Table 1.
Characteristics of vegetation in 0.32 ha plots of primary and secondary rainforest (all trees with DBH ≥ 5 cm).
Table 2.
Characteristics of arboreal ant communities in 0.32 ha plots of primary and secondary rainforest (in trees with DBH ≥ 5 cm).
Table 3.
Multivariate analyses of the effect of forest plot (primary and secondary) and tree size traits (DBH, tree height, trunk height, crown height, crown width, total leaves weight; log-transformed data) on ant species composition.
Fig 1.
Sample-based rarefaction curves of the number of ant species in trees.
Ant species richness (a) in 0.32 ha plot of primary and (b) in 0.32 ha plot of secondary rainforest. The curves for the numbers of species observed are shown separately for ant nests, foraging ant species and for both groups combined respectively. Overall diversity of ants in each forest plot is estimated using Chao 2 (mean ± SD).
Fig 2.
Number of ant species per tree in primary and secondary forest plot.
Box-plots show median and mean values per a tree (black line and dot respectively) with 25–75% quartiles and whiskers represent 1.5 interquartile ranges for all species combined (All records), foraging species (F), nesting ant species (N) and species that foraged but did not nest on tree (F-N). Average species richness per tree is significantly higher for F than for N ants in both habitat types (paired t-test, log-transformed data, P < 0.001). See also Table 2 for the mean and SE values.
Fig 3.
Ordination diagram of ant species composition in studied trees.
Ordination based on CCA analysis (see Table 3 for the significance and % of variance related to the ordination canonical axes). Variation of ant community composition (all presence—absence records of ant species in trees) is related to the explanatory variables Forest plot on the first axis (primary and secondary rainforest) and Tree size on the second axis (DBH after logarithmic transformation). Solid symbols indicate tree-nesting species and empty symbols the species found only foraging in trees (in green: occurrence of the species in primary forest, in blue: secondary forest, in black: both forests). The enlarged symbols with species abbreviations refer to the most common ant species (i.e. present in > 20 trees, see S1 Table for their full names).
Fig 4.
Abundance of tree-foraging and nesting ant species in each forest plot.
Total individual abundance in % of number of foraging individuals (left column) and of arboreal nests (right column) collected in trees in primary (a) and secondary (b) forest plot. Different color patterns express the proportions of the five most abundant species and of the rest of the species respectively. Invasive species are cross-hatched (see S1 Table for full species names and their individual abundances).
Fig 5.
Cumulative probability of distances to the nearest nest of tree-foraging species.
Curves based on observed nest data (F-N species records) for primary and secondary forest plots are compared to the area given by minimal and maximal values of 100 curves randomly generated by the permutations of the presence-absence matrix of nests in trees (see Methods for details and S3 Table for the data). Cumulative probability of distances between all trees in each forest plot is indicated by dashed lines.
Fig 6.
Probability of nesting of tree-foraging species in surrounding trees.
Mean nesting probability of ant foragers (F-N species records) is calculated with increasing maximum distance from tree where they forage (but not nest) in the primary and secondary forest plot. Means are shown for observed data and 100 random permutations of the presence-absence matrix of nests, including 95% confidence interval envelopes (2.5% to 97.5 quantile range) of the model (see Methods and S1 Text for details on calculation).