A comparison of habitat use by bats in natural forest fragments and Eucalyptus plantations in Brazilian Savanna

Sónia M.S. Pina, Christoph Meyer, Marlon Zortéa


We compared bat assemblages in semideciduous forest fragments in Brazilian Cerrado with those of eucalyptus plantations (Eucalyptus spp.). Specifically, we assessed differences in species richness, diversity, abundance, and species composition between habitat types and evaluated bat responses with respect to a variety of landscape characteristics and local-scale variables. Bats were sampled with mist nets in four replicates each of eucalyptus plantations and fragments of native Cerrado. Of the total of 209 individuals captured, 75 were caught in plantations and 134 in native forest areas. Cerrado assemblages were characterized by higher species richness, diversity, and evenness compared to those in eucalypt plantations. Gleaning animalivorous phyllostomids seem to be most sensitive to plantation forests, as they were not captured in this habitat. Non-metric multidimensional scaling indicated no strong separation between habitat types and there were no significant differences in species composition based on an analysis of similarities. The proportion of forest cover (both Cerrado and Eucalyptus) in the landscape was the only landscape-scale variable which had a significant influence on species composition. In contrast, species composition was unrelated to geographical distance between forest fragments. A generalized linear mixed effects model for the three most frequently captured species (Carollia perspicillata, Artibeus planirostris, Platyrrhinus lineatus) only showed significant differences in abundance for Artibeus planirostris, the species being more abundant in native Cerrado. Our results demonstrate the impoverishment of bat assemblages in eucalypt plantations and highlight the importance of examining the effects of habitat conversion at the individual species level as responses are often species-specific.

Full Text:



  • There are currently no refbacks.