Taxonomic, functional, and phylogenetic diversity of anuran assemblages across habitats and seasons in a Neotropical savanna

Differences between habitats and seasonal changes in climate are some of the main factors responsible for shaping diversity patterns and biological community structures. In this context, anurans are considered excellent models for ecological studies due to their high sensitivity to environmental changes. This study analyzed how differences between open and forested formations and seasonality influence the taxonomic, functional, and phylogenetic diversity of anuran assemblages in an area of Neotropical savanna. Parameters of taxonomic (species richness—SR), functional (functional diversity—SES.FD, functional dispersion—FDis and functional redundancy—FR), and phylogenetic diversity (phylogenetic species variability—PSV and phylogenetic species richness—PSR) were used to quantify spatiotemporal changes in the anuran assemblages. Sampling was carried out over 15 months, using pitfall traps and active searches. In total, we recorded 598 individuals belonging to 21 species and seven families. Species composition, SR and PSR differed between habitats and seasons, the latter two being higher in open formations during the rainy season. SES.FD differed seasonally, and was higher in the dry season. FDis, FR and PSV did not differ between treatments. Anurans from forested formations were functionally and phylogenetically clustered during the rainy season, while those from open formations were functionally clustered during the same season. Habitat type and seasonal variation were important in determining the taxonomic, functional, and phylogenetic diversity of the anuran assemblages. However, changes in climate and land use are expected to negatively impact the ecological and evolutionary diversity of these organisms in the Cerrado, reinforcing the importance of conservation policies and actions.

Methods
The study was carried out in a federal protected area (Serra das Araras Ecological Station – SAES), located in the southwest of the state of Mato Grosso, municipalities of Cáceres and Porto Estrela (15°33 – 57°03 N; 15°39 – 57°19 E; Figure 1). The area, which is situated in a transition zone between Cerrado and Amazonia, is composed of a vegetation mosaic. Sampling was carried out in five types of vegetation, classified as open formations (Cerrado sensu stricto and parque de cerrado) and forested formations (riparian forest, semi-deciduous dry forest, and cerrado woodland, also known in Brazil as cerradão). 

Sampling was carried out at ten sampling points (two sampling points in each type of vegetation), at least 200 meters apart. We sampled a total of four points in open formations and six points in forested formations. Data were collected between April 2009 and June 2010, using two complementary methodologies—pitfall traps with drift fences and active searches.

We installed a set of pitfall traps at each sampling point (each set consisting of ten 60-liter plastic buckets, seven meters apart, buried along a straight line. The buckets were inspected daily, early in the morning, for five consecutive days each month, totaling 78 non-consecutive days of sampling. The total sampling effort was 6,820 buckets/day.

At each sampling point, we performed active searches along pre-existing trails (nearly 1.5 m wide). Three to five observers slowly surveyed each trail on foot, during the daytime (between 08:20-10:30h) and at night (between 18:30 to 23:00h), along of all study. All the accessible microhabitats were visually inspected for the presence of amphibians. The sampling effort at each sampling site varied from 4.5 to 20 observer-hours, and the total sampling effort was 122.5 observer-hours.