Functional responses of avian frugivores to variation in fruit resources between natural and fragmented forests

Plant-frugivore interactions and morphological traits Observations of plant-bird interactions: We sampled plant-frugivore networks at six study sites covering all three elevations (1000, 2000 and 3000 m) and two habitat types (natural and fragmented forest). Every study site comprised three plots, resulting in a total of 18 study plots. Each plot was sampled once in the more humid and once in the less humid season per year over two years (2014 and 2015). During each observation period, we observed each plot for 25 h on four consecutive days between dawn and noon. Each study plot was observed for a total of 100 h, resulting in 300 h for every study site and 1800 h for all 18 study plots. We recorded plant-frugivore interaction events by direct observations with binoculars within a 30 x 100 m transect within each plot. We walked freely through the plot and observed each fruiting plant species for a similar amount of time. The study plots ran along steep slopes, which enabled the observation of frugivore activity in the canopy. Only legitimate seed dispersal events were considered, therefore only interactions with swallowed, pecked or transported fruits of a fruiting plant were recorded. Seed predation events were excluded from the analysis (< 1% of the observed interactions). In order to quantify resource switching between natural and fragmented forests, we selected the subset of bird species that occurred in both interaction networks (i.e., natural and fragmented forest) at each elevation (i.e., 63 species out of 133 species from all interaction networks). We then excluded those birds that were recorded just in a single interaction event in either of the two forest types. Consequently, plant species considered in this analysis were those ones consumed by the subset of birds for each habitat type at each elevation. This resulted in a total number of 52 bird species feeding on 83 plant species that we used for all subsequent analyses. Most bird species were present at only one elevation; only four out of the 52 species were recorded at two different elevations. Species morphological traits and metrics of functional diversity: We measured traits of all fleshy-fruited plant species based on plant individuals collected on the study plots. Fruit length, width and mass were measured on 15 fruits per plant species to obtain average fruit traits per plant species. In the field, we estimated the height of every individual fruiting plant to calculate the average height of each fruiting plant species. Bill and wing traits of birds were measured on four specimens (two female and two male) of each species in museum collections (Natural History Museum, Berlin, Germany; Museo Ecuatoriano de Ciencias Naturales, Quito, Ecuador; Zoological Research Museum Alexander Koenig, Bonn, Germany; Zoological Museum of the University of Copenhagen, Denmark). Bird body mass was compiled from Dunning (2007).