Although the expansion of open vegetation within Amazonia was the basis for the Forest Refugia hypothesis, studies of Amazonian biota diversification have focussed mostly on forest taxa. Here we compare the phylogeographic patterns and population history of two sister species associated with Amazonian open-vegetation patches, Elaenia cristata and Elaenia ruficeps (Aves: Tyrannidae). We sampled individuals across Amazonia for both species, and in the central Brazilian savannas (Cerrado) for E. cristata. We sequenced one mitochondrial (ND2) and two nuclear (BFib7 and ACO) markers. We tested for population structure, estimated migration rates and elucidated the historical demography of each species. The Amazon River is the strongest barrier for E. ruficeps and the Branco River is a secondary barrier. For the more broadly distributed E. cristata, there was no discernible population structure. Both species attained their current genetic diversity recently and E. cristata has undergone demographic expansion since the Last Glacial Maximum, The results suggest distinct effects of recent landscape change on population history for the two species. E. ruficeps, which only occurs in Amazonian white sand habitats, has been more isolated in open-vegetation patches than E. cristata, which occupies Amazonian savannas, and extends into the Central Brazilian Cerrado.

Current generation high-throughput sequencing technology has facilitated the generation of more genomic-scale data than ever before, thus greatly improving our understanding of avian biology across a range of disciplines. Recent developments in linked-read sequencing (Chromium 10×) and reference-based whole-genome assembly offer an exciting prospect of more accessible chromosome-level genome sequencing in the near future. We sequenced and assembled a genome of the Hairy-crested Antbird (Rhegmatorhina melanosticta), which represents the first publicly available genome for any antbird (Thamnophilidae). Our objectives were to (1) assemble scaffolds to chromosome level based on multiple reference genomes, and report on differences relative to other genomes, (2) assess genome completeness and compare content to other related genomes, and (3) assess the suitability of linked-read sequencing technology for future studies in comparative phylogenomics and population genomics studies. Our R. melanosticta assembly was both highly contiguous (de novo scaffold N50 = 3.3 Mb, reference based N50 = 53.3 Mb) and relatively complete (contained close to 90% of evolutionarily conserved single-copy avian genes and known tetrapod ultraconserved elements). The high contiguity and completeness of this assembly enabled the genome to be successfully mapped to the chromosome level, which uncovered a consistent structural difference between R. melanosticta and other avian genomes. Our results are consistent with the observation that avian genomes are structurally conserved. Additionally, our results demonstrate the utility of linked-read sequencing for non-model genomics. Finally, we demonstrate the value of our R. melanosticta genome for future researchers by mapping reduced representation sequencing data, and by accurately reconstructing the phylogenetic relationships among a sample of thamnophilid species.

Savannas are ecosystems maintained by fires, characterized by a fire-adapted biota, and savanna patches occur in Amazonia in patches surrounded by tropical forest. Different fire regimes can generate structurally diverse vegetation, and savanna birds are known to be closely related to vegetation structure. However, long-term approaches and interaction of fire with other environmental factors need to be explored for the better understanding of the effects of habitat fire on birds. In an Amazonian landscape composed by savanna and forest, we investigate the effects of different fire regimes in a 12-ha area in three periods through 23 years. We also examined the effects of frequency and extent of fires, tree cover, and distance to forest on bird composition in twelve 3.7ha savannas plots. Birds were surveyed with mist-nets and species were classified as to their habitat use by comparison of registers in forest and savanna plots through visual/acoustical surveys. After 13 years of fire suppression, many forest species colonized the area and some savanna species were lost. Fire frequency and tree coverage affect avifauna communities. The avifauna was sensitive to the occurrence of fires, independent of fire extent. After one fire event in a plot that had been protected from fire for 12 years, some savanna species returned. These results highlight the effects of the fire regime on birds species composition and indicate that many savanna bird species depend on the occurrence of regular fires.

No abstract