Beta diversity of frogs in the forests of New Guinea, Amazonia and Europe: contrasting tropical and temperate communities

Aim To test the hypothesis that animal communities within environmentally relatively uniform lowland forests are characterized by low beta diversity, both in tropical and in temperate areas. Location Lowland forests in the basins of the Sepik and Ramu rivers in New Guinea, the Amazon river in Bolivia, and the Elbe and Dyje rivers in the Czech Republic. Methods A network of 5–6 study sites spanning distances from 20–80 to 300–500 km in each study area was systematically surveyed for all frogs, using visual detection and call tracking. The community data were analysed for alpha and beta diversity. Results Local (alpha) diversity of frog communities was similar in the two tropical areas, New Guinea (mean ± SE of 22 ± 1.4 species per site) and Amazonia (24 ± 1.7 species), but was significantly lower in Europe (8 ± 0.8 species). In Amazonia, 36 of the total of 70 species were recorded from single sites. In contrast, widespread species dominated in Europe, whereas New Guinea exhibited an intermediate pattern with both local and widespread species well represented. The rate of species accumulation across different sites was lowest in Europe, intermediate in New Guinea and highest in Amazonia. The regional species diversity, expressed as the combined number of species from five study sites, was 1.5 times higher than the local species diversity at a single site in Europe, 2.0 times higher in New Guinea and 2.7 times higher in Amazonia. The proportion of species shared between communities decreased with geographic distance in New Guinea and Europe, but not in Amazonia. Main conclusions Frog communities in the lowland tropical rain forests of New Guinea and Amazonia had similar numbers of species, but differed in their beta diversity. More species in Amazonia had restricted distributions than in New Guinea. Both tropical areas had markedly higher alpha and beta diversity than the temperate area in Europe.