Nitrogen (N) additions have decreased species richness in hardwood forest herbaceous layers, yet the functional mechanisms for these decreases have not been explicitly evaluated. We tested two hypothesized mechanisms in the hardwood forest herbaceous layer of a long-term, plot-scale fertilization experiment in the central Appalachian Mountains, USA - the interspecific competitive exclusion (ICE) and assemblage level thinning (ALT) hypotheses. Under N additions, ICE argues that species superior in nutrient acquisition, growth rate, and other growing strategies will displace species with an inferior ability. With increased soil fertility, the superior species suppresses the growth of the subordinate species and different mortality rates between the two emerge. In contrast, ALT contends that mortality is equal among all species, and that the change in species composition under increased fertility is an effect of enhanced density-dependent mortality, where uncommon species are lost by chance. Using a bootstrap random thinning approach, we simulated changes in species densities under ALT and compared the simulated densities to the observed densities among N-fertilized (+N), N-fertilized and limed (+N+L), and reference (REF) plots in regenerating forest stands. We also tested the response of nitrophilic species by creating a nitrophily database from published literature on each species.
We found a decline in richness among +N and +N+L treatments, and determined that the majority of species exhibited changes in their status as competitors because they occurred at densities that were either higher or lower than expected due to random thinning. Differences in competitive ability was also observed between +N and +N+L treatments, providing evidence that species can respond to either the fertilization or acidification effects of N. However, any competitive benefit conveyed by liming appeared to be short lived. Species identified as nitrophilic responded to N additions by achieving greater densities in the +N and +N+L treatments, when compared to REF. Additionally, deviations from random thinning became larger through time and were more pronounced when REF was compared to +N treatments. However, a radical shift in community composition was not observed after 15 years of treatment, suggesting that declines in richness under N fertilization are due to the dominance and competitive exclusion by a small number of nitrophilic species. Overall, our results support the N homogenization hypothesis and indicate that lime additions may not mitigate the loss of species richness and biodiversity in hardwood forests due to N deposition.