Competition at the population level along a standing crop gradient: a field experiment in successional grassland

I measured competitive responses of experimentally-established populations of the perennial grass, Andropogon gerardi, across a complex gradient of standing crop and species composition in the successional grasslands of southwest Michigan. The goal was to assess whether long-term (three year) population-level responses of Andropogon to competition matched the inferences made from a previous phytometer study that examined transplant responses to competition across this same gradient over a single growing season. Replicate experimental populations of Andropogon were established at seven grassland sites by sowing seed into 0.5 x 0.5 m plots that had been denuded of all vegetation. During the first year of the study, all Andropogon populations were maintained as monocultures by hand weeding. At the end of the first growing season, half of the monocultures were selected for continued weeding and half were left open to invasion by competitors for three years. Invasion of the unweeded populations by neighboring plants varied strongly among sites and was positively correlated with standing crop. Increased susceptibility to invasion and competition resulted in the extinction of the unweeded Andropogon populations at the two most productive sites, supporting the hypothesis that Andropogon is restricted by competition to low productivity sites in these grasslands. The finding that the intensity of competition was positively correlated with standing crop is consistent with the previous transplant study, suggesting that short term experimental assays of competition on the growth of individual transplants may have predictive value for longer-term outcomes of competition at the population level.