in the ambient RaMPs and Out plots were significantly more clustered on the cladogram than is expected from random, whereas, individuals in the altered RaMPs did not differ from a random distribution indicating greater genetic diversity. Our results suggest that despite a reduction of number of A. gerardii genotypes with the altered precipitation treatment, these remaining genotypes are less related to each other. This increase in genetic diversity may be the result of diversifying selection. Moreover, increased genetic diversity could have important consequences for ecosystem functioning, as we have found genetic diversity positively correlates with aboveground productivity in this system.