Arboreal camera trapping: a reliable tool to monitor plant‐frugivore interactions in the trees on large scales.

Although arboreal camera trapping is a growing field, it has rarely been used for monitoring plant‐frugivore interactions in the trees. Frugivore foraging behavior generally occurs in trees, hence arboreal camera trapping can be a potentially useful tool for frugivory research. We developed a camera trap sampling method to monitor plant‐frugivore interactions during mature fruiting periods. We used this method to monitor 318 individuals (camera sites) of 18 fleshy‐fruit plant species on 22 subtropical land‐bridge islands in the Thousand Island Lake, China. We recorded a total of at least 52 frugivorous animals, including a ground‐foraging bird species (Lophura nycthemera) and several mammals with foraging behaviors in the trees. We also recorded 4399 independent interaction events, including 275 unique plant‐bird interactions. We proposed a framework to classify interaction types and performed a sampling completeness test. We found that a sampling strategy that covered approximately a third of the fruit maturation period when most fruits were ripe was sufficient to sample plant‐frugivore interactions. Our results demonstrated that our sampling method with camera transects is reliable to monitor plant‐frugivore interactions in a fragmented landscape. This study helps to lay the methodological foundation for building networks of plant‐frugivore interactions with arboreal camera trapping on large spatial/temporal scales. As a non‐invasive, labor‐saving, and largely unbiased sampling method, the field application of arboreal camera trapping in different regions can advance the technology of biodiversity monitoring and lead to more accurate biodiversity inventories in arboreal environments. [ABSTRACT FROM AUTHOR]