Insectivorous predators, including birds and bats, play crucial roles in trophic cascades. However, previous research on these cascades has often relied on permanent predator exclosures, which prevent the isolation of specific effects of birds and bats, given their different activity patterns throughout the day. Moreover, limited knowledge exists regarding the variations in individual effects of these predators under different biotic and abiotic conditions, such as changes in elevation. To address these uncertainties, our study aimed to investigate the distinct effects of bats and birds on arthropod densities in foliage and herbivory damage in lowland and highland rainforests of Papua New Guinea (PNG). Predator exclosures were established for one month to exclude diurnal or nocturnal predators across 120 saplings (ca. 2.5-4 m tall) selected from two lowland and two highland forests (i.e., 30 saplings per study site) along the Mt. Wilhelm transect in PNG. Arthropods were collected and measured, and herbivory damage was analysed at the end of the experiment. Birds significantly reduced arthropod densities by 30 %, particularly in arthropods longer than 10 mm, regardless of elevation. Additionally, both birds and bats appeared to mitigate herbivory damage in highland forests, with protected saplings displaying up to 189 % more herbivory. Our results support previous studies that have demonstrated the ability of insectivorous predators to reduce leaf damage through the control of arthropods. Furthermore, our approach highlights the importance and necessity of further research on the role of seasons and elevations in trophic cascades.

We conducted predator exclosure experiments at four elevational study sites along the Mt. Wilhelm elevational gradient in Papua New Guinea. The experiments were carried out on 120 saplings from 11 tree species. The predator exclosures, constructed from PVC tubes and agricultural nylon netting, were used to individually exclude birds and bats from the saplings. Each exclosure was accompanied by control saplings. After one month of exclusion, we collected all arthropods from the saplings to estimate their densities and sizes and assessed the extent of herbivory damage. We compared the experimental outcomes with the richness and abundance of vertebrate predators (birds and bats) using previously published data from this location.