Forest area predicts all dimensions of small mammal and lizard diversity in Amazonian insular forest fragments

Context Although hydropower development is one of the primary drivers of habitat loss and insular fragmentation, its impacts on species identity and their functional and phylogenetic roles have often been overlooked. Objectives Here we use an integrative approach, considering taxonomic, functional and phylogenetic dimensions at multiple scales, to understand the processes underlying species (dis)assembly of two taxa exhibiting relatively low dispersal ability: small mammals and lizards. Methods We surveyed 26 islands within the Balbina Hydroelectric Reservoir, and adjacent continuous forest, in Central Amazonia. Each dimension of diversity was related to spatial and habitat variables. We also examined functional composition using community-weighted mean trait values, and community redundancy using functional uniqueness. beta-diversity was partitioned into their richness (beta(rich)) and replacement (beta(repl)) components. Results Functional and phylogenetic alpha-diversities of both taxa mirrored the taxonomic dimension, all of which increased with forest area. Individual small mammal (body mass and matrix tolerance), and lizard traits (body length, heliothermic mode and habitat type) were also predicted by forest area. For both groups, functional uniqueness decreased with forest area, and all dimensions of beta-diversity were predominantly partitioned in beta(rich). Conclusions The environmental filter created by forest area resulted in the low conservation value associated with small forest islands, only occupied by a small set of species comprised by generalist lizards and matrix-tolerant small mammals. On the other side, large forest sites ensured ecosystem resilience to disturbance. To maintain ecosystem integrity, creating myriad small islands over large expanses of floodwaters should be avoided in future hydropower development.