Secondary Succession and the Pattern of Plant Dominance Along Experimental Nitrogen Gradients

In 1982, experimental nitrogen gradients were established on both existing and disturbed (disked) vegetation in three fields (abandoned 14, 25, and 48 yr) and on existing vegetation in native oak savannah. Each of these seven gradients contained five or six replicates of each of nine treatments that differed in the annual rate of nitrogen addition. In none of the fields did plant biomass, height, species richness, or light penetration respond to addition of P, K, Ca, Mg, S, and trace metals. In contrast, plant biomass and height increased significantly, and light penetration and species richness decreased significantly, with added nitrogen along all seven gradients. On average, >60% of the species had been displaced from high—nitrogen treatments by 1985. Nitrogen addition led to a period of transient dominance by certain species. Species that reached peak relative abundance in high—nitrogen treatments in 1982 tended to be rare in all but the low—nitrogen treatments by 1985. In contrast, the relative abundances of most species that dominated the high—nitrogen treatments in 1985 did not increase along the nitrogen gradients in 1982. The relative or absolute abundances of most common species changed significantly along the experimental gradients in at least 1 yr. By 1985, many common species were differentiated in their distributions along the seven gradients. In general, early successional annuals and short—lived perennials and plants of short stature at maturity reached their peak abundance in low—nitrogen plots, whereas plots, receiving high rates of nitrogen addition were dominated by long—lived herbaceous and woody species that are taller at maturity. A survey of 22 old fields at Cedar Creek, Minnesota, showed that total and available soil nitrogen increased during succession and that major species had individualistic, fairly Gaussian distributions along this temporal nitrogen gradient. The distributions along the experimental gradients of most of the common species were consistent with the pattern observed in the old—field survey, demonstrating that nitrogen influences the pattern of secondary succession at Cedar Creek. The major exception was Agropyron repens, and early successional grass that dominated high—nitrogen treatments on six of the seven gradients. Comparisons of species responses on the disturbed plots with those on plots of existing (undisturbed) vegetation showed that, by 1985, most species responded similarly to the nitrogen gradients despite great differences in their initial abundances. For instance, Agropyron repens, an initial dominant of Field A but rare in Fields B and C, was dominant in the high—nitrogen treatments in both the disturbed and undisturbed plots of these three fields. It invaded into and rapidly increased in abundance in numerous high—nitrogen plots from which it was initially absent. Schizachyrium scoparium declined along the nitrogen gradients both in undisked plots in which it was initially dominant and in disked plots in which it was initially rare. Such similarities suggest that the outcome of interspecific interactions among old—field plants is highly dependent on nitrogen supply rates, but fairly independent of initial plant abundances.