Determinants of abundance and species richness among lianas differing in climbing mechanisms
Across all 122 plots, lianas accounted for a total of 4500 individual stems (≥ 2.5 cm dbh). Lianas with an active climbing mechanism accounted for 3686 individual stems, belonging to 592 species and 194 genera. The remaining 814 individual stems represented lianas with a passive climbing mechanism, belonging to 170 species and 86 genera. The final multilevel model (structural equation model), including the density and species richness (Fig. 3, 4) for both active and passive climbing species, was very well supported by the data (Fisher’s C = 0.396; d.f. = 2; P = 0.82) (Table S3). As the SEM results considering the basal area of active and passive climbing species were qualitatively similar to the SEM results for abundance (Appendix S2, Fig. S2), we refer only to the abundance and species richness results in the main text.
The results of our SEM models revealed a clear effect of forest structure and climate on the abundance and richness of active and passive climbing species, after controlling for the biogeographic region (Fig. 3). As the results of the SEM models including the climatic water deficit were similar to those including the mean annual precipitation (Table S3), we refer (in the main text) only to the results including the mean annual precipitation. The gap forest index (stem size distribution) had a positive effect on the abundance and richness of active species, but only on the richness of passive climbing species (Fig. 3 a, b). Canopy height, another metric of forest structure, had a significant negative effect only on the richness of the passive climbing species (Fig. 3 b). Among the climate variables, temperature was the strongest predictor of abundance and richness (marginally significant) of the active climbing species, with a greater abundance and richness of active climbing species being associated with higher temperatures (Fig. 3, 4). The temperature did not have a significant effect on either the abundance or richness of passive climbing species, while precipitation was positively associated with the richness of passive climbing species (Fig. 4 b). Soil fertility (CEC) was negatively associated with the richness of both active and passive climbing species, but not with the abundance of both active and passive climbing species (Fig. 4 a, b). We found no significant indirect effects, via the canopy height, of the climate (temperature and precipitation) and soil fertility (CEC) on the abundance of either active or passive climbing species, nor was there such an effect on the richness of the active climbing species, however, soil fertility, temperature and precipitation did have an indirect effect on the richness of passive climbing species (Fig. 4 b).