Trunk Structural Complexity Determines The Diversity Of Bark-Dwelling Spiders In A Tropical Forest

Received 18 May 2020, accepted 9 September 2020Complex environments often have diverse animal communities. The structural complexity of plants has been directly and indirectly responsible for the abundance and diversity of spiders, allowing the establishment of different types of webs and influencing prey availability. Piptadenia gonoacantha and Croton floribundus trees are native species of the Atlantic Forest and have trunks with different structural complexities. The structural characteristics of the former vary according to the stem diameter, while the latter has a relatively simple trunk. This variation makes these plant species a good natural model for natural experiments designed to study how the structural components of trunks can affect the arthropod community, especially spiders. For this purpose, we sampled 30 trees of each species to determine their structural complexity and the abundance and richness of associated spiders. We classified the trees by level of complexity based on the abundance of their structural categories. As expected, the structural complexity of tree trunks explained almost 50% of the variation in species richness and abundance of spiders. Spider diversity, evaluated by the Shannon-Wiever index, was higher in trunks of P. gonoacantha included in the groups of higher complexity. This pattern however, was not observed in relatively simpler trunks of C. floribundus. On the other hand, the species-area relationship pattern was not supported by our data as we observed a nonlinear relationship between trunk surfaces and species richness. Differences in spider richness in the trunks appeared to be caused mainly by their specific microhabitat demands, which differed between groups according to their complexity. For some spiders, the overall trunk structure may play an essential role in the establishment of individuals. For others, specific structures such as loose bark or branches may be essential variables.