Species Richness of Experimental Productivity Gradients: How Important is Colonization Limitation?

The biodiversity of a site should depend on the interplay of local colonization (gain) and extinction (loss) rates, but few theoretical explanations of diversity patterns on productivity gradients have included effects of productivity on colonization. In an 11—yr study, experimental increases in productivity via nitrogen addition generally led to decreased species richness in four grassland fields. Decreased diversity in productive plots was caused as much by lower rates of species gain as by greater rates of loss of existing species. Annual grasses and forbs had high gain and loss rates, but these were independent of productivity. In contrast, the rates of gain of perennial grass and forb species declined with productivity, and their rates of loss increased. Species richness was dependent on litter mass and on light penetration, but not on aboveground living plant mass, suggesting that there was no direct effect of productivity on diversity. Diversity in periodically burned prairie, which had low litter mass, was independent of productivity by the 10th and 11th yr of the study. Results suggest that diversity is lower in productive grasslands because accumulated litter, and possibly lower light penetration, inhibit germination and/or survival of seedlings, and thus decrease rates of establishment by new species. Higher productivity also leads to higher rates of loss, presumably via competitive displacement, of existing species. Results do not support the hypothesis that soil N heterogeneity controls diversity on the temporal and spatial scales of this study.