Connectivity explains local ant community structure in a Neotropical forest canopy: a large‐scale experimental approach

Abstract

Understanding how habitat structure and resource availability affect local species distributions is a key goal of community ecology. Where habitats occur as a mosaic, variation in connectivity among patches influences both local species richness and composition, and connectivity is a key conservation concern in fragmented landscapes. Similarly, availability of limiting resources frequently determines species coexistence or exclusion. For primarily cursorial arthropods like ants, gaps between neighboring trees are a significant barrier to movement through the forest canopy. Competition for limited resources such as nest sites also promotes antagonistic interactions. Lianas (woody vines) connect normally isolated neighboring tree crowns and often have hollow stems inhabited by ants. We used two large‐scale liana‐removal experiments to determine how connectivity and nest site availability provided by lianas affect arboreal ant species richness, species composition, and β‐diversity in a lowland tropical forest in Panama. Removing lianas from a tree crown reduced ant species richness up to 35%, and disproportionately affected species that require large foraging areas. Adding artificial connectivity to trees mitigated the effects of liana removal. Ant colonization of artificial nests was higher (73% occupied) in trees without lianas vs. trees with lianas (28% occupied). However, artificial nests typically were colonized by existing polydomous, resident ant species. As a result, nest addition did not affect ant community structure. Collectively, these results indicate that lianas are important to the maintenance of arboreal ant diversity specifically by providing connectivity among neighboring tree crowns. Anticipated increases in liana abundance in this forest could increase the local (tree‐level) species richness of arboreal ants, with a compositional bias toward elevating the density of broad‐ranging specialist predators.

Do lianas shape ant communities in an early successional tropical forest?

“…Lianas provide nest sites and reliable food resources for ants (Blüthgen et al., ; Philpott, ; Tanaka & Itioka, ), and these resources should be important for ants regardless of forest age. Lianas also provide physical connections among treecrowns that are consistently used as foraging corridors by arboreal ants (Adams, Schnitzer, et al., ; Yanoviak, ). However, the functional role of connectivity should be relatively more important in older forests, where ant communities include more canopy specialists (Dejean et al., ; Tschinkel & Hess, ), and treecrowns typically lack physical contact with each other (Ng, ; Putz et al., ).…”

“…Additionally, connectivity among treecrowns was also not important to the ant communities likely because of the high frequency of ground‐nesting ants, especially in the 2012 survey. Research on liana‐based resources in a nearby older forest (>60 years old) found that connectivity was particularly important to the maintenance of arboreal ant diversity (Adams, Schnitzer, et al., ). The increase in arboreal ant species in the 2015 survey suggests that lianas likely will become more important to the arboreal ant community as this forest continues to mature.…”

“…Thus, changes in liana abundance and species composition during succession (Barry, Schnitzer, van Breugel, & Hall, ; Lai et al., ; Letcher & Chazdon, ) should have substantial bottom‐up effects on animal community structure via changes in liana‐based resources (Estrada‐Villegas & Schnitzer, ). Lianas also influence arboreal animal distributions by connecting neighboring treecrowns (Adams, Schnitzer, & Yanoviak, , ; Chiarello et al., ; Yanoviak, ), which are otherwise commonly isolated in space due to “crown shyness” (the tendency for a physical gap to exist between neighboring trees within a forest canopy; Ng, ; Putz, Parker, & Archibald, ). However, the contribution of lianas to inter‐crown connectivity presumably is minimal during early succession due to the inherently small stature and dense packing of treecrowns in young forest (Aide et al., ; Montgomery & Chazdon, ).…”

“…As individual trees age and grow taller, the resident ant community also changes to include fewer ground‐nesting species (Dejean, Djiéto‐Lordon, Céréghino, & Leponce, ; Tschinkel & Hess, ). In older forests, lianas physically connect isolated treecrowns and support higher local (i.e., per‐tree) ant species richness, specifically benefiting ant species with wide‐ranging foraging strategies (e.g., Neoponera spp., Paraponera clavata ; Davidson, ; Yanoviak & Schnitzer, ; Adams, Schnitzer, et al., , ). However, the ecological effects of lianas on ants are unknown for early stages of forest succession.…”

“…We used a large‐scale liana removal experiment in 17 forest sites to determine the effects of lianas on arboreal ant community structure along a 31‐year chronosequence in a neotropical forest. Based on observations and liana removal experiments in older forests (Adams et al., ; Adams, Schnitzer, et al., ), we hypothesized that lianas positively influence ant community structure (e.g., species composition, species richness, β‐diversity, and species co‐occurrence) in early secondary tropical forests. Accordingly, we evaluated four associated predictions.…”