Ecological studies are conducted at a number of spatial and temporal scales. However, the scale of individual studies often depends on the species of interest and the ease with which it can be manipulated. This bias limits our ability to determine how processes occurring at the individual and neighborhood level interact with factors that vary at landscape and regional scales. Previous work has shown that plant establishment along topographic gradients is limited by differences in water and nutrient availability along those gradients. However, processes like competition and facilitation influence plant establishment at a much more local scale. I conducted a series of observational studies and manipulative experiments in order to determine whether the pattern of distribution and abundance of Helianthus petiolaris in a Sandhills mixed-grass prairie is determined by microsite differences, landscape-level variation, or the interaction of processes occurring at each scale. The observational studies revealed that adult H. petiolaris abundance was highest on disturbed ground. Plants were also found more frequently in valleys than anywhere else along the topographic gradient. In a manipulative experiment, seedling survival time increased when the surrounding vegetation was removed but decreased when seedlings were planted on a simulated gopher mound. The decrease in survival on mounds was most pronounced on ridges. These results suggest that manipulative experiments, especially those conducted at more than one spatial scale provide clearer insights for understanding patterns in nature.

Key words: disturbance, Helianthus petiolaris, seedling establishment, spatial scale