Arboreal habitat use by amphibians and reptiles in lowland and montane rainforests of Southeast Asia

Background/Question/Methods

Tropical rainforests are among the most biologically rich communities on Earth, but their canopies and subcanopies are largely unstudied.  Here we present the first comprehensive study of arboreal habitat use by amphibians and reptiles in the tropics. From February-October 2011, we conducted ground-to-canopy surveys and traditional ground-based surveys across an elevation gradient (900-2100 m a.s.l.) on a forest preserve in southern Luzon, the Philippines and in lowland dipterocarp forests of Singapore. We compared composite distributions for arboreal guilds (all species combined) using a novel kernel density estimation technique to decipher how amphibians and reptiles vertically stratify in a rainforest and how this stratification changes across elevations. Additionally, we compared distributions from mountainous rainforests of the Philippines to lowland rainforest of Singapore and between taxonomic groups (amphibians and reptiles).

Results/Conclusions

We found that over 75% of amphibian and reptile individuals were found above ground, suggesting that they play an underappreciated role in tropical forest dynamics. Amphibians and reptiles were most abundant between the ground and sub-canopy heights for the Philippines (50% of individuals occurred between ground and 10 m height), and this remained consistent for all elevations. The same vertical positioning was observed in Singapore, where 50% of amphibians and reptiles occurred between ground and 17 m height. Differences of structuring by taxonomic group were present. Notably, even representation of amphibians and reptiles occurred throughout the forest stratum in the Philippines, whereas reptiles were the predominant above-ground group found in the lowland rainforests of Singapore. Although there is relative consistency in usage of above-ground habitats between our two localities, patterns of arboreal habitat occupancy are taxonomically and biogeographically variable.  Traditional ground sampling methods were also conducted and compared to our ground-to-upper canopy surveys; our results indicate that over 65% of all individual amphibians and reptiles are omitted from traditional survey techniques commonly employed from the ground. The lack of data derived from canopy-based surveys is likely contributing to poor management of arboreal guilds and their habitats.