INTERACTIVE EFFECTS OF NUTRIENTS AND DISTURBANCE - AN EXPERIMENTAL TEST OF PLANT STRATEGY THEORY

The relative impacts of nutrient availability, disturbance intensity, and interspecific competition on species distribution and growth were studied by applying five levels of nutrients and five intensities of clipping to simple artificial communities (of Dactylis glomerata, Holcus lanatus, Lolium perenne, and Trifolium repens) in a factorial design for 16 mo. This was done to test the hypothesis that the effects of interspecific competition would decline along gradients as the amount of nutrients decreased and the degree of disturbance increased. Nutrients and disturbance, alone and in interaction, produced significant effects on percent cover of all species, and all species achieved maximum percent cover in the undisturbed regime. The experimental design permitted us to distinguish between percent cover reductions caused directly by nutrients and disturbance from those caused by interspecific competition. When grown in mixture, interspecific competition produced marked changes in the distribution patterns of percent cover that were in addition to the direct impacts of nutrients and disturbance. The absolute reduction in percent cover due to interspecific competition was greatest in the undisturbed regime, and for many mixtures this competitive effect was significantly reduced or even eliminated in the highly disturbed areas. Along the nutrient gradient the responses were variable and neighbor dependent, but for all species, the absolute reduction in percent cover due to interspecific competition was generally most severe where percent cover in pure stands was highest. When the data were standardized for differences in percent cover in pure stands and competition was assessed as the percent reduction in percent cover between pure and mixed stands, then the effect of interspecific competition on Trifolium declined along both gradients. The three grasses had a reasonably constant competitive effect across the disturbance gradient, but along the nutrient gradient the highest competitive effects were generally achieved where percent cover in pure stands was highest. We argue that this standardized index is less appropriate because it obscures the obvious differences in interspecific competition effects that are clearly operating within these communities.