Evolutionary history of marginal habitats regulates the diversity of tree communities in the Atlantic Forest.

BACKGROUND AND AIMS: The Atlantic Forest biodiversity hotspot is a complex mosaic of habitat types. However, the diversity of the rain forest at the core of this complex has received far more attention than that of its marginal habitats, such as cloud forest, semi-deciduous forest or restinga. Here, we investigate broad-scale angiosperm tree diversity patterns along elevation gradients in the south-east Atlantic Forest and test if the diversity of marginal habitats is shaped from the neighbouring rain forest, as commonly thought. METHODS: We calculated phylogenetic indices that capture basal [mean pairwise phylogenetic distance (MPD)] and terminal [mean nearest taxon distance (MNTD)] phylogenetic variation, phylogenetic endemism (PE) and taxonomic and phylogenetic beta diversity (BD and PBD) for 2074 angiosperm tree species distributed in 108 circular sites of 10 km diameter across four habitat types i.e. rain forest, cloud forest, semi-deciduous forest and coastal vegetation known as restinga. We then related these metrics to elevation and environmental variables. KEY RESULTS: Communities in wetter and colder forests show basal phylogenetic overdispersion and short phylogenetic distances towards the tips, respectively. In contrast, communities associated with water deficit and salinity show basal phylogenetic clustering and no phylogenetic structure toward the tips. Unexpectedly, rain forest shows low PE given its species richness, whereas cloud and semi-deciduous forests show unusually high PE. The BD and PBD between most habitat types are driven by the turnover of species and lineages, except for restinga. CONCLUSIONS: Our results contradict the idea that all marginal habitat types of the Atlantic Forest are sub-sets of the rain forest. We show that marginal habitat types have different evolutionary histories and may act as 'equilibrium zones for biodiversity' in the Atlantic Forest, generating new species or conserving others. Overall, our results add evolutionary insights that reinforce the urgency of encompassing all habitat types in the Atlantic Forest concept.

Phylogenetic structure of understorey annual and perennial plant species reveals opposing responses to aridity in a Mediterranean biodiversity hotspot.

Aridity is a critical driver of the diversity and composition of plant communities. However, how aridity influences the phylogenetic structure of functional groups (i.e. annual and perennial species) is far less understood than its effects on species richness. As perennials have to endure stressful conditions during the summer drought, as opposed to annuals that avoid it, they may be subjected to stronger environmental filtering. In contrast, annuals may be more susceptible to interannual climatic variability. Here we studied the phylogenetic structure of the annual and perennial components of understorey plant communities, along a regional aridity gradient in Mediterranean drylands. Specifically, we asked: (1) How do species richness (S) and phylogenetic structure (PS) of annuals and perennials in plant communities respond to aridity? (2) What is the contribution of other climatic and topo-edaphic variables in predicting S and PS for both components? (3) How does the taxonomic and phylogenetic turnover of annuals and perennials vary with spatial and environmental distances? We assessed annuals' and perennials' species richness, the phylogenetic structure at deep and shallow phylogenetic levels, and taxonomic and phylogenetic turnover along spatial and environmental distances. We found no relationship between annuals' richness and aridity, whereas perennials' richness showed a unimodal pattern. The phylogenetic structure of annuals and perennials showed contrasting responses to aridity and negatively correlated with topo-edaphic variables. We found phylogenetic clustering at intermediate-to-higher aridity levels for annuals, and at lower aridity levels for perennials. Both taxonomic and phylogenetic turnover in annuals and perennials correlated with the environmental distance rather than with spatial distance between communities, suggesting adaptation to local factors. Overall, our results show a decoupling in the response of the phylogenetic structure of annual and perennial components of plant communities to aridity in Mediterranean drylands. Our findings have significant implications for land management strategies under climate change.