In the southeastern USA, harvest of pine straw sometimes involves mechanical raking of natural Pinus palustris (longleaf pine) communities. Little is known about the effects of raking nor how these effects may vary in time and space. In a two yr experiment, we examined the effects of mechanized raking on Pinus palustris dominated communities (scrub oak, dry savanna, and mesic savanna) by monitoring vegetation at seven spatial scales (0.01-100 m²). We measured floristic similarity and the proportion of species initially present that were gained (i.e. new species) or lost during four sampling periods. Relationships between spatial scale and these community attributes were analyzed using a repeated measures approach and functional response curves. Spatial scale clearly affected observed rates of species loss and floristic similarity; losses declined and floristic similarity increased as scale increased. We relate these patterns to expanding population sizes with scale and our inability to detect species reductions in large populations. Scale had little influence on species gains. The effects of raking did not differ across scales, but raking caused greater mean losses of species and greater mean changes in floristic similarity when mean values were calculated over all scales. Raking also increased the mean rate of species gains in the mesic savanna during one period. Otherwise, interaction effects of community and raking were largely absent from both mean values and response curves. Despite significant short-term effects of raking, changes in species richness were minor.

Nomenclature: Kartesz (1994).