Taxonomic and functional approaches reveal different responses of ant assemblages to land-use changes

Land-use change is well documented to cause species loss. However, our understanding of the effects of land-use change on other aspects of biodiversity is still limited. We evaluated if different land-use changes (Eucalyptus plantation and planted pasture) affect ant species and functional groups in similar ways across three Cerrado vegetation types (grassland, savanna and savanna-forest). We found that ant species and functional responses differed with land-use change in relation to frequency of occurrence and habitat specificity and fidelity. Land-use change affected species frequency of occurrence but not functional groups, indicating that species are more sensitive than functional groups to habitat transformation. Native habitats had different indicator species and functional groups compared with converted habitats. However, we did not find functional group indicators of converted habitats in any vegetation type; indicating that there is no specificity and fidelity of functional group to converted habitats and that such an approach is less sensitive to land-use changes. In savanna and savanna-forest, species and functional groups showed the same response in relation to composition with differences between native and converted habitats. Thus, functional groups will be lost when ant species are lost. In grasslands, functional group composition was similar between native and converted habitats indicating a turnover of species within functional groups. We demonstrate that both Eucalyptus plantation and planted pasture affect ant species and functional groups in different ways, with negative impacts both taxonomically and less so functionally. Therefore, we show that the two aspects of biodiversity can respond independently to land-use changes and, hence, the importance of using both taxonomic and functional group approaches to evaluate the effects of land-use change on biodiversity in savanna systems