Forestry practices that remove fallen coarse woody debris (CWD) from forests could alter the abundance and diversity of organisms inhabiting the leaf litter. Such a possible effect is supported by surveys that reveal abundances of several litter-dwelling spiders to be different between sites adjacent to, or far from, CWD. However, this indirect evidence has yet to be fully corroborated by an experimental approach in which the impact on litter spiders is assessed by reducing the amount of CWD.  We conducted a field experiment to test the hypothesis that the presence of CWD affects the density, diversity and community composition of litter-dwelling spiders. Twenty-two 12 x 12-m open plots on the forest floor were randomly assigned to either a Control treatment (no manipulation of CWD) or a Removal treatment from which CWD > 5 cm in diameter was removed first in August and then five more times over the next 12- months. Densities and activity-densities of spiders were determined for litter both near and far from CWD in Control and pre-manipulation Removal plots. In the latter treatment, litter was also sampled over the experiment from areas far from, and near to, where the CWD had been located before it was removed.

Results/Conclusions

Non-experimental studies had indicated that distance from CWD influences the overall density and diversity of litter-dwelling spiders. We observed similar gradients in these two response variables; nevertheless, the removal of CWD had no significant effect on total spider density nor on several standard diversity measures.  In contrast, both removal of CWD as well as distance from CWD influenced the composition of the spider community as measured by Jaccard and/or Bray-Curtis dissimilarities. Analysis of individual taxa reveals a very complex pattern, in which some taxa responded to CWD removal, but displayed no difference in samples taken adjacent to, or far from, CWD.  Some other spider taxa displayed the opposite pattern: a density gradient with respect to distance from CWD, but no density response to the removal of CWD. These results suggest that inferences derived from patterns of density gradients with respect to CWD may not reliably predict the overall impact of CWD on spider densities and community structure in forest-floor leaf litter.