An experimental manipulation of vegetation structure: Consequences for desert spiders

We examined the influence of changes in the structure of the shrub layer in a desert habitat on the community of web-building spiders inhabiting this vegetation. Vegetation structure was modified to simulate changes predicted with increasing aridity. We predicted that changing the vegetation structure would affect the diversity and abundance of spiders by means of (1) changing available web supports and (2) changing the abundance of available prey. Using a randomized block design, we performed two manipulations: pruning the shrubs to one-half their height and thinning the plot to one-half its initial shrub density. We surveyed the distribution, abundance, and species identity of spiders before the manipulation, and twice after the manipulation. Potential arthropod prey were censused during the two post-manipulation spider surveys. We found no influence of the treatments on the potential prey. Six weeks after the manipulation, spider abundance was reduced significantly in plots of both treatments, and species diversity was significantly lower in the pruned plots. The reduced species diversity in pruned plots may be explained by the propensity of spiders of different species to construct their webs at different heights in the vegetation. Pruning selectively eliminates the crown of a shrub, and eliminates the potential web sites of species which prefer the crown, while thinning removes entire shrubs, eliminating web sites for all species of spiders equally. Neither spider abundance nor diversity differed among treatments in the second survey, ten months after the manipulation. We suggest that the lack of a treatment effect on spider species diversity is related to the fact that the second survey was in spring, when the cooler microclimate found in tall shrubs was less important than in summer. The lack of treatment effect on spider abundance may be a result of low spider densities, such that web sites were not limiting. The changes in species diversity and abundance are consistent with the hypothesis that the physical structure of the vegetation influences the spider community of the shrub layer, independently of any trophic influences.