Environmental and forest structure variables
We used forest structure, climate and soil variables as predictors of liana abundance (Figure S1). Canopy height, as derived from spaceborne LiDAR, has been used as a proxy for forest height (Lefsky, 2010; Simard et al., 2011) and has also been used in macroecological analyses to approximate the amount of vertical habitat and niche space, for example, of climbing palms (Couvreur et al., 2015), lianas (Meyer et al., 2020) and primates (Gouveia et al., 2014). We used the 1-km resolution canopy height map from Simard et al. (2011) and extracted the mean canopy height for each point location correspondent to the plots with information for the abundance of the lianas (Gentry’s transects). We calculated the gap phase index (GP) based on the relative size distribution of stems per each 0.1 ha plot following Feeley et al. (2007). GP was calculated as follows: GP = ln([BA30 + 1] / [BA10 + 1]), where BA30 is the total basal area of all stems with dbh ≥ 30 cm (dbh – diameter at 1.30 m above ground) and BA10 is the total basal area of all stems with dbh ≥ 2.5 cm and ≤ 10 cm (Feeley et al., 2007). For these calculations we considered only the stems of trees and shrubs that could provide more suitable support for the lianas to climb on and which would not include other growth forms such as hemiepiphytes or herbs, which were also sampled in the methodology used by Gentry.
As bioclimatic variables may be highly colinear with each other and as we were primarily interested in the influence of specific climatic variables, per se, we quantified the influence of climate on the abundance and richness of the liana climbing mechanisms by extracting the mean annual precipitation (in millimeters per year) and mean annual temperature (in degrees Celsius) for each plot location from the 30 arcsec resolution CHELSA dataset (Krager et al., 2007). We also included the climatic water deficit, a proxy for the monthly water deficit experienced by plants (Aragão et al., 2007). For edaphic conditions, we extracted the cation exchange capacity (CEC) (a general proxy for soil fertility; cmol (+) kg−1) at 250 m resolution from the SoilGrids website (https://soilg rids.org/).