Habitat characteristics may override climatic influences on ant assemblage composition: a study using a 300-km climatic gradient

We assessed the relative importance of a variety of climatic and habitat variables in structuring ant communities along a 300-km climatic gradient. Sampling was conducted in semiarid, transitional and cool temperate climatic zones in New South wales, Australia. Ants were sampled at three paired sites of two habitats (pastures and conservation 'remnants') in each of the climatic zones (herein referred to as 'zones') using pitfall traps. Remnants represented original open forests, while pastures were a mix of grassland vegetation and cleared woodland. We tested the effects of habitat type, region (representing different climatic zones) and environmental variables on assemblages using distance-based similarity measures (PERMANOVA and PERMDISP) and canonical analysis of principal coordinates. Assemblage composition differed between habitats and zones, but we found no interaction effects. Assemblage dispersion (between-site heterogeneity) differed between habitats but not among zones. Pasture habitats supported more homogeneous assemblages than remnant habitats. Our findings suggest that habitat type, and structure, homogenise assemblages in pastures, thus overriding the effects of local climate apparent in remnants. As remnants are isolated within the biologically homogeneous pastures, movement of unique species between remnants in response to climate changes may be limited, thus landscape connectivity is likely to be important in reducing species loss.