| Abstract: | Diverse plant communities are predicted to have higher abundance of predators as compared to species-poor ones. In this study we explored whether this prediction holds true for the abundance of predatory arthropods in forest ecosystems, which have been poorly studied in this respect. We collected ground-dwelling arthropods using pitfall traps from six long-term forest diversity experiments in Finland, Sweden and UK. The effects of tree species diversity on abundance of five main groups of predatory arthropods (ants, spiders, carabids, staphylinids and opilionids) were examined by means of meta-analysis. Overall, the diversity of tree species did not affect abundances of predators with the exception of staphylinids, which were more abundant in mixed stands than in monocultures. However, the effects of stand diversity on predator abundance became apparent when analyses were conducted on tree species basis. Preference for stands containing particular tree species was clear in the case of opilionids, carabids and staphylinids, and these preferences overruled the effects of tree species diversity in comparisons between monocultures and mixed stands containing the same tree species. Tree species diversity effects on predator abundance were furthermore mediated by the tree age, plot size and planting density used in the experiment. Overall, our results show that predator abundance does not increase uniformly with increase in tree species diversity, but rather suggest that predators have distinct preferences for stands composed of particular tree species and that these species preferences may overrule the effects of diversity. |
